List of Archived Posts

2012 Newsgroup Postings (09/29 - 10/24)

General Mills computer
STOP PRESS! An Auditor has been brought to task for a failed bank!
General Mills computer
General Mills computer
Another Light goes out
General Mills computer
General Mills computer
General Mills computer
General Mills computer
How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
Jedi Knights
General Mills computer
Why Auditors Fail To Detect Frauds?
System/360--50 years--the future?
System/360--50 years--the future?
System/360--50 years--the future?
System/360--50 years--the future?
System/360--50 years--the future?
System/360--50 years--the future?
How to get a tape's DSCB
General Mills computer
8-bit bytes and byte-addressed machines
Preparing for War with China
Jedi Knights
System/360--50 years--the future?
System/360--50 years--the future?
Is there a correspondence between 64-bit IBM mainframes and PoOps editions levels?
System/360--50 years--the future?
390 vector instruction set reuse, was 8-bit bytes
Jedi Knights
General Mills computer
390 vector instruction set reuse, was 8-bit bytes
390 vector instruction set reuse, was 8-bit bytes
390 vector instruction set reuse, was 8-bit bytes
System/360--50 years--the future?
390 vector instruction set reuse, was 8-bit bytes
390 vector instruction set reuse, was 8-bit bytes
PDP-10 and Vax, was System/360--50 years--the future?
Jedi Knights
PDP-10 and Vax, was System/360--50 years--the future?
history of Programming language and CPU in relation to each other
System/360--50 years--the future?
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each
Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each
Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each other
The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years
U.S. Sues Wells Fargo, Accusing It of Lying About Mortgages
Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Bull by the Horns
2012 History Conference
history of Programming language and CPU in relation to each
The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years
Should you support or abandon the 3270 as a User Interface?
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each other
Should you support or abandon the 3270 as a User Interface?
Fred Brooks on OS/360 "JCL is the worst language ever"
Fred Brooks on OS/360 "JCL is the worst language ever"
history of Programming language and CPU in relation to each other
Should you support or abandon the 3270 as a User Interface?
Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
history of Programming language and CPU in relation to each other
Mainframes are still the best platform for high volume transaction processing
Fred Brooks on OS/360 "JCL is the worst language ever"
history of Programming language and CPU in relation to each
history of Programming language and CPU in relation to each other
history of Programming language and CPU in relation to each other
U.S. banks on high alert against cyberattacks
Beyond the 10,000 Hour Rule
history of Programming language and CPU in relation to each other
history of Programming language and CPU in relation to each other
history of Programming language and CPU in relation to each other
Beyond the 10,000 Hour Rule
Protected: R.I.P. Containment
OT: Tax breaks to Oracle debated

General Mills computer

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Sat, 29 Sep 2012 16:18:49 -0400
Walter Bushell <proto@panix.com> writes:
Ah, so that's why Bush went in favor of that program despite it being against his stated principles and that of his party. You start by trying to destroy Social Security (Which was a holy Crusade for his branch of the GOP) and later add Medicare Part D?

re:
https://www.garlic.com/~lynn/2012m.html#33 General Mills computer
https://www.garlic.com/~lynn/2012m.html#71 General Mills computer

military-industrial-congressional complex ... MICC,
financial-regulatory-congressional complex ... FRCC,
pharmaceutical-regulatory-congressional complex ... PRCC

medicare part-d had 18 congressional members (& their senior staff) shepherding bill through and last minute add one sentence that prevents competitive bidding, they also prevent CBO from distributing analysis of the one sentence, until after vote passes. CBS 60mins shows identical drugs from VA (which allows competitive bidding) 1/3rd the price of medicare pard-d (w/o competitive bidding). after passes, all 18 have resigned and are on drug industry payroll. comptroller general characterises part-d a long-term $40T unfunded mandate that comes to swamp all other budget items. also seems to be part of the comptroller general's motivation for characterising nobody in congress capable of middle school arithmetic (first major legislation after congress allows fiscal responsibility act to expire in 2002 ... which required tax revenues match spending).

congress passing 401k puts large amounts of retirement funds into wallstreet hands ... the performance of 401k avg much worse than large retirement funds and the wallstreet fees/commissions for 401k is much larger than large retirement funds i.e. congressional 401k is enormous gift to wallstreet ... in much the same way that having S&L cut reserve requirements in half was enormous gift to wallstreet. The proposals from the last decade regarding privatising social security would have effectively put additional enormous additional funds into wallstreet for them to skim off (i.e. huge amounts of social security would have evaporated into various hands in much the same way that 401k funds evaporated last decade).

In much the same way the DEA followed the money trail involved in purchasing 737 jets used in drug smuggling back to too-big-to-fail doing the money laundering ... follow the money trail regarding efforts behind privatising social security (who benefits the most from the privatising of social security? and has enormous piles of money to heavily lobby washington).

past post mentioning privatising social security
https://www.garlic.com/~lynn/2008p.html#8 Global Melt Down

past posts mentioning too-big-to-fail caught money laundering (motivating reference too-big-to-jail)
https://www.garlic.com/~lynn/2010m.html#24 Little-Noted, Prepaid Rules Would Cover Non-Banks As Wells As Banks
https://www.garlic.com/~lynn/2010q.html#58 Programmer Charged with thieft (maybe off topic)
https://www.garlic.com/~lynn/2011.html#50 What do you think about fraud prevention in the governments?
https://www.garlic.com/~lynn/2011f.html#52 Are Americans serious about dealing with money laundering and the drug cartels?
https://www.garlic.com/~lynn/2011n.html#49 The men who crashed the world
https://www.garlic.com/~lynn/2012e.html#16 Wonder if they know how Boydian they are?
https://www.garlic.com/~lynn/2012e.html#35 The Dallas Fed Is Calling For The Immediate Breakup Of Large Banks
https://www.garlic.com/~lynn/2012e.html#37 The $30 billion Social Security hack
https://www.garlic.com/~lynn/2012f.html#88 Defense acquisitions are broken and no one cares
https://www.garlic.com/~lynn/2012g.html#9 JPM LOSES $2 BILLION USD!
https://www.garlic.com/~lynn/2012g.html#20 Psychology Of Fraud: Why Good People Do Bad Things
https://www.garlic.com/~lynn/2012i.html#14 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012j.html#25 This Is The Wall Street Scandal Of All Scandals
https://www.garlic.com/~lynn/2012k.html#37 If all of the American earned dollars hidden in off shore accounts were uncovered and taxed do you think we would be able to close the deficit gap?
https://www.garlic.com/~lynn/2012m.html#30 General Mills computer

--
virtualization experience starting Jan1968, online at home since Mar1970

STOP PRESS! An Auditor has been brought to task for a failed bank!

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 29 Sep 2012
Subject: STOP PRESS! An Auditor has been brought to task for a failed bank!
Blog: Financial Cryptography
STOP PRESS! An Auditor has been brought to task for a failed bank!
https://financialcryptography.com/mt/archives/001387.html

references:

The Big Four accounting firms; Shape shifters; With the audit market maturing, accounting firms become consultancies
http://www.economist.com/node/21563726

Griftopia
https://www.amazon.com/Griftopia-Machines-Vampire-Breaking-ebook/dp/B003F3FJS2

CFTC required significant positions to play because speculators result in wild irrational price swings. Then 19 secret letters were issued allowing specific speculators to play ... accounting for various things ... including the wild spike of oil way over $100 the summer of 2008 (and fall back below).

then Senator releasing data showing speculators causing the wild irrational price swings (aka they promote volatility making money on the swing up as well as the swing back down) and is heavily criticized in several corners

Sen. Bernie Sanders Leaks Oil Trading Data: Americans Have A Right To Know Who Drove Up Gas Prices
http://crooksandliars.com/susie-madrak/sen-bernie-sanders-leaks-oil-trading-
Bernie Sanders Demands Action From Obama On Wall Street Oil 'Gambling'
http://www.huffingtonpost.com/2011/04/29/bernie-sanders-demands-ac_n_855495.html
FIA "shocked and outraged after Senator leaks oil trading data"
http://www.finextra.com/news/Fullstory.aspx?newsitemid=22911

CFTC is now allowing speculators to drive extreme volatility and wild price swings in commodities ... ala Grfitopia book and other news items.

Drop back a little, Born, head of CFTC, suggests that it needs to regulate CDS (as insurance instruments, there may be a couple trillion at risk ... however as gambling bets, there is now over $700T in CDS ... hundreds times more than any at risk insurance).

So drop back a little more, person is competing with Gerstner to be next head of AMEX ... and looses out, then leaves with his protoge Jamie Dimon. They go to Baltimore and take over a loan business, then over a few years make several acquisitions eventually culminating in Citibank ... which is in violation of Glass-Steagall. Greenspan gives Citi an exemption while the new CEO lobbies washington for repeal of Glass-Steagall (Sec. of treasury who happens to be former chairman of Goldman-Sachs, head of senate banking committee and others). Head of senate banking committee gets repeal included in GLBA. After GLBA passes, sec. of treasury resigns and joins Citi becoming co-chariman.

Various people oppose regulating CDS including Enron. Born is replaced as head of CFTC by wife of the head of senate banking committee, while her husband gets legislation included that prohits CFTC from regulating CDS. When that passes, his wife (head of CFTC) resigns and joins Enron board and member of the audit committee.

Enron and Worldcom explode and congress passes Sarbanes-Oxley supposedly to prevent Enron/Worldcom from happening again; however it requires SEC to do something. Apparently GAO doesn't believe SEC is doing anything and does reports of public company fraudulent financial filings ... even showing uptic after passage of SOX (question, does SOX have 1) no effect on fraudulent filings, 2) encourages fraudulent filings, 3) if it wasn't for SOX there would have been even larger increase in fraudulent filings).
http://www.gao.gov/products/GAO-03-395R .
http://www.gao.gov/products/GAO-06-678 .
https://www.gao.gov/products/gao-06-1079sp .

supposedly under SOX, all the executives and auditors would be doing jail time.

Lack of CDS regulation by CFTC goes on to play major role in triple-A rated toxic CDOs, AIG, etc.

recent post with news items about husband&wife involvement in CFTC, Enron, CDS, Glass-Steagall, etc
https://www.garlic.com/~lynn/2012c.html#31

some other recent posts mentioning audit
https://www.garlic.com/~lynn/2012e.html#79 What's the takeaway on Audit?
https://www.garlic.com/~lynn/2012j.html#60 Auditors All Fall Down; PFGBest and MF Global Frauds Reveal Weak Watchdogs
https://www.garlic.com/~lynn/2012j.html#85 Study: One in Five Firms Misrepresent Earnings
https://www.garlic.com/~lynn/2012k.html#89 Auditors Don't Know Squat!
https://www.garlic.com/~lynn/2012k.html#90 Auditors Don't Know Squat!
https://www.garlic.com/~lynn/2012l.html#4 Auditors Don't Know Squat!
https://www.garlic.com/~lynn/2012l.html#6 Some fun with IBM acronyms and jargon (was Re: Auditors Don't Know Squat!)
https://www.garlic.com/~lynn/2012m.html#19 Why Auditors Fail To Detect Frauds?

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Sat, 29 Sep 2012 21:14:36 -0400
Anne & Lynn Wheeler <lynn@garlic.com> writes:
addtional references:
http://fabiusmaximus.com/2012/09/27/liscio-income-inequality-research-polls43607/
http://tlrii.typepad.com/theliscioreport/2012/03/recent-work-on-income-disparity.html


re:
https://www.garlic.com/~lynn/2012m.html#65 General Mills computer

somebody's update today:

A New American Record: The 1 Percent Have Officially Stolen More of the National Income
http://johnhively.wordpress.com/2012/09/29/a-new-american-record-the-1-percent-have-officially-stolen-more-of-the-national-income/
along with
http://johnhively.wordpress.com/2012/09/17/eight-key-charts-about-the-growing-income-inequality-in-the-united-states/

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

Refed: **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Sun, 30 Sep 2012 09:04:13 -0400
jmfbahciv <See.above@aol.com> writes:
It was going on during the S&L crisis, too. Just because the banks are the center of a similar problems implies the same kind of theft will occur. These types of opportunists are paid very well to figure out ways around each Congressional law they pass.

re:
https://www.garlic.com/~lynn/2012m.html#68 General Mills computer
https://www.garlic.com/~lynn/2012m.html#71 General Mills computer

or just the opposite ... constantly besieging congress with huge bags of money ... from today:

Scott Brown: ATM For The Big Banks
http://baselinescenario.com/2012/09/30/scott-brown-atm-for-the-big-banks/

from above:
It was not paper money he was handing out, of course, it was something much more valuable -- rule changes that conferred a greater ability to take on reckless risk, damage consumers, and impose higher future costs on the taxpayer.

... snip ...

lots of activity predates most recent congress. However, Glass-Steagall had been passed to separate the safety and soundness of regulated depository institutions from the reckless unregulated behavior of investment banks. Repeal of Glass-Steagall opened the way for too-big-to-fail and risky behavior putting depository institutions at risk ... and apparently never ending assistance to such ill-considered behavior ... leading to moral hazard ... the expectation that there is no real risk in outrageous gambling ... if they win, they keep the beneifits, if they loose, the gov. steps in to save them. The current round is to make sure that there is no real limitations put on the risks that too-big-to-fail are able to take.

most recent take on lots of deregulation and lack of enforcement resulted in the economic mess:
https://www.garlic.com/~lynn/2012n.html#1 STOP PRESS! An Auditor has been brought to task for a failed bank!

above from financial cryptography was somewhat facetious ... since the "pleading no contest", paying a fine, and promising to stop doing it ... has become something of a farce ... the fines are so small compared to the amounts involved (becoming small part of the cost of doing businesses) and several of the same institutions pleading no contest to the same violations year after year (promise to stop violations meaningless).

misc. past posts mentioning moral hazard
https://www.garlic.com/~lynn/2008g.html#64 independent appraisers
https://www.garlic.com/~lynn/2008j.html#71 lack of information accuracy
https://www.garlic.com/~lynn/2008j.html#76 lack of information accuracy
https://www.garlic.com/~lynn/2008k.html#16 dollar coins
https://www.garlic.com/~lynn/2008l.html#51 Monetary affairs on free reign, but the horse has Boulton'd
https://www.garlic.com/~lynn/2008l.html#67 dollar coins
https://www.garlic.com/~lynn/2008m.html#83 Fraud due to stupid failure to test for negative
https://www.garlic.com/~lynn/2008m.html#86 WSJ finds someone to blame.... be skeptical, and tell the WSJ to grow up
https://www.garlic.com/~lynn/2008n.html#0 Blinkylights
https://www.garlic.com/~lynn/2008n.html#3 Blinkylights
https://www.garlic.com/~lynn/2008n.html#65 Whether, in our financial crisis, the prize for being the biggest liar is
https://www.garlic.com/~lynn/2008n.html#69 Another quiet week in finance
https://www.garlic.com/~lynn/2008o.html#37 The human plague
https://www.garlic.com/~lynn/2010q.html#66 Ernst & Young sued for fraud over Lehman
https://www.garlic.com/~lynn/2011.html#84 The Imaginot Line
https://www.garlic.com/~lynn/2011f.html#52 Are Americans serious about dealing with money laundering and the drug cartels?
https://www.garlic.com/~lynn/2011h.html#22 Is BitCoin a triple entry system?
https://www.garlic.com/~lynn/2011j.html#41 Advice from Richard P. Feynman
https://www.garlic.com/~lynn/2011l.html#40 The Next Convergence: The Future of Economic Growth in a Multispeed World
https://www.garlic.com/~lynn/2011m.html#18 computer bootlaces
https://www.garlic.com/~lynn/2011n.html#49 The men who crashed the world
https://www.garlic.com/~lynn/2011p.html#96 Republicans Propose Bill to Treat Mexican Drug Cartels as 'Terrorist Insurgency'
https://www.garlic.com/~lynn/2012b.html#35 Entropy and #SocialMedia
https://www.garlic.com/~lynn/2012c.html#32 US real-estate has lost $7T in value
https://www.garlic.com/~lynn/2012g.html#28 REPEAL OF GLASS-STEAGALL DID NOT CAUSE THE FINANCIAL CRISIS - WHAT DO YOU THINK?
https://www.garlic.com/~lynn/2012g.html#61 Why Hasn't The Government Prosecuted Anyone For The 2008 Financial recession?

--
virtualization experience starting Jan1968, online at home since Mar1970

Another Light goes out

Refed: **, - **, - **
From: lynn@GARLIC.COM (Anne & Lynn Wheeler)
Subject: Re: Another Light goes out
Newsgroups: bit.listserv.ibm-main
Date: 30 Sep 2012 07:02:25 -0700
shmuel+gen@PATRIOT.NET (Shmuel Metz , Seymour J.) writes:
If your bonuses are based on 90 day cash flow, what are you going to optimise? As long as corporate decisions are driven by short term fluctuations in the stock prices, you are going to see decisions with adverse long-term consequences.

published earlier this month
https://www.amazon.com/Who-Stole-American-Dream-ebook/dp/B007MEWAX2

has a number of quotes from former IBM vp (and head of research) Ralph Gomory ... like loc.4213-16:
In this new era of globalization, the interests of companies and countries have diverged. In contrast with the past, what is good for America's global corporations is no longer necessarily good for the American people. -- RALPH GOMORY, former IBM vice president

... snip ...

has some amount on executive compensation and bonuses have become misaligned with long-term corporate and public interest. there are some number of stories about executives becoming myopicly focused on managing the quarterly reports to exclusion of running the company. there is folklore that after the company went into the red in the early 90s ... that the several hundred top executives on the executive bonus plan focused solely on moving expenses from the following year to the year in the red (w/o regard to any consideration of actually running the company). The issue is that put the company slightly out of the red the following year and the way the bonuses were calculated as change from previous year, that year's executive bonus was more than twice as large as the largest bonus ever paid (i.e. the executives actually made more money than if the company hadn't gone into the red).

In the late 70s & early 80s I was blamed for online computer conferencing on the internal network (larger than arpanet/internet from just about the beginning until late '85 or early '86). There was some amount of discussion about the rise of MBAs and the need for common, generic, simplified measurement criteria ... allowing MBAs to run companies w/o actually needing to know what the company does ... was leading to the downfall of US business.

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Sun, 30 Sep 2012 10:20:24 -0400
JimP. <pongbill127@cableone.net> writes:
If any of you watch 'how the states got their shapes' on History channel, you'll learn about all kinds of arguments and battles between the states over borders. Some still on-going.

recent URL item

America's State Borders Not Set in Stone
http://news.yahoo.com/americas-state-borders-not-set-stone-131555853.html

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Sun, 30 Sep 2012 15:43:54 -0400
re:
https://www.garlic.com/~lynn/2012m.html#65 General Mills computer
https://www.garlic.com/~lynn/2012n.html#2 General Mills computer

Is QE3 Yet Another Stealth Bank Bailout?
http://www.nakedcapitalism.com/2012/09/is-qe3-yet-another-stealth-bank-bailout.html

The triple-A rated toxic CDOs had gone for 22cents on the dollar ... but as part of bail-out, the Fed was buying at or near face-value (this time says 100cents on the dollar, there was some reference in the past that some were bought at only 98cents on the dollar).

The emerging headache of QE3: Andy Xie
http://www.marketwatch.com/story/the-emerging-headache-of-qe3-andy-xie-2012-09-26

from above:
That suggests that QE3 will last for over two years, and the total amount of MBS purchases will exceed $1 trillion, more than QE1's $1 trillion.

... snip ...

considering that estimated $27T were done during the bubble and end of 2008, just the four largest TBTF were carrying $5.2T "off-book" (if they had been forced to bring back on the balance sheet, the institutions would have been declared insolvent and forced to be liquidated).
Evil Wall Street Exports Boomed With 'Fools' Born to Buy Debt
https://www.bloomberg.com/news/articles/2008-10-27/evil-wall-street-exports-boomed-with-fools-born-to-buy-debt

Let's Shift Inflation to Medical, Rent, Education, Food, and Communications -- QE3 Of, By, and For the 1%
http://www.phibetaiota.net/2012/09/gold-transformer-screw-the-99-lets-shift-inflation-to-medical-rent-education-food-and-communications-qe3-of-by-and-for-the-1/

When it costs more to be poor -- Fed and government shifting inflation onto rent, medical care, and food. QE3 to widen the gap between the poor and the wealthy.
http://www.mybudget360.com/inflation-middle-class-poor-fed-qe3-impact-on-inflation-cpi-components/

The Data Don't Support That The U.S. Bank Bailouts Were Better Than the Swedish Model
http://www.nakedcapitalism.com/2012/09/the-data-dont-support-that-the-u-s-bank-bailouts-were-better-than-the-swedish-model.html

recent references to bailout chose the zombie japan bank model rather than swedish model
https://www.garlic.com/~lynn/2012.html#21 Zombie Banks
https://www.garlic.com/~lynn/2012.html#47 Avoiding a lost decade
https://www.garlic.com/~lynn/2012h.html#25 US economic update. Everything that follows is a result of what you see here
https://www.garlic.com/~lynn/2012h.html#36 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#37 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#53 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012j.html#86 Study: One in Five Firms Misrepresent Earnings
https://www.garlic.com/~lynn/2012k.html#79 Romney and Ryan's Phony Deficit-Reduction Plan
https://www.garlic.com/~lynn/2012l.html#48 The Payoff: Why Wall Street Always Wins
https://www.garlic.com/~lynn/2012l.html#53 CALCULATORS
https://www.garlic.com/~lynn/2012l.html#64 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012m.html#61 General Mills computer

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Sun, 30 Sep 2012 17:47:05 -0400
re:
https://www.garlic.com/~lynn/2012m.html#38 General Mills computer
https://www.garlic.com/~lynn/2012m.html#42 General Mills computer
https://www.garlic.com/~lynn/2012m.html#44 General Mills computer

Are Average Investors Getting Bilked by Wall Street Supercomputers? While high-frequency trading has propped up volume in the market, the method rewards shortsighted investors looking to make a quick profit
http://business.time.com/2012/09/24/are-average-investors-getting-bilked-by-wall-street-supercomputers/

recent past posts in other threads:
https://www.garlic.com/~lynn/2012c.html#13 Study links ultrafast machine trading with risk of crash
https://www.garlic.com/~lynn/2012f.html#86 The Dangers of High-Frequency Trading; Wall Street's Speed Freaks

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Mon, 01 Oct 2012 09:34:11 -0400
re:
https://www.garlic.com/~lynn/2012m.html#50 General Mills computer
https://www.garlic.com/~lynn/2012m.html#69 General Mills computer
https://www.garlic.com/~lynn/2012n.html#6 General Mills computer

other pressure on Federal Reserve for lots of nearly free money

Examining the Ubiquity of Private Equity
http://dealbook.nytimes.com/2012/09/17/examining-the-ubiquity-of-private-equity/
Kravis Shared DNA With Roberts Makes Buyout Men Unique
http://www.bloomberg.com/news/2012-09-27/kravis-shared-dna-with-roberts-makes-buyout-men-unique.html

above references recent book: The New Tycoons: Inside The Trillion Dollar Private Equity Industry That Owns Everything
https://www.amazon.com/The-New-Tycoons-Everything-ebook/dp/B008KPMBIG

earlier book describing private equity industry (from Nov2009):

The Buyout of America: How Private Equity Will Cause the Next Great Credit Crisis
https://www.amazon.com/Buyout-America-Private-Equity-Credit/dp/1591842859/

talks about FED keeping cost of money at nearly zero has been big driver for private equity industry. The firms borrow money to do reverse IPOs and buy up companies. They take enormous fees and comissions on the deal. Later they do IPO of the company ... difference from home speculators flipping houses, is rather than paying off the original loan (mortgage), the loan stays with the newly IPO'ed company (if rates rise later, the newly IPO'ed company may have severe problems servicing the debt). The rise of speculators and flipping houses the last decade was driven by the housing bubble (some parts of the country saw 20-30% inflation). Private equity firms can make enormous profits even if they flip a company at a loss (since the original loan doesn't have to be paid off and goes with the flipped company)

Gerstner is in competition to be next CEO at AMEX. The other person looses and leaves. AMEX private equity arm is in competition with KKR to buy RJR and looses. After take-over of RJR by KKR, KKR hires Gerstner away to turn-around RJR. Gerstner then is hired as CEO to resurrect IBM. After leaving IBM, Gerstner serves as CEO of another large private equity company (Carlyle) from 2003-2008.

gerstner posts
https://www.garlic.com/~lynn/submisc.html#gerstner
private equity posts
https://www.garlic.com/~lynn/submisc.html#private.equity

recent posts mentioning private equity industry
https://www.garlic.com/~lynn/2012.html#37 Romney's Opponents Intensify Attacks as Voting Nears
https://www.garlic.com/~lynn/2012.html#40 Age of Greed: The Triumph of Finance and the Decline of America, 1970 to the Present
https://www.garlic.com/~lynn/2012b.html#43 Where are all the old tech workers?
https://www.garlic.com/~lynn/2012g.html#82 How do you feel about the fact that today India has more IBM employees than US?
https://www.garlic.com/~lynn/2012g.html#87 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#27 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#29 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#45 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#55 The Invention of Email
https://www.garlic.com/~lynn/2012i.html#1 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012k.html#31 History--punched card transmission over telegraph lines
https://www.garlic.com/~lynn/2012l.html#63 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012m.html#24 Does the IBM System z Mainframe rely on Security by Obscurity or is it Secure by Design

--
virtualization experience starting Jan1968, online at home since Mar1970

How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 01 Oct 2012
Subject: How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
Blog: IBMers
re:
https://www.garlic.com/~lynn/2012m.html#23 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012m.html#67 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

The funny thing in the Enterprise discussion, is that the discussion started out that mainframe software, services, & storage is generating more than six times the revenue of the mainframe itself ... i.e. max. Z196 goes for $28M and 25/4 multiplier makes total revenue $175M ... and then the discussion wanders off into how little the total cost of mainframe is (it is inconsistent to argue that total revenue from mainframes is enormous ... but apparently somebody besides customers are paying the bill if mainframe is so economical for customers).

The "Who Stole American Dream" goes into some detail of IBM effectively replacing large number of Americans with Indians ... going from 6k in 2003 to possibly 130k by 2010 ... and got so good at replacing American's with Indians that it set up a business unit to advise other companies on how to do it. As referenced upthread some number of comments in the book also by former IBM VP (and head of research).

also regarding ficon, zhpf, z10, z196, ec12, e5-2600, etc

post with IBM URLs for PDF documents about FICON, zHPF, SSCH,
https://www.garlic.com/~lynn/2012m.html#4

Lawrence Livermore National lab started work on fibre channel standard in 1988. I've been involved with national lab on&off since 1979 when trying to sell them large clusters of 4341s. I was active in fibre channel included when POK channel engineers become involved in the 90s ... working on layering FICON on top that significantly cut throughput (for the FICON layer compared to base fibre channel). old email referencing 4300s
https://www.garlic.com/~lynn/lhwemail.html#4341

fibre channel has always been able to do concurrent in both directions so latest 16gbit fibre channel theoretically does 32gbit aggregate or equivalent of 4M 4kbytes/sec, 2m concurrent in both direction. IBM documents has z196 doing 2m 4kbytes peak using 104 FICON channels to 14 storage subsystems. emulex reference is their latest fibre channel for e5-2600 is capable of over 1m IOPS (aka single channel). Also says that z196 peaks at 2.2m SSCH/sec with all 14 SAPs running at 100% busy ... but recommends keeping SAPs at no more than 70% busy ... or 1.5m SSCH/sec. IBM zHPF document discussion is that it operates the same for z196 and ec12.

I did HA/CMP and was doing cluster scale-up in conjunction with fibre channel standard (for both numerical intensive/scientific including with numerous national labs as well as commercial) ... old post referencing to early jan92 meeting in ellison's conference room
https://www.garlic.com/~lynn/95.html#13
and old email regarding cluster scale-up
https://www.garlic.com/~lynn/lhwemail.html#medusa

within hrs of last email mentioned above (end of Jan92), cluster scale-up was transferred and we were told we couldn't work on anything with more than four processors. within a couple weeks it was announced as supercomputer for numeric intensive/scientific *ONLY*. we then decide to leave. other past posts mentioning ha/cmp
https://www.garlic.com/~lynn/subtopic.html#hacmp

as an aside, in the 80s was working with NSF on what would become the NSFNET backbone (tcp/ip is the technology basis for modern internet, NSFNET backbone was the operational basis for the modern internet and CIX was the business basis for the modern internet). some old email
https://www.garlic.com/~lynn/lhwemail.html#nsfnet

when the NSFNET backbone rfp was released, internal politics prevented us from bidding (some of the same executives later involved in transferring cluster scale-up were also involved in preventing us from bidding on NSFNET backbone). The director of NSF tried to help by writing the company a letter 3Apr1986, NSF Director to IBM Chief Scientist and IBM Senior VP and director of Research, copying IBM CEO), but that just made the internal politics worse

another old reference "From Annals of Release No Software Before Its Time" ... finally seeing some stuff released ... doing commercial dbms scale-up circa 1990/1991 (as well as some internal mainframe cluster stuff from late 70s/early 80s):
https://www.garlic.com/~lynn/2009p.html#43

other random tidbits ... past posts about being involved in the original relational DBMS implementation ... System/R
https://www.garlic.com/~lynn/submain.html#systemr

and past posts getting dragged into also playing disk engineer in bldg.14&15. part of it was that they were doing stand-alone, pre-scheduled 7x24 testing of disk/controller under development. At one time they tried MVS for supporting, multiple concurrent testing ... but found MVS had 15min MTBF with single testcell. I offered to rewrite I/O supervisor to make it absolute bullet-proof and never fail ... so that they could do on-demand, anytime, concurrent testing ... significantly improving engineering development productivity. As a result, they were constantly asking me to help with various design and engineering and also act as interface with POK channel engineers
https://www.garlic.com/~lynn/subtopic.html#disk

recent posts & references on EC12 announcement (including compared to z196)
https://www.garlic.com/~lynn/2012l.html#41
https://www.garlic.com/~lynn/2012l.html#56
https://www.garlic.com/~lynn/2012l.html#67
https://www.garlic.com/~lynn/2012l.html#71
https://www.garlic.com/~lynn/2012l.html#90

posts (many with URL sources/references) regarding e5-2600 blade @527BIPS compared to z196 @50BIPS
https://www.garlic.com/~lynn/2012e.html#94
https://www.garlic.com/~lynn/2012g.html#36
https://www.garlic.com/~lynn/2012g.html#38
https://www.garlic.com/~lynn/2012h.html#4
https://www.garlic.com/~lynn/2012h.html#20
https://www.garlic.com/~lynn/2012h.html#35
https://www.garlic.com/~lynn/2012h.html#52
https://www.garlic.com/~lynn/2012h.html#62
https://www.garlic.com/~lynn/2012h.html#70
https://www.garlic.com/~lynn/2012i.html#11
https://www.garlic.com/~lynn/2012i.html#16
https://www.garlic.com/~lynn/2012i.html#84
https://www.garlic.com/~lynn/2012i.html#88
https://www.garlic.com/~lynn/2012j.html#66
https://www.garlic.com/~lynn/2012j.html#95
https://www.garlic.com/~lynn/2012j.html#96
https://www.garlic.com/~lynn/2012k.html#41
https://www.garlic.com/~lynn/2012l.html#20
https://www.garlic.com/~lynn/2012l.html#27
https://www.garlic.com/~lynn/2012l.html#30
https://www.garlic.com/~lynn/2012l.html#42
https://www.garlic.com/~lynn/2012l.html#51
https://www.garlic.com/~lynn/2012l.html#56
https://www.garlic.com/~lynn/2012l.html#81
https://www.garlic.com/~lynn/2012l.html#87
https://www.garlic.com/~lynn/2012l.html#88
https://www.garlic.com/~lynn/2012l.html#90
https://www.garlic.com/~lynn/2012m.html#13
https://www.garlic.com/~lynn/2012m.html#28

--
virtualization experience starting Jan1968, online at home since Mar1970

Jedi Knights

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 01 Oct 2012
Subject: Jedi Knights
Blog: Boyd Strategy
Certain to Win (Chet Richards) loc1336-40:
Before the First Gulf War, the Army established a year-long course, the School of Advanced Military Studies (SAMS) at Ft. Leavenworth, Kansas, where a select group of Command and Staff College graduates study ways to apply warfare based on rapid decision cycles -- "maneuver warfare." Boyd worked with SAMS's founder, then-Colonel Huba Wass de Czege, and lectured there on several occasions. SAMS alumni call themselves "Jedi Knights," and Gen Schwarzkopf's staff in Riyadh contained upwards of 60 of them.

... snip ...

Around Desert Storm, US News & World Report had article on Boyd titled "The Fight To Change How America Fights" ... also mentioning his Jedi Knights (6May1991)

other Desert Strom references

Long Strange Journey
https://www.amazon.com/Long-Strange-Journey-Intelligence-ebook/dp/B004NNV5H2

Has DOD testifying before congress that there wasn't any chemical/biological weapons during desert storm ... when they knew there were. The apparent explanation was that they didn't want it known that they knowingly sent in US military without adequate protection (and tried to have VA stonewall admitting the soldiers were suffering effects from chemical/biological weapons). Then a decade later, when Iraq no longer had WMDs, they use the excuse that they had WMDs, to go in again.

loc7297-7301:
Retirement had been kind to Norman Schwarzkopf. Scattered post-war revelations--Patriot missiles missing every SCUD fired by Iraq, Saddam still entrenched in Baghdad--had failed to shake the public's favorable perception of "Stormin' Norman" as triumphant war hero. The former general enjoyed lucrative speaking engagements around the country and he served as a paid consultant to NBC News. Six years after America's "victory" in Desert Storm, the only nettlesome issue confronting Schwarzkopf was "a few" sick soldiers with unexplained illnesses.

... snip ...

Boyd: The Fighter Pilot Who Changed the Art of War
https://www.amazon.com/Boyd-Fighter-Pilot-Changed-ebook/dp/B000FA5UEG

loc7362-64
But in Dick Cheney the Pentagon had a rare SecDef. Cheney had enough one-on-one sessions with Boyd to give him the knowledge and self-confidence to second-guess even a headstrong four-star general such as Norman Schwarzkopf. Simply put, Cheney knew more about strategy than did his generals.

... snip ... and loc7390-92:
Schwarzkopf and several generals have since spent much of their time blaming each other, but it was the slavish adherence to an outmoded attrition-warfare doctrine that allowed the Iraqis to escape. Boyd's earlier predictions about synchronization in the Army were proven true.

... snip ...

URLs and/or posts mentioning Boyd
https://www.garlic.com/~lynn/subboyd.html

past Jedi Knights &/or US News & World Reports article references:
https://www.garlic.com/~lynn/94.html#8 scheduling & dynamic adaptive ... long posting warning
https://www.garlic.com/~lynn/99.html#120 atomic History
https://www.garlic.com/~lynn/2000e.html#34 War, Chaos, & Business (web site), or Col John Boyd
https://www.garlic.com/~lynn/2001b.html#65 Java as a first programming language for cs students
https://www.garlic.com/~lynn/2002c.html#14 OS Workloads : Interactive etc
https://www.garlic.com/~lynn/2003h.html#58 employee motivation & executive compensation
https://www.garlic.com/~lynn/2004l.html#34 I am an ageing techy, expert on everything. Let me explain the
https://www.garlic.com/~lynn/2005j.html#15 The 8008
https://www.garlic.com/~lynn/2006q.html#41 was change headers: The Fate of VM - was: Re: Baby MVS???
https://www.garlic.com/~lynn/2008c.html#9 was: 1975 movie "Three Days of the Condor" tech stuff
https://www.garlic.com/~lynn/2008c.html#39 Current Officers
https://www.garlic.com/~lynn/2008c.html#40 Current Officers
https://www.garlic.com/~lynn/2008c.html#43 Current Officers
https://www.garlic.com/~lynn/2009o.html#37 Young Developers Get Old Mainframers' Jobs
https://www.garlic.com/~lynn/2009p.html#60 MasPar compiler and simulator
https://www.garlic.com/~lynn/2009q.html#38 The 50th Anniversary of the Legendary IBM 1401
https://www.garlic.com/~lynn/2010i.html#68 Favourite computer history books?
https://www.garlic.com/~lynn/2011j.html#14 Innovation and iconoclasm
https://www.garlic.com/~lynn/2011n.html#90 There is much we can learn from TE Lawrence
https://www.garlic.com/~lynn/2012h.html#17 Hierarchy

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Tue, 02 Oct 2012 13:29:10 -0400
Anne & Lynn Wheeler <lynn@garlic.com> writes:
Abigail Field: Mortgage Settlement Institutionalizes Foreclosure Fraud
http://www.nakedcapitalism.com/2012/03/abigail-field-mortgage-settlement-institutionalizes-foreclosure-fraud.html


re:
https://www.garlic.com/~lynn/2012m.html#68 General Mills computer

Blatant Violation of Mortgage Settlement Elicits Predictable Limp-Wristed Response From Mortgage Monitor
http://www.nakedcapitalism.com/2012/09/blatant-violation-of-mortgage-settlement-elicits-predictable-limp-wristed-response-from-mortgage-monitor.html

from above:
Gretchen Morgenson has a good piece up today which again proves that no matter how bad you think the mortgage settlement is, it's worse.

... snip ...

How to Erase a Debt That Isn't There
http://www.nytimes.com/2012/09/30/business/when-banks-erase-a-debt-that-isnt-there.html

other past posts mentioning mortgage fraud
https://www.garlic.com/~lynn/2012b.html#95 Bank of America Fined $1 Billion for Mortgage Fraud
https://www.garlic.com/~lynn/2012c.html#4 Bank of America Fined $1 Billion for Mortgage Fraud
https://www.garlic.com/~lynn/2012f.html#13 The White House and Mortgage Fraud: So Far It's All Talk, No Action
https://www.garlic.com/~lynn/2012f.html#80 The Failure of Central Planning
https://www.garlic.com/~lynn/2012g.html#6 Adult Supervision
https://www.garlic.com/~lynn/2012g.html#63 SEC: Taking on Big Firms is 'Tempting,' But We Prefer Picking on Little Guys
https://www.garlic.com/~lynn/2012g.html#64 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012g.html#71 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012g.html#76 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#77 Interesting News Article
https://www.garlic.com/~lynn/2012i.html#13 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012i.html#14 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012k.html#75 What's the bigger risk, retiring too soon, or too late?
https://www.garlic.com/~lynn/2012k.html#89 Auditors Don't Know Squat!
https://www.garlic.com/~lynn/2012m.html#51 General Mills computer
https://www.garlic.com/~lynn/2012m.html#56 General Mills computer
https://www.garlic.com/~lynn/2012m.html#58 General Mills computer

--
virtualization experience starting Jan1968, online at home since Mar1970

Why Auditors Fail To Detect Frauds?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 03 Oct, 2012
Subject: Why Auditors Fail To Detect Frauds?
Blog: Financial Crime Risk, Fraud and Security
re:
https://www.garlic.com/~lynn/2012m.html#19 Why Auditors Fail To Detect Frauds?

cross-over from financial cryptography

STOP PRESS! An Auditor has been brought to task for a failed bank!
http://financialcryptography.com/mt/archives/001387.html
and
https://www.garlic.com/~lynn/2012n.html#1 STOP PRESS! An Auditor has been brought to task for a failed bank!

from above:
In what is stunning news - because we never believed it would happen - it transpires that, as reported by The Economist: In 2002 the Sarbanes-Oxley act limited what kind of non-audit services an American accounting firm can offer to client.

... snip ...

In the global financial crisis, Mark 1, which cost the world a trillion or two, Deloitte paid out $20m and denied any wrongdoing ... references:
http://www.economist.com/node/21563726

There is estimate of $27T in securitized loans done during the bubble ... tv business news programs yesterday and today are that the amounts so far discussed can be treated as (very small relative) cost of the way they do business (even some references about settling to try and wall off real substantial penalties and fines).

then there is this ...

PricewaterhouseCoopers Paid $1 Billion as Consultant on Mortgage Settlement
http://www.nakedcapitalism.com/2012/10/pricewaterhousecoopers-paid-1-billion-as-consultant-on-mortgage-settlement.html

Evil Wall Street Exports Boomed With 'Fools' Born to Buy Debt
https://www.bloomberg.com/news/articles/2008-10-27/evil-wall-street-exports-boomed-with-fools-born-to-buy-debt

from above:
More than $27 trillion of these securities have been sold since 2001, according to the Securities Industry Financial Markets Association, an industry trade group. That's almost twice last year's U.S. gross domestic product of $13.8 trillion.

... snip ...

estimate that just the four largest too-big-to-fail were still carrying $5.2T "off-balance" the end of 2008 (if they had been forced to bring back on balance, they would have been declared insolvent and forced to been liquidated)

Banks' Hidden Junk Menaces $1 Trillion Purge
Bank's Hidden Junk Menaces $1 Trillion Purge
http://www.bloomberg.com/apps/news?pid=newsarchive&sid=akv_p6LBNIdw&refer=home

from above:
At the end of 2008, for example, off-balance-sheet assets at just the four biggest U.S. banks -- Bank of America Corp., Citigroup Inc., JPMorgan Chase & Co. and Wells Fargo & Co. -- were about $5.2 trillion, according to their 2008 annual filings.

... snip ...

TARP funds were originally appropriated for the purchase of these toxic assets ... but with only $700B appropriated, it wouldn't have even come close to making a dent in the problem (at least not if purchased at face value). Note the fall of 2008, toxic assets were going for 22cents on the dollar ... the $700B would have covered a little over half of the $5.2T for just the four largest too-big-to-fail at 22cents on the dollar ... but that would require facing that they were failed institutions ... and still wouldn't have been able to clear the problem. Note that a lot of the fed money (Bloomberg spent two years in court with federal reserve getting them to release what they were doing) was purchasing toxic assets at between 98cents and 100cents on the dollar.

Note a big motivation for wallstreet pushing these instruments (and looking the other way as corners were cut) was the commissions&fees from the sales of the instruments ... i.e. they didn't actually care what was being sold ... just so long as they could churn out the transaction sales. Towards the end, there were even stories of custodians of the off-balance toxic asset warehouses effectively exchanging equal amounts of assets in sales to each other (as way of keeping the commissions&fees coming in).

note that securitized mortgages had been used during S&L crisis to obfuscate fraudulent mortgages ... however, w/o the triple-A ratings there was not much of market. Being able to pay the rating agencies for triple-A ratings (when both the rating agencies and sellers knew they weren't worth triple-A ... from congressional hearing testimony into the role that the rating agencies playing in the financial mess). The triple-A ratings enormously increase the market for the toxic assets ... especially large institutional funds restricted to only dealing in "safe" investments ... like the large retirement funds.

Note triple-A rating trumps documentation. Late last century we were asked to look at improving the integrity of supporting documentation in securitized mortgages as countermeasure to the use of securitized mortgages for fraudulent mortgages (done during the S&L crisis). However, with the (paid-for) triple-A ratings trumping documentation, loan originators could do no-documentation mortgages ... and with no-documentation there is no longer an issue regarding documentation integrity.

This also comes up with MERS and subsequent robo-signing to generate fraudulent documents after the fact

Note that 10% skim on the $27T transactions would be $2.7T ... the increase in wallstreet wealth during the bubble would indicate a take more like between 15% and 20% on the $27T in transactions ... of course some of that may have come from other manipulations ... including creating toxic CDOs that they knew would fail, paying for the triple-A ratings ... and then making CDS bets that they would fail. There have been claims that wallstreet as percent of GDP, tripled during the bubble ... indicating that their increased take during the bubble was something over $5T.
Evil Wall Street Exports Boomed With 'Fools' Born to Buy Debt
https://www.bloomberg.com/news/articles/2008-10-27/evil-wall-street-exports-boomed-with-fools-born-to-buy-debt

recent posts mentioning the bloomberg articles:
https://www.garlic.com/~lynn/2012b.html#19 "Buffett Tax" and truth in numbers
https://www.garlic.com/~lynn/2012c.html#31 US real-estate has lost $7T in value
https://www.garlic.com/~lynn/2012c.html#45 Fannie, Freddie Charge Taxpayers For Legal Bills
https://www.garlic.com/~lynn/2012c.html#46 PC industry is heading for more change
https://www.garlic.com/~lynn/2012d.html#5 PC industry is heading for more change
https://www.garlic.com/~lynn/2012d.html#32 PC industry is heading for more change
https://www.garlic.com/~lynn/2012d.html#42 China's J-20 Stealth Fighter Is Already Doing A Whole Lot More Than Anyone Expected
https://www.garlic.com/~lynn/2012e.html#40 Who Increased the Debt?
https://www.garlic.com/~lynn/2012e.html#42 Who Increased the Debt?
https://www.garlic.com/~lynn/2012e.html#57 speculation
https://www.garlic.com/~lynn/2012e.html#58 Word Length
https://www.garlic.com/~lynn/2012f.html#31 Rome speaks to us. Their example can inspire us to avoid their fate
https://www.garlic.com/~lynn/2012f.html#69 Freefall: America, Free Markets, and the Sinking of the World Economy
https://www.garlic.com/~lynn/2012f.html#75 Fed Report: Mortgage Mess NOT an Inside Job
https://www.garlic.com/~lynn/2012g.html#6 Adult Supervision
https://www.garlic.com/~lynn/2012g.html#20 Psychology Of Fraud: Why Good People Do Bad Things
https://www.garlic.com/~lynn/2012g.html#28 REPEAL OF GLASS-STEAGALL DID NOT CAUSE THE FINANCIAL CRISIS - WHAT DO YOU THINK?
https://www.garlic.com/~lynn/2012g.html#59 Why Hasn't The Government Prosecuted Anyone For The 2008 Financial recession?
https://www.garlic.com/~lynn/2012g.html#70 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012g.html#77 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#25 US economic update. Everything that follows is a result of what you see here
https://www.garlic.com/~lynn/2012h.html#26 US economic update. Everything that follows is a result of what you see here
https://www.garlic.com/~lynn/2012i.html#14 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012i.html#94 Naked emperors, holy cows and Libor
https://www.garlic.com/~lynn/2012j.html#28 Why Asian companies struggle to manage global workers
https://www.garlic.com/~lynn/2012j.html#65 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012k.html#38 Four Signs Your Awesome Investment May Actually Be A Ponzi Scheme
https://www.garlic.com/~lynn/2012k.html#75 What's the bigger risk, retiring too soon, or too late?
https://www.garlic.com/~lynn/2012l.html#64 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012m.html#56 General Mills computer

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Wed, 03 Oct 2012 10:12:48 -0400
hancock4 writes:
Nearly fifty years ago IBM announced a new mainframe architecture that remains in widespread service to this day (as the Z series). Despite predictions it was dead, it will still be in service on its 50th anniversary. Pretty amazing.

But how long after that?

Will the IBM mainframe continue, or will customers finally switch everything over to 'client server'? IBM has made many enhancements over the last 50 years, and today's system includes support for Linux, encryption, and other features not even dreamed of 50 years ago.

Also, how much of IBM, if any, will continue in manufacturing? Does IBM still make its own circuit cards and chips? Does it make any peripherals? (Weren't disk drives, which IBM invented, sold off? Also, weren't printers sold off?)

(The IBM website is somewhat cryptic on current technologies, using rather esoteric descriptions for its "solutions". It's hard to find a "system summary" publication like they used to publish for S/360 and S/ 370 listing the boxes, their speeds, and peripherals.)


recent thread that IBM earns around $5B on mainframe sales ... a maxed out z196 goes for $28M ... the $5B sales would be equivalent of approx. 180 maxed. out z196 machines. It notes that is approx. 4% of total sales ... but total mainframe revenue (including software, services and storage) accounts for 25% of revenue (and 40% of profit) ... that then is ratio of 25/4 for total mainframe revenue compared to mainframe computer revenue ... or customer with maxed out z196 (at $28M) would be paying more like $175M. maxed out z196 is rated at aggregate of 50BIPs ... or $175M/50 = $3.5M/BIPS
https://www.garlic.com/~lynn/2012m.html#31 Still think the mainframe is going away soon: Think again. IBM mainframe computer sales are 4% of IBM's revenue; with software, services, and storage it's 25%
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#67 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

for comparison, IBM has base price of $1815 for e5-2600 blade which have ratings of 527BIPS or $3.44/BIPS (IBM is taken in a million times per processor cycle for mainframe than e5-2600)

max. out z196 mainframe peaks out at 2M IOPS with 104 (FICON) channels and 14 storage subsystems. The 14 system assist processors are capable of 2.2M SSCH/sec at 100% busy ... but recommendation is to operate the system assist processors at no more than 70% cpu busy or 1.5M SSCH/sec

this is with ibm FICON channels ... which is layer built on top of fibre channel standard ... which drastically reduces throughput and efficiency of the underlying fibre channel standard. By comparison emulex says it has a fibre channel for e5-2600 capable of over million IOPS on a single fibre channel. recent references:
https://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#5 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#13 Intel Confirms Decline of Server Giants HP, Dell, and IBM
https://www.garlic.com/~lynn/2012m.html#28 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

LLNL had some serial technology and around 1988 they started effort to standardize as fibre channel standard. I had been involved with national lab off&on since late 70s ... including doing some benchmarks for national lab looking at possibly getting large number of 4341s for computational work ... some old email
https://www.garlic.com/~lynn/lhwemail.html#4341

and got dragged into various parts of the fibre channel standard ... have a bunch of archive from fibre channel standard mailing list ... including some amount of discored when POK channel engineers got involved doing the work to layer FICON on top of underlying fibre channel standard (significantly cutting throughput and efficiency).

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Wed, 03 Oct 2012 10:29:08 -0400
scott@slp53.sl.home (Scott Lurndal) writes:
Look around for the hot-chips 2012 presentations (the 2012 proceedings are currently embargoed (for participants only), but will be released to the public in december).

The Z12 presentation has good detail on the processor. The range of peripherals compared to 30 years ago has been considerably reduced (tiered storage and networking interfaces for the most part).


re:
https://www.garlic.com/~lynn/2012n.html#13 Systeme/360--50 years--the future?

max z10 has 64 processors and rated at 30BIPS or 469MIPS/processor

max z196 has 80 processors and rated at 50BIPS or 625MIPS/processor

claim is possibly half of per processor increase between z10 & z196 is introduction of out-of-order instruction execution

max ec12 claims 101 processor and rating of 75BIPS or 743MIPS/processor

part of the improvement from z196 to latest ec12 is further enhancements in out-of-order instruction execution.

recent posts mentioning ec12:
https://www.garlic.com/~lynn/2012l.html#39 The IBM zEnterprise EC12 announcment
https://www.garlic.com/~lynn/2012l.html#41 The IBM zEnterprise EC12 announcment
https://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#67 Strings (hijacked from: The IBM zEnterprise EC12 announcment)
https://www.garlic.com/~lynn/2012l.html#71 Strings (hijacked from: The IBM zEnterprise EC12 announcment)
https://www.garlic.com/~lynn/2012l.html#72 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#74 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#77 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#79 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#81 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#82 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#90 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#99 PDP-10 system calls, was 1132 printer history
https://www.garlic.com/~lynn/2012l.html#100 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#0 PDP-10 system calls, was 1132 printer history
https://www.garlic.com/~lynn/2012m.html#13 Intel Confirms Decline of Server Giants HP, Dell, and IBM
https://www.garlic.com/~lynn/2012m.html#28 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

by comparison e5-2600 blade has rating of 527BIPS with two 8-core chips, or 33BIPS/processor. there is some guestimate that e5-4600 blade will have rating of over 1000BIPS with four 8-core chips (but I haven't seen any published numbers yet). recent posts mentioning e5-2600:
https://www.garlic.com/~lynn/2012d.html#50 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012d.html#64 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012e.html#3 NASA unplugs their last mainframe
https://www.garlic.com/~lynn/2012e.html#4 Memory versus processor speed
https://www.garlic.com/~lynn/2012e.html#94 Can Mainframes Be Part Of Cloud Computing?
https://www.garlic.com/~lynn/2012e.html#99 Can Mainframes Be Part Of Cloud Computing?
https://www.garlic.com/~lynn/2012e.html#105 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012f.html#0 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012f.html#4 Can Mainframes Be Part Of Cloud Computing?
https://www.garlic.com/~lynn/2012f.html#7 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012g.html#36 Should IBM allow the use of Hercules as z system emulator?
https://www.garlic.com/~lynn/2012g.html#38 Should IBM allow the use of Hercules as z system emulator?
https://www.garlic.com/~lynn/2012h.html#4 Think You Know The Mainframe?
https://www.garlic.com/~lynn/2012h.html#20 Mainframes Warming Up to the Cloud
https://www.garlic.com/~lynn/2012h.html#35 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#52 How will mainframers retiring be different from Y2K?
https://www.garlic.com/~lynn/2012h.html#62 What are your experiences with Amdahl Computers and Plug-Compatibles?
https://www.garlic.com/~lynn/2012i.html#11 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#16 Think You Know The Mainframe?
https://www.garlic.com/~lynn/2012i.html#84 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#88 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#1 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#34 Can anybody give me a clea idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#46 Word Length
https://www.garlic.com/~lynn/2012j.html#66 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012j.html#95 printer history Languages influenced by PL/1
https://www.garlic.com/~lynn/2012j.html#96 The older Hardware school
https://www.garlic.com/~lynn/2012k.html#41 Cloud Computing
https://www.garlic.com/~lynn/2012l.html#20 X86 server
https://www.garlic.com/~lynn/2012l.html#27 PDP-10 system calls, was 1132 printer history
https://www.garlic.com/~lynn/2012l.html#28 X86 server
https://www.garlic.com/~lynn/2012l.html#30 X86 server
https://www.garlic.com/~lynn/2012l.html#34 X86 server
https://www.garlic.com/~lynn/2012l.html#42 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012l.html#51 Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#81 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#87 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#88 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#90 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#100 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#5 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#13 Intel Confirms Decline of Server Giants HP, Dell, and IBM
https://www.garlic.com/~lynn/2012m.html#28 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012m.html#31 Still think the mainframe is going away soon: Think again. IBM mainframe computer sales are 4% of IBM's revenue; with software, services, and storage it's 25%
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#67 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Wed, 03 Oct 2012 14:25:33 -0400
"Charles Richmond" <numerist@aquaporin4.com> writes:
In 1995, I bought an Apple Macintosh 635 computer. It came with a 350 *mega*byte Winchester drive. The drive was manufactured by IBM in Thailand and was encased in a hard black nylon case. The drive is now used in my Linux box (with an AMD chip) and is used exclusively as swap space.

One point is... what is an IBM drive doing in a Mac??? Another point... manufactured by IBM can mean many things. In this case, it meant made in Thailand.


re:
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#14 System/360--50 years--the future?

also remember that apple became part of AIM (apple, ibm, motorola) ... aka somerset ... moving to power/pc. the folklore is that apple then moved off power/pc to intel ... when it looked like power/pc would just be concentrating on producing server chips and not targeting the laptop (later tablet) market.

misc. past posts mentioning 801, romp, rios, iliad, power, power/pc, somerset, etc
https://www.garlic.com/~lynn/subtopic.html#801

a few years later the disk division gets sold off to hitachi which later it is sold off by hitachi

late 80s, a senior disk engineer gets a talk scheduled at the annual, internal, world-wide communication group conference and opens the talk with statement that the communication group was going to be responsible for the demise of the disk division. recent posts repeated the story:
https://www.garlic.com/~lynn/2012.html#73 Reject gmail
https://www.garlic.com/~lynn/2012.html#92 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012b.html#8 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012b.html#32 New IBM mainframe instructions
https://www.garlic.com/~lynn/2012c.html#40 Where are all the old tech workers?
https://www.garlic.com/~lynn/2012e.html#39 A bit of IBM System 360 nostalgia
https://www.garlic.com/~lynn/2012e.html#61 "25 Years of IBM's OS/2"
https://www.garlic.com/~lynn/2012g.html#32 IBM bans Siri: Privacy risk, or corporate paranoia at its best?
https://www.garlic.com/~lynn/2012g.html#34 Co-existance of z/OS and z/VM on same DASD farm
https://www.garlic.com/~lynn/2012g.html#72 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#12 How do you feel about the fact that today India has more IBM employees than US?
https://www.garlic.com/~lynn/2012h.html#57 How will mainframers retiring be different from Y2K?
https://www.garlic.com/~lynn/2012i.html#28 Top Ten Reasons Why Large Companies Fail To Keep Their Best Talent
https://www.garlic.com/~lynn/2012k.html#23 How to Stuff a Wild Duck
https://www.garlic.com/~lynn/2012k.html#46 Slackware
https://www.garlic.com/~lynn/2012k.html#70 END OF FILE
https://www.garlic.com/~lynn/2012k.html#78 END OF FILE
https://www.garlic.com/~lynn/2012m.html#31 Still think the mainframe is going away soon: Think again. IBM mainframe computer sales are 4% of IBM's revenue; with software, services, and storage it's 25%

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Wed, 03 Oct 2012 15:47:33 -0400
John Levine <johnl@iecc.com> writes:
Well, you know, it's all client server now, with the mainframe being the server. The question is more likely whether they'll switch over to big racks of cheap PCs. Probably not.

A few years ago I was talking to people from ITA software and Orbitz about the guts of their reservation systems. Until then, airline reservations had all been on mainframes, more or less the same design as the original SABER in the early 1960s. ITA's system does very fast searches with racks of cheap PCs searching in parallel, and then combining the results. (You can see why Google bought them.)

But actually making reservations and selling tickets can't be done in parallel, because a ticket is either sold or it isn't, which requires a database with locking, and that needs a system with a small number of fast and reliable processors. So their rez and sales system uses Oracle running on Suns or something like that. If Suns aren't fast enough, you need a mainframe. Big databases are only getting bigger, so airlines and banks and other high performance database systems are going to need mainframes for a long time.

The other thing about mainframes is that their reliabity is phenomenal. I read about a TPF system that was up continuously for ten years, through multiple hardware and software upgrades. The crash after ten years was just bad luck, a hardware failure while the part of the system that normally would have taken over the load was offline for scheduled maintenance, and even then it was back up in less than a minute.


re:
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#14 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#15 System/360--50 years--the future?

well over half of airline reservation system processing is route & fare lookup ... much of which can be done in parallel.

TPF has been function limited system to get its thruput ... airlines had been doing infrastructure & data maintenance on more traditional systems ... and then periodically rebuilding TPF data from scratch (scheduled downtime). 20yrs ago, one of the offspring worked for air freight forwarder where he had access to major airline res systems ... not for seats ... but for space in the belly of the plane. rebuild was typically sunday night a couple times a month ... but sometimes it overan its allocated window impacting scheduling on monday morning.

past posts mentioning rewriting/redoing routes for airline res system
https://www.garlic.com/~lynn/99.html#17 Old Computers
https://www.garlic.com/~lynn/99.html#100 Why won't the AS/400 die? Or, It's 1999 why do I have to learn how to use
https://www.garlic.com/~lynn/99.html#103 IBM 9020 computers used by FAA (was Re: EPO stories (was: HELP IT'S HOT!!!!!))
https://www.garlic.com/~lynn/99.html#136a checks (was S/390 on PowerPC?)
https://www.garlic.com/~lynn/2000f.html#20 Competitors to SABRE?
https://www.garlic.com/~lynn/2001.html#26 Disk caching and file systems. Disk history...people forget
https://www.garlic.com/~lynn/2001d.html#69 Block oriented I/O over IP
https://www.garlic.com/~lynn/2001g.html#45 Did AT&T offer Unix to Digital Equipment in the 70s?
https://www.garlic.com/~lynn/2001g.html#49 Did AT&T offer Unix to Digital Equipment in the 70s?
https://www.garlic.com/~lynn/2001j.html#17 I hate Compaq
https://www.garlic.com/~lynn/2001n.html#0 TSS/360
https://www.garlic.com/~lynn/2001n.html#3 News IBM loses supercomputer crown
https://www.garlic.com/~lynn/2002g.html#2 Computers in Science Fiction
https://www.garlic.com/~lynn/2002g.html#3 Why are Mainframe Computers really still in use at all?
https://www.garlic.com/~lynn/2002h.html#12 Why did OSI fail compared with TCP-IP?
https://www.garlic.com/~lynn/2002h.html#43 IBM doing anything for 50th Anniv?
https://www.garlic.com/~lynn/2002i.html#83 HONE
https://www.garlic.com/~lynn/2002j.html#83 Summary: Robots of Doom
https://www.garlic.com/~lynn/2002m.html#67 Tweaking old computers?
https://www.garlic.com/~lynn/2003.html#48 InfiniBand Group Sharply, Evenly Divided
https://www.garlic.com/~lynn/2003c.html#30 difference between itanium and alpha
https://www.garlic.com/~lynn/2003d.html#67 unix
https://www.garlic.com/~lynn/2003k.html#3 Ping: Anne & Lynn Wheeler
https://www.garlic.com/~lynn/2003n.html#47 What makes a mainframe a mainframe?
https://www.garlic.com/~lynn/2004b.html#6 Mainframe not a good architecture for interactive workloads
https://www.garlic.com/~lynn/2004e.html#44 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004f.html#58 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#14 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004l.html#6 Xah Lee's Unixism
https://www.garlic.com/~lynn/2004o.html#23 Demo: Things in Hierarchies (w/o RM/SQL)
https://www.garlic.com/~lynn/2004o.html#29 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2004p.html#26 IBM 3614 and 3624 ATM's
https://www.garlic.com/~lynn/2004q.html#85 The TransRelational Model: Performance Concerns
https://www.garlic.com/~lynn/2005.html#22 The Soul of Barb's New Machine (was Re: creat)
https://www.garlic.com/~lynn/2005.html#41 something like a CTC on a PC
https://www.garlic.com/~lynn/2005c.html#67 intel's Vanderpool and virtualization in general
https://www.garlic.com/~lynn/2005e.html#47 Using the Cache to Change the Width of Memory
https://www.garlic.com/~lynn/2005f.html#22 System/360; Hardwired vs. Microcoded
https://www.garlic.com/~lynn/2005n.html#44 What was new&important in computer architecture 10 years ago ?
https://www.garlic.com/~lynn/2005o.html#44 Intel engineer discusses their dual-core design
https://www.garlic.com/~lynn/2005q.html#7 HASP/ASP JES/JES2/JES3
https://www.garlic.com/~lynn/2006d.html#5 IBM 610 workstation computer
https://www.garlic.com/~lynn/2006j.html#6 The Pankian Metaphor
https://www.garlic.com/~lynn/2006k.html#9 Arpa address
https://www.garlic.com/~lynn/2006o.html#4 How Many 360/195s and 370/195s were shipped?
https://www.garlic.com/~lynn/2006o.html#18 RAMAC 305(?)
https://www.garlic.com/~lynn/2006r.html#18 50th Anniversary of invention of disk drives
https://www.garlic.com/~lynn/2007e.html#52 US Air computers delay psgrs
https://www.garlic.com/~lynn/2007g.html#22 Bidirectional Binary Self-Joins
https://www.garlic.com/~lynn/2007g.html#41 US Airways badmouths legacy system
https://www.garlic.com/~lynn/2007h.html#41 Fast and Safe C Strings: User friendly C macros to Declare and use C Strings
https://www.garlic.com/~lynn/2007k.html#72 The top 10 dead (or dying) computer skills
https://www.garlic.com/~lynn/2007n.html#8 nouns and adjectives
https://www.garlic.com/~lynn/2008c.html#53 Migration from Mainframe to othre platforms - the othe bell?
https://www.garlic.com/~lynn/2008i.html#19 American Airlines
https://www.garlic.com/~lynn/2008j.html#32 CLIs and GUIs
https://www.garlic.com/~lynn/2008p.html#41 Automation is still not accepted to streamline the business processes... why organizations are not accepting newer technologies?
https://www.garlic.com/~lynn/2009j.html#33 IBM touts encryption innovation
https://www.garlic.com/~lynn/2009l.html#55 IBM halves mainframe Linux engine prices
https://www.garlic.com/~lynn/2009l.html#66 ACP, One of the Oldest Open Source Apps
https://www.garlic.com/~lynn/2009r.html#59 "Portable" data centers
https://www.garlic.com/~lynn/2010b.html#73 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#74 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#80 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010c.html#19 Processes' memory
https://www.garlic.com/~lynn/2010j.html#52 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010m.html#42 IBM 3883 Manuals
https://www.garlic.com/~lynn/2010n.html#16 Sabre Talk Information?
https://www.garlic.com/~lynn/2010n.html#81 Hashing for DISTINCT or GROUP BY in SQL
https://www.garlic.com/~lynn/2011c.html#17 If IBM Hadn't Bet the Company
https://www.garlic.com/~lynn/2011c.html#42 If IBM Hadn't Bet the Company
https://www.garlic.com/~lynn/2011d.html#14 Sabre; The First Online Reservation System
https://www.garlic.com/~lynn/2011d.html#43 Sabre; The First Online Reservation System
https://www.garlic.com/~lynn/2011e.html#8 Multiple Virtual Memory
https://www.garlic.com/~lynn/2011i.html#77 program coding pads
https://www.garlic.com/~lynn/2011n.html#35 Last Word on Dennis Ritchie
https://www.garlic.com/~lynn/2011n.html#92 Innovation and iconoclasm
https://www.garlic.com/~lynn/2012c.html#8 The PC industry is heading for collapse
https://www.garlic.com/~lynn/2012e.html#70 Disruptive Thinkers: Defining the Problem
https://www.garlic.com/~lynn/2012f.html#7 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012k.html#23 How to Stuff a Wild Duck
https://www.garlic.com/~lynn/2012k.html#50 1132 printer history

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Wed, 03 Oct 2012 20:17:03 -0400
hancock4 writes:
I'm a bit confused--are you saying that there are roughly only 180 mainframes out there today? Even considering that a mainframe box today might represent five older machines, that still leaves the equivalent of only 1,000 machines. Back in 1401 days they sold 10,000 of those, and I think they sold more S/360-30's.

re:
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#14 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#15 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#16 System/360--50 years--the future?

there is estimate of 10,000 mainframe systems in the world today ... no breakdown as avg. age of those systems, what kind, etc.

what i referred to was a max. z196 goes for $28M .... and $5B/year in mainframe sales ... which is the equivalent of 180 max z196 machines/year. there is no statement about the breakdown/distribution for the actual machines sold for that $5B ... probably a larger number of smaller (less expensive) mainframe configurations.

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Thu, 04 Oct 2012 10:06:17 -0400
Peter Flass <Peter_Flass@Yahoo.com> writes:
I would assume most customers would *not* buy a maxed-out machine. You always want to have some wiggle roon for upgrades before you have to replace the system.

re:
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#14 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#15 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#16 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#17 System/360--50 years--the future?

part is left-over calculation about upper bound on total mainframe mips in the world ... aggregate being less than the aggregate of processing power in any of the major cloud megadatacenters. if you assume every one of the estimataed 10,000 mainframes was maxed out z196 @50BIPS, that places upper limit of 500TIPS ... but the avg. is probably closer to 1-2BIPS ... placing total aggregate mainframe processing in the world today around 1-2TIPS. 500TIPS would possibly be approx same order of magnitude as some of the individual major cloud megadatacenters.

note that long-time loyal customer of mainframes are big financial institutions that frequently have large number of maximum configured mainframes in single datacenters (frequently sized to complete work in overnight batch window). they could be considered driving factor behind increasing max. number of processors in single configuration ... going from 64 processors with z10 to 80 processors with z196 to 101 processors for ec12.

I've pontificated periodically that many of these institutions spent billions in the 90s on failed re-engineering mainframe overnight batch applications for straight-through processing on large number of parallel killer micors

old posts about doing working on performance optimization for major mainframe customer a little over a decade ago. They had 40+ maxed out mainframes (of the period, max age of their mainframes was 18m) at around $30M (well over $1B). The size/number of mainframes were dedicated by large 500k statement cobol application that ran in overnight batch window and needed to complete in the overnight window (some other financial institutions considered this a "small" operation)
https://www.garlic.com/~lynn/2006u.html#50 Where can you get a Minor in Mainframe?
https://www.garlic.com/~lynn/2007f.html#47 Is computer history taught now?
https://www.garlic.com/~lynn/2007f.html#48 Is computer history taught now?
https://www.garlic.com/~lynn/2007l.html#20 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007u.html#21 Distributed Computing
https://www.garlic.com/~lynn/2007v.html#64 folklore indeed
https://www.garlic.com/~lynn/2008c.html#24 Job ad for z/OS systems programmer trainee
https://www.garlic.com/~lynn/2008d.html#73 Price of CPU seconds
https://www.garlic.com/~lynn/2008l.html#81 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2009d.html#5 Why do IBMers think disks are 'Direct Access'?
https://www.garlic.com/~lynn/2009d.html#14 Legacy clearing threat to OTC derivatives warns State Street
https://www.garlic.com/~lynn/2009e.html#76 Architectural Diversity
https://www.garlic.com/~lynn/2009f.html#55 Cobol hits 50 and keeps counting
https://www.garlic.com/~lynn/2009g.html#20 IBM forecasts 'new world order' for financial services
https://www.garlic.com/~lynn/2009i.html#21 Why are z/OS people reluctant to use z/OS UNIX?
https://www.garlic.com/~lynn/2009s.html#9 Union Pacific Railroad ditches its mainframe for SOA
https://www.garlic.com/~lynn/2010.html#77 Korean bank Moves back to Mainframes (...no, not back)
https://www.garlic.com/~lynn/2010i.html#41 Idiotic programming style edicts
https://www.garlic.com/~lynn/2011c.html#35 If IBM Hadn't Bet the Company
https://www.garlic.com/~lynn/2011c.html#41 If IBM Hadn't Bet the Company
https://www.garlic.com/~lynn/2011e.html#63 Collection of APL documents
https://www.garlic.com/~lynn/2011f.html#32 At least two decades back, some gurus predicted that mainframes would disappear
https://www.garlic.com/~lynn/2011p.html#60 Spontaneous conduction
https://www.garlic.com/~lynn/2012.html#50 Can any one tell about what is APL language
https://www.garlic.com/~lynn/2012i.html#25 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?

--
virtualization experience starting Jan1968, online at home since Mar1970

How to get a tape's DSCB

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: lynn@GARLIC.COM (Anne & Lynn Wheeler)
Subject: Re: How to get a tape's DSCB
Newsgroups: bit.listserv.ibm-main
Date: 4 Oct 2012 08:14:54 -0700
shmuel+gen@PATRIOT.NET (Shmuel Metz , Seymour J.) writes:
No! The channel should pass CCW opcodes[1] on to the controller and let the controller handle them.

[1] Other than TIC.


note that more recent zHPF for FICON with TCW ... batch up multiple channel commands for download. this is similar to the underlying fibre channel standard (originally from late 80s/early 90s) before FICON (inefficiency) was layered on top (from my perspective involved with fibre channel standard ... it appeared that the POK channel engineers that had gotten involved, were trying to revert to ESCON but using the faster fibre channel transfer rate). reference
http://www-03.ibm.com/systems/z/hardware/connectivity/ficon_performance.html
and
ftp://ftp.software.ibm.com/common/ssi/sa/wh/n/zsw03059usen/ZSW03059USEN.PDF

from above:
zHPF improves upon FICON by providing a Transport Control Word (TCW) that facilitates the processing of an I/O request by the channel and the control unit. The TCW has a capability that enables multiple channel commands to be sent to the control unit as a single entity instead of being sent as separate commands as in FICON CCW. In addition, the channel is no longer expected to process and keep track of each individual CCW. Instead, the channel forwards a chain of commands to the control unit to execute. Reduction in this overhead cost increases the maximum I/O rate possible on the channel and improves the utilization of the various sub-components along the path traversed by the I/O request.

... snip ...

original os/360 paradigm was channel programs were built by application (or libraries) and passed to the system with EXCP for execution. the mainframe channel program paradigm uses "real" addresses for execution.

the original virtual machine cp67 in the 60s ... (for SIO simulation) had to scan virtual machine channel programs, making copy of the channel program, that replaced the virtual machine virtual addresses with "real" addresses. when initial VS2 was being built ... it started off with SVS (single virtual storage) ... effectively laying out an MVT in a 16mbyte virtual address space ... minimum changes hacked into the side of MVT to setup the virtual address space table, handle page fault, do paging operations. Possibly the biggest effort for initial mapping of OS/360 to virtual memory was hacking CP67's "CCWTRANS" into EXCP processing to handle scanning of the passed channel program (that had virtual addresses) to create a copy that had real addresses.

EXCPVR call (aka virtual=real) was added to bypass scanning & creating copy ... that invoking application had done a bunch of special stuff to fix virtual pages in real storage and create channel programs with the real addresses ... rather than virtual addresses. reference to use of excpvr
http://pic.dhe.ibm.com/infocenter/zos/v1r12/topic/com.ibm.zos.r12.idas300/efcprs.htm
and
http://pic.dhe.ibm.com/infocenter/zos/v1r12/topic/com.ibm.zos.r12.ieaa600/excpvrm.htm

past posts mentioning EXCPVR
https://www.garlic.com/~lynn/2007q.html#8 GETMAIN/FREEMAIN and virtual storage backing up
https://www.garlic.com/~lynn/2007s.html#2 Real storage usage - a quick question
https://www.garlic.com/~lynn/2008i.html#68 EXCP access methos
https://www.garlic.com/~lynn/2011p.html#118 Start Interpretive Execution

recent posts mentioning zHPF/TCW
https://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#5 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#11 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#13 Intel Confirms Decline of Server Giants HP, Dell, and IBM
https://www.garlic.com/~lynn/2012m.html#28 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Thu, 04 Oct 2012 15:43:20 -0400
scott@slp53.sl.home (Scott Lurndal) writes:
Oil prices actually declined for the three years after the article which I quoted (dated 2008). They've only this year exceeded the 2008 values.

oil spike way over $100/barrel summer of 2008 and then came back down.

griftopia
https://www.amazon.com/Griftopia-Machines-Vampire-Breaking-ebook/dp/B003F3FJS2

had chapter that CFTC required significant positions to play because speculators resulted in wild irrational price swings. Then 19 secret letters were issued allowed specific speculators to play ... accounting for various things ... including the wild spike of oil way over $100 the summer of 2008 (and fall back below).

then senator releases data show oil speculators causing the wild irrational price swings (aka speculators promote volatility, making money on the swing up as well as the swing back down) ... who is then heavily criticized in several corners for providing evidence of the wild irrational price swings caused by speculators

Sen. Bernie Sanders Leaks Oil Trading Data: Americans Have A Right To Know Who Drove Up Gas Prices
http://crooksandliars.com/susie-madrak/sen-bernie-sanders-leaks-oil-trading-
Bernie Sanders Demands Action From Obama On Wall Street Oil 'Gambling'
http://www.huffingtonpost.com/2011/04/29/bernie-sanders-demands-ac_n_855495.html
FIA "shocked and outraged after Senator leaks oil trading data"
http://www.finextra.com/news/Fullstory.aspx?newsitemid=22911

CFTC is now allowing speculators to drive volatility and wild irrational price swings in commodities ... ala Griftopia book and other news items

Drop back a little, Born, head of CFTC, suggests that it needs to regulate CDS (as insurance instruments, there may be a couple trillion at risk ... however as gambling bets, there is now over $700T in CDS ... hundreds times more than any at risk insurance).

So drop back a little more, person is competing with Gerstner to be next head of AMEX ... and looses out, then leaves with his protoge Jamie Dimon. They go to Baltimore and take over a loan business, then over a few years make several acquisitions eventually culminating in Citibank ... which is in violation of Glass-Steagall. Greenspan gives Citi an exemption while the new CEO lobbies washington for repeal of Glass-Steagall (Sec. of treasury who happens to be former chairman of Goldman-Sachs, head of senate banking committee and others). Head of senate banking committee gets repeal included in GLBA. After GLBA passes, sec. of treasury resigns and joins Citi becoming co-chariman.

Various people oppose regulating CDS including Enron. Born is replaced as head of CFTC by wife of the head of senate banking committee, while her husband gets legislation included that prohits CFTC from regulating CDS. When that passes, his wife (head of CFTC) resigns and joins Enron board and member of the audit committee.

Enron and Worldcom explode and congress passes Sarbanes-Oxley supposedly to prevent Enron/Worldcom from happening again; however it requires SEC to do something. Apparently GAO doesn't believe SEC is doing anything and does reports of public company fraudulent financial filings ... even showing uptic after passage of SOX (question, does SOX have 1) no effect on fraudulent filings, 2) encourages fraudulent filings, 3) if it wasn't for SOX there would have been even larger increase in fraudulent filings).
http://www.gao.gov/products/GAO-03-395R
http://www.gao.gov/products/GAO-06-678
https://www.gao.gov/products/gao-06-1079sp

supposedly under SOX, all the executives and auditors would be doing jail time.

Lack of CDS regulation by CFTC goes on to play major role in triple-A rated toxic CDOs, AIG, etc.

recent post with news items about husband&wife involvement in CFTC, Enron, CDS, Glass-Steagall, too-big-to-fail, deregulation, etc
https://www.garlic.com/~lynn/2012c.html#31

past posts mentioning release of oil trading data:
https://www.garlic.com/~lynn/2011k.html#76 FIA shocked and outraged after Senator leaks oil trading data
https://www.garlic.com/~lynn/2011l.html#21 HOLLOW STATES and a CRISIS OF CAPITALISM
https://www.garlic.com/~lynn/2011m.html#2 computer bootlaces
https://www.garlic.com/~lynn/2011m.html#18 computer bootlaces
https://www.garlic.com/~lynn/2011m.html#33 Deloitte sued for $7.6 billion, accused of missing fraud
https://www.garlic.com/~lynn/2011m.html#71 Don't Dump the Volcker Rule Just Because It's Not Perfect
https://www.garlic.com/~lynn/2011n.html#1 Banks Awash in Cash, Which Isn't Good News
https://www.garlic.com/~lynn/2011n.html#80 A Close Look at the Perry Tax Plan
https://www.garlic.com/~lynn/2011o.html#47 Civilization, doomed?
https://www.garlic.com/~lynn/2011o.html#61 Civilization, doomed?
https://www.garlic.com/~lynn/2011o.html#62 Civilization, doomed?
https://www.garlic.com/~lynn/2011p.html#72 The men who crashed the world
https://www.garlic.com/~lynn/2011p.html#130 vampires in financial infrastructure
https://www.garlic.com/~lynn/2012b.html#84 A Conversation with Peter Thiel
https://www.garlic.com/~lynn/2012c.html#13 Study links ultrafast machine trading with risk of crash
https://www.garlic.com/~lynn/2012c.html#25 Goldman Sachs P.R. Chief's Accidental Exit Interview
https://www.garlic.com/~lynn/2012c.html#39 Greek knife to Wall Street
https://www.garlic.com/~lynn/2012d.html#61 Why Republicans Aren't Mentioning the Real Cause of Rising Prices at the Gas Pump
https://www.garlic.com/~lynn/2012g.html#79 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012i.html#92 Naked emperors, holy cows and Libor

--
virtualization experience starting Jan1968, online at home since Mar1970

8-bit bytes and byte-addressed machines

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 8-bit bytes and byte-addressed machines
Newsgroups: comp.arch
Date: Fri, 05 Oct 2012 19:27:35 -0400
Robert Wessel <robertwessel2@yahoo.com> writes:
I have to disagree at least partially with that. While you could allow 32 and 64 bit x86 code to share an address space (at least 4GB of it), transitions and mixing of 32 and 64 bit code is not nearly as easy as it is on S/360.

On S/360..zArch, there is only a single ISA, and you can execute an AGR (64 bit register-to-register add) in any address mode - 24, 31 or 64 bit. Likewise all registers (and all 64 bits of those registers) are available in all address modes. And the application can transition between address modes with a single (unprivileged) instruction (usually a branch, but there's a direct set-mode instruction as well). On x86, such things would require at least some help from the OS.

The evolution of the S/360 OS's is/was rather different too. While there were definitely (major) kernel changes required during the 24-31 and 31-64 bit transitions, the actual evolution of the OSs was much slower. At first many system services remained callable only in the older mode, and the OS support for the longer addressing modes increased over time. But early applications that needed to make use of the larger addressing modes could do so, but needed to build in their own thunking to get to system services (and other subsystems) that didn't support the newer mode yet. And some of that evolution has been pretty slow - there are still a few control blocks (structures) that need to be allocated in 24 bit storage (some 29 years after the first 31 bit models shipped), and many services are still not available in 64 bit mode (even the ability to put code, as opposed to data, above the 2GB line is severely restricted in zOS). Contrast that with 32 or 64 bit Linux or Windows, where the OS is basically one mode or the other, but with a thunking layer for 32 bit code.

In a sense IBM was forced into that approach since a big-bang conversion of something like MVS would have been impossible.


MVS had a spearate problem that motivated some extensions in parallel with larger virtual addressing modes.

MVS inherited a environment that was strongly pointer passing API ... theoretically requiring the services to occupy the same address space as the applications (from real-storage days). The initial transition to virtual memory was VS2/SVS which basically laid it the real-storage "MVT" in a single 16mbyte virtual address space ... with abbreviated support that created the virtual address table and handle page faults (not a whole lot different from running MVT in a 16mbyte virtual machine under cp67). The slight caveat is I/O programs are built in application space with "real addresses" and passed to supervisor for execution via EXCP. Initially SVS borrowed CCWTRANS from CP67 that was used to scan a virtual channel program, creating a copy, replacing the virtual addresses with real addresses (as addition to EXCP processing).

The next step transition from VS2/SVS to VS2/MVS (single virtual stroage to multiple virtual storage), each application got its own 16mbyte virtual address space. However, because of the pointer passing API requirements ... there was an image of the MVS kernel mapped into 8mbyte of every application 16mbyte virtual address space. There was also some number of system services that resided outside the kernel ... that were in their own separate virtual address space ... that continued to support pointer passing API from applications. To support pointer passing API between applications and services between different virtual address space, the COMMON SEGMENT (initially 1mbyte) image was created that occupied every virtual address space (where application and services could allocate information and pass pointers back&forth).

In parallel with this, work began on "370-xa" extensions, 31bit virtual addresssing ... but also multiple virtual address support (access registers) ... leading to description about home, primary, secondary, etc "address spaces". Initially, this would involve application making kernel call for a service ... the kernel would load the service virtual address space and put the application virtual address space as secondary. Eventually lots of services code would be modified to reference pointer data in secondary address space (rather than common segment primary address space).

Before 370-xa ships, larger 370 systems were running into significant common segment constraints ... increasing number of system services in private address space and larger number of concurrent applications resulted with installations having to define 5-6mbyte common segments (leaving only 2-3mbytes of every 16mbyte address space for application) ... and under threat that it might need to increase to 7-8mbyte common segment (possibility that nothing is left for application).

The person that later goes on to be major force behind Itanium, retrofits part of 370-xa to high-end "3033" 370 as "dual-address-space" mode ... allowing some MVS services to get an early start on accessing application information as secondary address space rather than CSA (originally "common segment area " ... later "common system area")

part of full 370-xa extensions was also program call&return. Kernel would setup table for program call ... that had virtual address space pointer for each services. Eventually application could do a "program call" which would perform the switch to the specific services address space ... moving calling application address space to secondary (effectively almost library call/return ... except there is virtual address space fiddling goes on at the same time).

Parts of MVS is still stuck with 24bit mode ... in this case supporting "Format 0 CCWs" ... which is 24bit address fields. ... furthermore channel command words ... are (24bit) "real" addresses
http://pic.dhe.ibm.com/infocenter/zos/v1r12/topic/com.ibm.zos.r12.ieaa600/excpvrm.htm

this is discussion of different kinds of addresses
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dz9zr003/3.2.1?SHELF=&DT=20040504121320&CASE=

this is discussion of how program table works:
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dz9zr003/5.5?DT=20040504121320

possibly more than you want to know about access register
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dz9zr003/5.7?DT=20040504121320

description of program call instruction
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dz9zr003/10.34?DT=20040504121320

program call allows switching to services virtual address space by hardware w/o having the overhead of going through software processing of kernel call.

--
virtualization experience starting Jan1968, online at home since Mar1970

Preparing for War with China

Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 06 Oct 2012
Subject: Preparing for War with China
Blog: Boyd Strategy
re:
http://lnkd.in/nFRhH3
and
https://www.garlic.com/~lynn/2012l.html#44 Preparing for War with China

a couple recent items

LRASM Missiles: Reaching for a Long-Reach Punch
http://www.defenseindustrydaily.com/LRASM-Missiles-Reaching-for-a-Long-Reach-Punch-06752/

from above:
The US Navy is beginning to acknowledge a growing problem that threatens its freedom of the seas: its strike reach is shrinking and aging, while potential opponents' attack reach is expanding and modernizing.

... snip ...

Goes along with reports that MICC got enormous budget increases last decade, half for the wars and its is not clear where the other half went since forces got smaller and avg. equipment age increased.
http://nation.time.com/2012/10/01/adventures-in-babbleland-desperate-rhetoric-for-mundane-times/
and
http://nation.time.com/2012/10/02/adventures-in-babbleland-technological-bloat/

The Navy's New Class of Warships: Big Bucks, Little Bang
http://nation.time.com/2012/10/05/the-navys-new-class-of-warships-big-bucks-little-bang/

from above:
The Navy's new Littoral Combat Ship (LCS) is not only staggeringly overpriced and chronically unreliable but -- even if it were to work perfectly -- cannot match the combat power of similar sized foreign warships costing only a fraction as much.

... snip ...

--
virtualization experience starting Jan1968, online at home since Mar1970

Jedi Knights

Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 06 Oct 2012
Subject: Jedi Knights
Blog: Boyd Strategy
re:
http://lnkd.in/xZigE
and
https://www.garlic.com/~lynn/2012n.html#10 Jedi Knights

fyi, couple pictures Boyd's display case at Quantico
https://www.facebook.com/photo.php?fbid=1640474247359&l=142bf5f95d
and
https://www.facebook.com/photo.php?fbid=1640474287360&l=777294a852

as mentioned various times previously ... I was fortunate enough to sponsor Boyd's briefings at IBM .... and forced :) to spend some amount of time alone with him, more intense than the IBM executive I reported to (making me his direct report was possibly punishment for both of us).

some cross-over from this discussion here:
http://lnkd.in/9q5Zvq

One of my co-workers had called up Chuck Spinney after seeing this article ... gone behind paywall, but lives (mostly) free at wayback machine (except pg8)
https://web.archive.org/web/20070320170523/http://www.time.com/time/magazine/article/0,9171,953733,00.html
also
https://content.time.com/time/magazine/article/0,9171,953733,00.html

Spinney suggested he talk to Boyd (Boyd would tell story that SECDEF blamed Boyd for the article and tried to have him transferred to Alaska and banned from the Pentagon for life). The co-worker then con'ed me into sponsoring Boyd's briefings ... which I then did a number of times.

Spinney's tribute to Boyd
http://radio-weblogs.com/0107127/stories/2002/12/23/genghisJohnChuckSpinneysBioOfJohnBoyd.html

other recent posts mentioning above:
https://www.garlic.com/~lynn/2012b.html#75 The Winds of Reform
https://www.garlic.com/~lynn/2012b.html#93 Original Thinking Is Hard, Where Good Ideas Come From
https://www.garlic.com/~lynn/2012c.html#51 How would you succinctly desribe maneuver warfare?
https://www.garlic.com/~lynn/2012d.html#36 McCain calls for U.S.-led airstrikes in Syria
https://www.garlic.com/~lynn/2012e.html#25 We are on the brink of historic decision [referring to defence cuts]
https://www.garlic.com/~lynn/2012e.html#68 Glory Days of Army Acquisition Were Not So Glorious
https://www.garlic.com/~lynn/2012e.html#70 Disruptive Thinkers: Defining the Problem
https://www.garlic.com/~lynn/2012e.html#72 Sunday Book Review: Mind of War
https://www.garlic.com/~lynn/2012f.html#5 Hardware v. People
https://www.garlic.com/~lynn/2012f.html#23 Time to competency for new software language?
https://www.garlic.com/~lynn/2012f.html#31 Rome speaks to us. Their example can inspire us to avoid their fate
https://www.garlic.com/~lynn/2012f.html#88 Defense acquisitions are broken and no one cares
https://www.garlic.com/~lynn/2012h.html#17 Hierarchy
https://www.garlic.com/~lynn/2012h.html#21 The Age of Unsatisfying Wars
https://www.garlic.com/~lynn/2012h.html#24 Baby Boomer Guys -- Do you look old? Part II
https://www.garlic.com/~lynn/2012h.html#63 Is this Boyd's fundamental postulate, 'to improve our capacity for independent action'?
https://www.garlic.com/~lynn/2012h.html#68 Interesting News Article
https://www.garlic.com/~lynn/2012i.html#0 Interesting News Article
https://www.garlic.com/~lynn/2012i.html#51 Is this Boyd's fundamental postulate, 'to improve our capacity for independent action'? thoughts please
https://www.garlic.com/~lynn/2012k.html#40 Core characteristics of resilience
https://www.garlic.com/~lynn/2012k.html#79 Romney and Ryan's Phony Deficit-Reduction Plan
https://www.garlic.com/~lynn/2012l.html#2 Cultural attitudes towards failure
https://www.garlic.com/~lynn/2012l.html#45 Introducing John Boyd
https://www.garlic.com/~lynn/2012l.html#58 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012m.html#64 Guest Post: Beakley on Boyd, Aerial Combat and the OODA-Loop

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Sun, 07 Oct 2012 09:11:12 -0400
Anne & Lynn Wheeler <lynn@garlic.com> writes:
I've pontificated periodically that many these institutions spent billions in the 90s on failed re-engineering mainframe overnight batch applications for straight-through processing on large number of parallel killer micors

re:
https://www.garlic.com/~lynn/2012n.html#18 System/360--50 years--the future?

the nearly 500k cobol statement overnight batch application throughput was primary thing dictated the 40+ maxed out mainframes @$30M ... comes to over $1.2B ... and with no system over 18m ... amounted to over $800m/yr (in just hardware sales, with multiplier of six, cames out to mainframe revenue of over $3.2B).

the application had a very large performance group helping with the application's care&feeding for a number of decades ... so they were somewhat surprised when I identified 14% improvement (represented well over $100M/yr in just hardware) ... this was little over decade ago.

there were other financial industry operations with even larger number of max'ed out mainframes ... half-dozen such operations easily accounts for over half of mainframe revenue. in much the same way that big cloud megadatacenters are driving many characteristics of server evolution ... the big financial industry mainframe operations help drive evolution of maxed configuration mainframes.

this would also help explain that it was somebody from the financial industry that was brought in for the resurrection of IBM. misc. recent posts mentioning resurrection of IBM.
https://www.garlic.com/~lynn/2012.html#92 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012.html#103 Google works on Internet standards with TCP proposals, SPDY standardization
https://www.garlic.com/~lynn/2012.html#104 Can a business be democratic? Tom Watson Sr. thought so
https://www.garlic.com/~lynn/2012b.html#12 Sun Tzu, Boyd, strategy and extensions of same
https://www.garlic.com/~lynn/2012b.html#41 Are rotating register files still a bad idea?
https://www.garlic.com/~lynn/2012b.html#59 Original Thinking Is Hard, Where Good Ideas Come From
https://www.garlic.com/~lynn/2012b.html#74 IBM Doing Some Restructuring?
https://www.garlic.com/~lynn/2012d.html#23 IBM cuts more than 1,000 U.S. Workers
https://www.garlic.com/~lynn/2012d.html#35 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012f.html#84 How do you feel about the fact that India has more employees than US?
https://www.garlic.com/~lynn/2012g.html#4 Hard drives: A bit of progress
https://www.garlic.com/~lynn/2012g.html#34 Co-existance of z/OS and z/VM on same DASD farm
https://www.garlic.com/~lynn/2012g.html#58 Cartons of Punch Cards
https://www.garlic.com/~lynn/2012g.html#72 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012g.html#74 Why So Many Formerly Successful Companies Are Failing
https://www.garlic.com/~lynn/2012g.html#82 How do you feel about the fact that today India has more IBM employees than US?
https://www.garlic.com/~lynn/2012g.html#87 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#4 Think You Know The Mainframe?
https://www.garlic.com/~lynn/2012h.html#16 Hierarchy
https://www.garlic.com/~lynn/2012h.html#35 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#45 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#54 How will mainframers retiring be different from Y2K?
https://www.garlic.com/~lynn/2012h.html#55 The Invention of Email
https://www.garlic.com/~lynn/2012i.html#25 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#69 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012k.html#19 SnOODAn: Boyd, Snowden, and Resilience
https://www.garlic.com/~lynn/2012k.html#21 Is there a connection between your strategic and tactical assertions?
https://www.garlic.com/~lynn/2012k.html#31 History--punched card transmission over telegraph lines
https://www.garlic.com/~lynn/2012k.html#34 History--punched card transmission over telegraph lines
https://www.garlic.com/~lynn/2012k.html#49 1132 printer history
https://www.garlic.com/~lynn/2012k.html#70 END OF FILE
https://www.garlic.com/~lynn/2012l.html#12 First Battle: Operation Starlite and the Beginning of the Blood Debt in Vietnam
https://www.garlic.com/~lynn/2012l.html#20 X86 server
https://www.garlic.com/~lynn/2012l.html#63 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012l.html#69 Cultural attitudes towards failure
https://www.garlic.com/~lynn/2012n.html#8 General Mills computer

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Sun, 07 Oct 2012 09:25:02 -0400
Walter Bushell <proto@panix.com> writes:
And I've read that 4% of IBMs revenues come from these. I would imagine much more of the profits are from these.

re:
https://www.garlic.com/~lynn/2012n.html#18 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#24 System/360--50 years--the future?

this post
https://www.garlic.com/~lynn/2012m.html#31 Still think the mainframe is going away soon: Think again. IBM mainframe computer sales are 4% of IBM's revenue; with software, services, and storage it's 25%

is in this linkedin Enterprise discussion
http://lnkd.in/mjYX6H

which points out that the total mainframe is 25% of company revenue (as well as 40% of the company profit) ... i.e. company reaps $6.25 (aka 25/4) in total revenue from each dollar of mainframe computer sales (and nearly half its profit).

other posts in this thread:
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#14 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#15 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#16 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#17 System/360--50 years--the future?

other posts mentioning the 6.25 multiplier
https://www.garlic.com/~lynn/2012l.html#52 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#81 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#67 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?

--
virtualization experience starting Jan1968, online at home since Mar1970

Is there a correspondence between 64-bit IBM mainframes and PoOps editions levels?

Refed: **, - **, - **, - **, - **, - **
From: lynn@GARLIC.COM (Anne & Lynn Wheeler)
Subject: Re: Is there a correspondence between 64-bit IBM mainframes and PoOps editions levels?
Newsgroups: bit.listserv.ibm-main
Date: 7 Oct 2012 09:13:39 -0700
shmuel+ibm-main@PATRIOT.NET (Shmuel Metz , Seymour J.) writes:
Every generation believes that it invented sex. The 4300 may mark MVCIN becoming standard, but the instruction is much older than the 4300. The Technion had it on their 370/165 in 1972, and I believe that it was available on the 360/50.

there were lots of special instructions done for 360/50 ... the ibm boston programming center was doing conversational programming system which would include support for new instructions implemented on 360/50. some amount of the work was subcontracted to allen-babcock
http://www.bitsavers.org/pdf/allen-babcock/cps/CPS_Progress_Report_may66.pdf

lists (new microcode for 360/50):
EVAL Evaluate an arithmetic expression

CHB one byte item list search operation CHBE

CHH two byte item list search operation CHHE

LDM load multiple floating point STDM store multiple floating point

LUM load under mask

STUM store under mask

TAC Table ANDed Characters (extensions of translate&test) TOC Table ORed Characters

BS Binary Search

ADD Floating decimal instructions SUB COMP MULT DIV


...

IBM Boston Programming Center was on 3rd flr of 545 tech sq. When the cp67 group split off from the science center (on the 4th flr) and also started the morph from cp67 to vm370 ... they moved to the 3rd flr and took over the Boston Programming Center (including the CPS people) ... while they were now working on vm370/cms ... they did do a port of CPS to CMS (aka CPS could run on plain 360 w/o the new instructions).

The original virtual machine implementation had also been targeted for 360/50 ... (before 360/67 was available standard with virtual memory) ... but since nearly all spare 360/50s were going to the FAA Air Traffic Control effort (which had a whole different set of modifications), the science center had to settle for 360/40 ... where they did the hardware changes to provide virtual memory support ... and implemented virtual machine cp/40 (and lots of early cms development was done in parallel on the "bare" 360/40). Later when 360/67 was available, cp/40 morphed into cp/67.

as the vm370 group on the 3rd flr expanded ... they outgrew the space and moved out to the old SBC building in burlington mall (SBC having been transferred to CDC as part of settling some litigation). also as part of morph of cp67/cms to vm370/cms ... the ability for cms to execute on a real machine was crippled.

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Sun, 07 Oct 2012 19:49:56 -0400
hancock4 writes:
A lot of mainframe programmers weren't aware of some ways to improve batch performance. While not significant when processing relatively small files, large and complex processes would benefit. Heck, even I could suggest inprovements. These included proper use of COMP-3, COMP, and INDEXED in COBOL programs, use of certain coding techniques as recommended in the COBOL Programmers Guide*; use of a compiled COBOL program instead of a 4th generation program like Easytrieve, proper block size and record length, use of SYNCSORT to select off desired records before processing, for very records, reformat to a smaller length early in the process when practical. In certain situations an assembler subroutine was faster for certain binary efforts. For data base access, working with knowledgeable DBAs about the best way to access and update records is helpful. There are also systems programming issues, such as avoiding channel and head contention in files and ensuring adequate real core for buffers and programs.

re:
https://www.garlic.com/~lynn/2012n.html#24 System/360--50 years--the future?

the pretty much did all that stuff ... also frequently ran instruction address sampling looking for code "hot-spots" ... which would get extensive review for additional code rewrite/optimization.

not long before I got involved, they had also contracted with a consultant that had a sophisticated system model ... that had been fairly successful identifying performance issues at a number of other large mainframe shops. I've told this story periodically before, the system modeling had been acquired from IBM in the early 90s ... implemented in APL ... and run thru an APL to C-language converter. This was a many times descendent of the performance predictor that had originated at the science center and made available on HONE system (in mid-70s) for marketing people to use for "what-if" scenarios ... input detailed customer configuration and workload and ask what-if effects if configuration and/or workload changed (i.e. sell customer an extra megabyte or two of memory, add more disk arms, etc). misc. past posts mentioning hone
https://www.garlic.com/~lynn/subtopic.html#hone

In the 60s and 70s, science center had done lots of system monitoring and performance work ... instruction hot-spot monitoring (both detailed traces and sampling), workload profiling, detailed system simulation, analytic (APL) system modeling ... some of the stuff eventually evolves into capacity planning. There was also a lot of work with multiple regression analysis using huge amounts of system monitor data (decade of science center monitored data ... plus a few yrs from large number of different internal systems). misc. past posts mentioning science center
https://www.garlic.com/~lynn/subtopic.html#545tech

example of instrumenting vm370 kernel for instruction sequence hot-spot ... but was also augmented with instruction address sampling
https://www.garlic.com/~lynn/94.html#21 370 ECPS VM microcode assist

i leveraged a variation of the performance predictor doing my resource manager benchmarking ... misc. past refs
https://www.garlic.com/~lynn/submain.html#benchmark

in any case, i suggested i might be able to find something at the macro-level with multiple regression analysis ... that they were not seeing at the micro-level with all the coding techniques, hot-spot monitoring and system modeling.

what i found was a large application feature/function that multiple regression analysis showed was accounting for 21% of total application usage ... which nobody could explain or account for. The function was complex operation which invoked a large number of different application features ... each one of the individual application features had been extensively optimized. The eventual explanation was that the function was repeatedly executed three times for each operation ... when it should have only been invoked once. Elimination of the extra, unnecessary two iterations represented 14% savings.

misc. past posts mentioning multiple regression analysis
https://www.garlic.com/~lynn/2002l.html#62 Itanium2 performance data from SGI
https://www.garlic.com/~lynn/2004g.html#9 Need to understand difference between EBCDIC and EDCDIC
https://www.garlic.com/~lynn/2004k.html#31 capacity planning: art, science or magic?
https://www.garlic.com/~lynn/2005d.html#6 [Lit.] Buffer overruns
https://www.garlic.com/~lynn/2005k.html#17 More on garbage collection
https://www.garlic.com/~lynn/2005n.html#18 Code density and performance?
https://www.garlic.com/~lynn/2006f.html#22 A very basic question
https://www.garlic.com/~lynn/2006g.html#4 The Pankian Metaphor
https://www.garlic.com/~lynn/2006o.html#23 Strobe equivalents
https://www.garlic.com/~lynn/2006s.html#24 Curiousity: CPU % for COBOL program
https://www.garlic.com/~lynn/2006t.html#28 Why these original FORTRAN quirks?
https://www.garlic.com/~lynn/2006u.html#50 Where can you get a Minor in Mainframe?
https://www.garlic.com/~lynn/2006x.html#27 The Future of CPUs: What's After Multi-Core?
https://www.garlic.com/~lynn/2007g.html#71 PAAppViewer3 (AppViewer3)?
https://www.garlic.com/~lynn/2007u.html#21 Distributed Computing
https://www.garlic.com/~lynn/2008c.html#24 Job ad for z/OS systems programmer trainee
https://www.garlic.com/~lynn/2008l.html#81 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2009d.html#5 Why do IBMers think disks are 'Direct Access'?
https://www.garlic.com/~lynn/2009r.html#17 How to reduce the overall monthly cost on a System z environment?
https://www.garlic.com/~lynn/2010j.html#81 Percentage of code executed that is user written was Re: Delete all members of a PDS that is allocated
https://www.garlic.com/~lynn/2011p.html#60 Spontaneous conduction

--
virtualization experience starting Jan1968, online at home since Mar1970

390 vector instruction set reuse, was 8-bit bytes

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 390 vector instruction set reuse, was 8-bit bytes
Newsgroups: comp.arch
Date: Mon, 08 Oct 2012 10:41:54 -0400
Robert Wessel <robertwessel2@yahoo.com> writes:
The VF was introduced before the real end of the vector supercomputer era, and was not really that far off be competitive with the performance of contemporary Cray-2 or Cray-Y/MP. The ES/9000 facility also benchmarked passably. Of course the market for vector supercomputers did collapse just a few years later, although I think people were just starting to have their doubts by 1986 when the 3090s VF shipped for the 3090s, and certainly by the time the ES/9000s came online. I suspect this was more timing - this got started several years earlier, pretty much in the heyday of the vector supers (which were high profile, and big bucks), and much of the work to add the VF to the later ES/9000s would have been done while the market was still semi-healthy.

I have recollection of 3090 processor engineers claiming that scalar floating point had been optimized to the point that it ran at memory speed ... vector would get additional flops for only relatively few codes (slow floating point required large number of parallel units to saturate memory bus, fast scalar floating point could saturate memory bus) ... vector having turned into pure marketing

something similar for attaching HiPPI (100mbyte/sec) channel to 3090 ... it was something of rube goldberg hack since standard channel interface was incapable of that rate ... they had to hack into side of expanded store bus and I/O programming was done with peak/poke paradigm as expanded store bus transfer operations (4k byte transfer instruction).

a supercomputer effort was also started about the 3090 time-frame in kingston (separate from engineering & scientific lab) to design a "supercomputer" as well as getting involved with funding Chen's effort. In oct '91, the senior executive sponsoring the kingston supercomputer effort retired and kingston (and some number of his other efforts) were audited. There was then launched an effort to scour the company for something to be used as supercomputer.

I had been doing cluster scale-up for ha/cmp ... both numeric intensive and commercial ... old post mentioning early jan92 meeting in ellison's conference room regarding cluster scale-up for RDBMS.
https://www.garlic.com/~lynn/95.html#13

old email discussing cluster scale-up ... working with national labs, engineering&scientific support, commercial support, etc
https://www.garlic.com/~lynn/lhwemail.html#medusa

within hrs of the last email in above (end jan92), the effort is transferred to kingston and we are told we can't work on anything with more than four processors. within weeks it is announced as supercomputer for numeric and scientific *only*
https://www.garlic.com/~lynn/2001n.html#6000clusters1 920217
... scientific and technical only
and then
https://www.garlic.com/~lynn/2001n.html#6000clusters2 920511
... caught by surprise (!?)

we then decide to leave ... some of the executives involved in above had also been involved in preventing us from bidding on NSFNET backbone (tcp/ip is technology basis for modern internet, NSFNET backbone was operational basis for modern internet, CIX was business basis for modern internet), director of NSF tries to help by writing the company a letter 3Apr1986, NSF Director to IBM Chief Scientist and IBM Senior VP and director of Research, copying IBM CEO), but that just made the internal politics worse. some old email
https://www.garlic.com/~lynn/lhwemail.html#nsfnet

misc past posts mentioning working with engineering & scientific lab in kingston with lots FPS processors starting slightly before 3090 and then overlapped
https://www.garlic.com/~lynn/2000c.html#5 TF-1
https://www.garlic.com/~lynn/2000c.html#61 TF-1
https://www.garlic.com/~lynn/2001b.html#56 Why SMP at all anymore?
https://www.garlic.com/~lynn/2001m.html#25 ESCON Data Transfer Rate
https://www.garlic.com/~lynn/2002e.html#31 Hardest Mistake in Comp Arch to Fix
https://www.garlic.com/~lynn/2002j.html#30 Weird
https://www.garlic.com/~lynn/2003b.html#29 360/370 disk drives
https://www.garlic.com/~lynn/2003d.html#35 Why only 24 bits on S/360?
https://www.garlic.com/~lynn/2003m.html#20 360 Microde Floating Point Fix
https://www.garlic.com/~lynn/2006m.html#4 The Power of the NORC
https://www.garlic.com/~lynn/2010b.html#71 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#72 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010f.html#47 Nonlinear systems and nonlocal supercomputing
https://www.garlic.com/~lynn/2010f.html#61 Handling multicore CPUs; what the competition is thinking
https://www.garlic.com/~lynn/2011h.html#74 Vector processors on the 3090
https://www.garlic.com/~lynn/2011n.html#36 Last Word on Dennis Ritchie

--
virtualization experience starting Jan1968, online at home since Mar1970

Jedi Knights

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 08 Oct 2012
Subject: Jedi Knights
Blog: Boyd Strategy
re:
http://lnkd.in/xZigEc
and
https://www.garlic.com/~lynn/2012n.html#10 Jedi Knights
https://www.garlic.com/~lynn/2012n.html#23 Jedi Knights

mentions a son-in-law did two tours in the "2nd" Iraq (really long-winded discussion) ... Cultural attitudes towards failure
http://lnkd.in/9q5Zvq
and
https://www.garlic.com/~lynn/2012m.html#49 Cultural attitudes towards failure

Al Gray (& others) have tried hard ... this sort of plays some of it out
https://web.archive.org/web/20160326155408/https://www.mca-marines.org/gazette/attritionist-letters-archives

there are lots of conflicting interests. One such story is that DOD/Army forced Marines into taking Abrams main battle tanks ... Army budget had been cut below the tank discount threshold ... so it needed Marines to take enough tanks to get aggregate numbers for the per tank discount. Something like 90% of Marine mission profiles have 35ton load limit ... well below Abrams 65-70 tons.

mention that while Gray wasn't on the agenda, he wandered in last year and talked for two hrs ... totally blowing the schedule (again Culture attitudes towards failure discussion)
http://lnkd.in/9q5Zvq
and
https://www.garlic.com/~lynn/2012k.html#87 Cultural attitudes towards failure
also mentioned in this post in another fora
https://www.garlic.com/~lynn/2012b.html#38

Steele has recent references to Gray here
http://www.phibetaiota.net/2012/10/search-threat-matrix/
and here
http://www.phibetaiota.net/2012/10/robert-steele-introducing-dr-greg-newby-and-the-multinational-open-source-arctic-innovation-consortium-mosaic/

totally unrelated ... in one of the social media items referencing the above, I mentioned that if you are on the grid computing mailing Greg had been editor for quite awhile.

here I mention running into Steele periodically over the past couple decades and I would mention Boyd quotes during his talks ... to the extent he started to refer to me as "that Boyd person"
http://lnkd.in/ahxq5r
and
https://www.garlic.com/~lynn/2012i.html#51 Is this Boyd's fundamental postulate, 'to improve our capacity for independent action'? thoughts please

Of course now, he frequently references Spinney in his blog.
http://www.phibetaiota.net/?s=spinney

latest Steele/Spinney reference from today:
http://www.phibetaiota.net/2012/10/chuck-spinney-predictable-meltdown-in-afghanistan-strategic-decrepitude-and-lack-of-integrity-go-hand-in-hand/

above mentions roots of "perpetual war". x-over from this boyd discussion "What a Caveman Can Teach You About Strategy"
http://lnkd.in/TYahQ9
and
https://www.garlic.com/~lynn/2012l.html#97 What a Caveman Can Teach You About Strategy

references that the basis for the "Small Wars Manual"
https://en.wikipedia.org/wiki/Small_Wars_Manual

was written about the same time & based on same/similar experience as "War is a Racket"
https://en.wikipedia.org/wiki/War_Is_a_Racket

War is a Racket references:
https://en.wikipedia.org/wiki/Perpetual_war

which references Sprey, Spinney, et all ... also referenced here:
http://chuckspinney.blogspot.com/p/domestic-roots-of-perpetual-war.html

That explains why Army got M1s ... doesn't change that Marines find themselves with enough M1s so the total number (Army+Marines) come up to the number needed for the volume discount. Also set up some part of Marines on trajectory that is little different than adjunct to Army. One issue has been Marines justify unique roles that keeps them from eliminated as different organization.

There are claims that some amount of European infrastructure rebuild last century was specifically to handle M1 loads. That is tempting to drift into subject about how similar infrastructure in US was neglected during the period ... "Confidence Men" has a civil engineering professor talking to Volcker about last several decades civil engineering programs at univ. being eliminated (lack of building programs, lack of jobs for civil engineering graduates, no jobs means drop off in students) ... and Volcker responds: "Well, I said, 'The trouble with the United States recently is we spent several decades not producing many civil engineers and producing a huge number of financial engineers. And the result is s*tty bridges and a s*tty financial system!". Some number of recent US rebuild projects have gone to companies from foreign countries that are still producing civil engineering graduates.

--
virtualization experience starting Jan1968, online at home since Mar1970

General Mills computer

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: General Mills computer
Newsgroups: alt.folklore.computers
Date: Mon, 08 Oct 2012 20:11:13 -0400
hancock4 writes:
Back in the 1980s many shops wanted to improve programmer productivity. To do so, they pushed or mandated the shorthand languages (4GL) like Easytrieve, SAS, DYL, MarkIV, etc.

But as you said, if it's not a good fit, there will be trouble. I remember applications using very large files ran very poorly on a shorthand compared to COBOL, but management demanded everything be written in the 4GL. (The COBOL Report Writer feature was not included as a shortcut tool).


virtual machine service bureaus ... sort of the '60s "cloud service"

A Brief History of Fourth Generation Languages
http://www.decosta.com/Nomad/tales/history.html

from above
One could say PRINT ACROSS MONTH SUM SALES BY DIVISION and receive a report that would have taken many hundreds of lines of Cobol to produce. The product grew in capability and in revenue, both to NCSS and to Mathematica, who enjoyed increasing royalty payments from the sizable customer base.

... snip ...

past posts mentioning ramis, nomad, focus:
https://www.garlic.com/~lynn/2003d.html#15 CA-RAMIS
https://www.garlic.com/~lynn/2003d.html#17 CA-RAMIS
https://www.garlic.com/~lynn/2003n.html#15 Dreaming About Redesigning SQL
https://www.garlic.com/~lynn/2006k.html#37 PDP-1
https://www.garlic.com/~lynn/2007c.html#12 Special characters in passwords was Re: RACF - Password rules
https://www.garlic.com/~lynn/2007e.html#37 Quote from comp.object
https://www.garlic.com/~lynn/2009k.html#40 Gone but not forgotten: 10 operating systems the world left behind
https://www.garlic.com/~lynn/2010e.html#54 search engine history, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010e.html#55 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
https://www.garlic.com/~lynn/2010e.html#58 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
https://www.garlic.com/~lynn/2010n.html#21 What non-IBM software products have been most significant to the mainframe's success
https://www.garlic.com/~lynn/2010o.html#26 Global Sourcing with Cloud Computing and Virtualization
https://www.garlic.com/~lynn/2010q.html#63 VMSHARE Archives
https://www.garlic.com/~lynn/2011d.html#55 Maybe off topic
https://www.garlic.com/~lynn/2011m.html#69 "Best" versus "worst" programming language you've used?
https://www.garlic.com/~lynn/2011n.html#39 Beyond Patriot? The Multinational MEADS Air Defense Program
https://www.garlic.com/~lynn/2011p.html#1 Deja Cloud?
https://www.garlic.com/~lynn/2011p.html#43 z/OS's basis for TCP/IP
https://www.garlic.com/~lynn/2012b.html#60 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012d.html#51 From Who originated the phrase "user-friendly"?
https://www.garlic.com/~lynn/2012e.html#84 Time to competency for new software language?

--
virtualization experience starting Jan1968, online at home since Mar1970

390 vector instruction set reuse, was 8-bit bytes

Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 390 vector instruction set reuse, was 8-bit bytes
Newsgroups: comp.arch
Date: Mon, 08 Oct 2012 23:26:31 -0400
Anne & Lynn Wheeler <lynn@garlic.com> writes:
and then
https://www.garlic.com/~lynn/2001n.html#6000clusters2 920511
... "caught by surprise" (!?)


re:
https://www.garlic.com/~lynn/2012n.html#28 390 vector instruction set reuse, was 8-bit bytes

referenced item is in this post with a bunch of other tidbits about numerical intensive from late 80s and early 90s:
https://www.garlic.com/~lynn/2001n.html#70 CM-5 Thinking Machines, Supercomputers

other old comp.arch posts on subject:
https://www.garlic.com/~lynn/2000c.html#4 TF-1
https://www.garlic.com/~lynn/2000c.html#6 TF-1
https://www.garlic.com/~lynn/2000c.html#9 Cache coherence [was Re: TF-1]
https://www.garlic.com/~lynn/2000c.html#12 Cache coherence [was Re: TF-1]
https://www.garlic.com/~lynn/2000c.html#21 Cache coherence [was Re: TF-1]
https://www.garlic.com/~lynn/2000c.html#22 Cache coherence [was Re: TF-1]

--
virtualization experience starting Jan1968, online at home since Mar1970

390 vector instruction set reuse, was 8-bit bytes

Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 390 vector instruction set reuse, was 8-bit bytes
Newsgroups: comp.arch
Date: Tue, 09 Oct 2012 21:27:24 -0400
Quadibloc <jsavard@ecn.ab.ca> writes:
It's after the idea of a successor to the 370/195 was abandoned, and IBM saw its future as going with the 3033 instead, and so Gene Amdahl quit that IBM largely abandoned _that_ idea, though - so I don't know whether the vector instructions you note came before or after this happened.

re:
https://www.garlic.com/~lynn/2012n.html#28 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#31 390 vector instruction set reuse, was 8-bit bytes

370 was going to be completely replaced by "future system" ... as different from 360/370 as different as 360 was previous generations ... major motivation for "future system" was countermeasure to clone controllers ... which have very tight integration. had lots of horribly complex stuff ... much of it was blue sky w/o much substance. internal politics was killing off 370 products ... and the lack of 370 products during the period was credited with giving clone processors market footfold.

Amdahl gave talk in early 70s in large mit auditorium about starting his new clone processor company. one of the questions was what argument did he use to get funding. his comment was that customers had already invested enormous amounts in 360/370 software applications ... that even if ibm were to completely walk away from 360/370 (might be considered veiled reference to future system) ... that install base would be sufficient to keep him in business until the end of the century.

eventually day of reckoning for future system and it was killed ... followed by mad rush to get products back into 370 pipelines. one of the nails in the future system coffing was evaluation that a future system made from fastest available hardware (370/195), applications running on the machine would have thruput of 370/145 (about 10-30 times slow down) ... future system architecture had enormous hardware complexity. misc. past posts mentioning future system
https://www.garlic.com/~lynn/submain.html#futuresys

3033 started out being 168-3 wiring logic remapped to 20% faster chips (that were being developed for FS) ... chips also had ten times the circuits than used for 168-3 chips ... initially went unused. some logic rework leveraging higher circuit density got 3033 to 50% faster.

in parallel with Q&D effort to turn-out 3033 ... work started 370/xa and 3081 ... also using some warmed-over FS technology ... more discussion here:
http://www.jfsowa.com/computer/memo125.htm

370/195 peak was 10mips ... but lots of code ran 5mips ... in part because most branches stalled pipeline. this in part motivated design for hyperthreaded 370/195 ... emulated two processors with instructions in pipeline one bit tag as to which instruction stream (two i-streams @5mip keeping pipeline operating aggregate 10mip) ... this never shipped to customers. 370/168-3 was nominally 3mips ... and 3033 ran around 4.5mips. 3081s were going to be dual-processor only ... initial 3081d claimed to be two 5mip processors ... but there were lots of applications ran slower on 3081d processor than on 3033. cache size was doubled for 3081k supposedly giving two 7mip processors ... but lots of stuff was more like 5mip. also as detailed in memo125.htm ... 3081 technology compared poorly with clone processors.

it wasn't really until 3090 that could consider new design/implementation

ibm had earlier external box ... 3838 array processor announced 1976, couple past posts
https://www.garlic.com/~lynn/2007h.html#34 GA24-3639
https://www.garlic.com/~lynn/2011j.html#9 program coding pads
and
https://www.garlic.com/~lynn/2011h.html#74 Vector processors on the 3090

also above mentions 3090vf was announced 1oct1985. and before 3838 there was 2938 array processor (1969, also external box).

--
virtualization experience starting Jan1968, online at home since Mar1970

390 vector instruction set reuse, was 8-bit bytes

Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 390 vector instruction set reuse, was 8-bit bytes
Newsgroups: comp.arch
Date: Tue, 09 Oct 2012 23:33:55 -0400
Robert Wessel <robertwessel2@yahoo.com> writes:
I do have some trouble believing that - these machines had only four FP regs - the reduction in memory traffic from the much larger (and more numerous) vector registers would have been substantial. Also these machines all had unified L1 caches, and vector loads and stores (which supported non-unit stride vectors and gather/scatter) would have greatly reduced instruction traffic through the L1.

Of course the age old problem being that all that requires properly vectorizable code.


re:
https://www.garlic.com/~lynn/2012n.html#28 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#31 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#32 390 vector instruction set reuse, was 8-bit bytes

long-winded email from long ago and far away (repeated from discussion with 3090 processor engineers):

Date: Jun 15 11:27:41 1989
From: wheeler
To: distribution
Subject: 3090 vector

My understanding of 3090 vector was that it was primarily a marketing ploy. With a little additional smarts and 7-stage pipeline ... 3090+ would execute instructions at memory bandwidth speed. The advantage (over vector) is that both numerical instensive (done via vector or scalar) would proceed at memory bandwidth spped AND all other scalar would also operate at memory bandwidth rates also. I believe that Hitachi has such a machine ... i.e. vectorization has little additional performance improvement over scalar since the scalar pipelining was operating memory bus at peak thruput.

The advantage (in the past) that vectorization has had over scalar is that the pipelining algorithms haven't been sophisticated enough. The seven stage pipeline would require less (additional) circuits than vector unit and still operate the hardware at peak efficiency. The problem is trying to design hardware systems that maximize the utilization of all the system components ... as well as total system thruput.

At one time there was work going on to build a dual i-stream version of the 370/195. The problem was that the 195 pipeline would drain when a branch instruction was encountered. The net result was that the hardware very seldom would operate at peak thruput. Adding a second i-stream to the hardware ... but continuing to use the same amount of execution and pipeline hardware (just taging each instruction as to its i-stream) ... effectively doubled the number of available instructions in the pipeline (probability of pipeline totally draining because both i-streams were at branches was low). The 3090+ 7-stage would support looking down both paths of a banch ... in an attempt to keep the pipeline full.

In some sense, vectorization is its own form of pipelining in order to optimize the total utilization of all the system components (memory, memory bus, i/o, cpu, i-unit, caches, e-units, floating point, etc). Sophisticated scalar pipelining is also capable of similar results with the added advantage of being applicable to total mix of instructions being executed.

370 MP

However, both approaches reach thruput bottlenecks at single bus to memory. Parallel execution has the advantage that more of the hardware components are duplicated ... and therefor the peak possible sustained thruput have fewer single points of contention. Although this isn't alwas the case, most of the ibm mainframe MP implementations bog down because of cache and software syncrhonization problems ... which is somewhat an artifact of the software having minimally evolved from an UP design point. Large parts of the operating system are still very much UP design point ... adapted to MP environment by defining critical sections that are bracketed by TS instructions. Basic UP design-point system with little thot to data scoping will rely heavily on single view of all data across all processors (i.e. cache syncrhonization heavily utilized) as well as numerous serialized critical code sections with TS looping.

There was a lot of work done for atomic CS with the design point of (at least in the software) eliminating critical sections and TS bracketing. However over the last 15 years this has had minimal propagation into major portions of the system. Other contributing factors have been in several situations fully parallel system code has been maintenanced (over a period of years) into critical section code with TS bracketing (a major problem over a 15-20 year development period is making sure that all people that might touch a piece of code are as skilled and talented as the people who originally wrote it). Given a mix of programmers, there are some that barely understand concurrent multiprogramming (on a UP) and their solution to concurrency problems is to disable critical sections ... that background translated into MP environment results in bracketing with TS loops.

In any case, a major design opportunity ... starting at the lowest kernel level ... is adequate attention to scoping/partitioning data, control blocks, etc with the objective to minimize critical sections with TS loop bracketing. This would then translates into less hardware degradation because of cache syncrhonization overhead ... as well as less software degradation because of TS looping on bracketing critical code sections.


... snip ... top of post, old email index

as i've mentioned before, CS had been invented by Charlie at science center while doing fine-grain multiprocessor locking work on cp67. the initial attempt to get it included in 370 was rebuffed, the favorite son operating system people in POK saying TS was more than sufficient. The 370 architecture "owners" explained that to get CS included in 370, uses for the instruction would be needed other than the kernel multiprocessor operation. thus was born the uses for multi-threaded (multiprogramming) applications (regardless of whether or not running in multiprocessor environment) ... the examples still appear in the mainframe principles of operation. misc. past posts mentioning multiprocessor and/or compare&swap
https://www.garlic.com/~lynn/subtopic.html#smp

semi-related other old 3090 email

3090 was referred to as trout 1.5 or in the following 1.5 (from 3090 processor engineer)
https://www.garlic.com/~lynn/2003j.html#email831118

3090 sie
https://www.garlic.com/~lynn/2006j.html#email810630

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Tue, 09 Oct 2012 23:59:59 -0400
Dave Garland <dave.garland@wizinfo.com> writes:
Get in on the ground floor. Big market coming for wallets that have a conductive layer to provide shielding. They're already around, but the ones I've seen are rather clunky. My bet is on flexible shielding laminated between layers of "normal" wallet materials, leather or fabric. For handbags, too.

there was class of RFID chips adopted for inventory as enhancement over UPC barcode ... carrying unique EPC product code .... every aisle had receiver that could go down and back and record all the items on the shelves. could also be scanned at check-out.

form of plastic payment cards were developed that basically used the same kind of chip ... but instead of carrying an EPC product code ... it carried the information that is found on debit/credit magnetic stripe.

skimming of magnetic stripe information to make counterfeit cards has been around for decades .... however the characteristic of the class of RFID chips developed for EPC/inventory ... when applied to payment cards ... also greatly increasing the skimming exposure.

misc. past posts mentioning RFID EPC:
https://www.garlic.com/~lynn/2006o.html#16 Gen 2 EPC Protocol Approved as ISO 18000-6C
https://www.garlic.com/~lynn/2006o.html#17 Gen 2 EPC Protocol Approved as ISO 18000-6C
https://www.garlic.com/~lynn/2006o.html#20 Gen 2 EPC Protocol Approved as ISO 18000-6C
https://www.garlic.com/~lynn/2006r.html#17 Greatest Software Ever Written?
https://www.garlic.com/~lynn/2008i.html#61 Could you please name sources of information you trust on RFID and/or other Wireless technologies?
https://www.garlic.com/~lynn/2008l.html#61 Osama bin Laden gets a cosmetic makevover in his British Vanity Passport
https://www.garlic.com/~lynn/2008n.html#48 In your experience which is a superior debit card scheme - PIN based debit or signature debit?
https://www.garlic.com/~lynn/2008p.html#14 Can Smart Cards Reduce Payments Fraud and Identity Theft?
https://www.garlic.com/~lynn/2008p.html#19 Can Smart Cards Reduce Payments Fraud and Identity Theft?
https://www.garlic.com/~lynn/2008p.html#46 Would you say high tech authentication gizmo's are a waste of time/money/effort?
https://www.garlic.com/~lynn/2008p.html#55 Can Smart Cards Reduce Payments Fraud and Identity Theft?
https://www.garlic.com/~lynn/2009.html#72 Double authentification for internet payment
https://www.garlic.com/~lynn/2009b.html#35 The recently revealed excesses of John Thain, the former CEO of Merrill Lynch, while the firm was receiving $25 Billion in TARP funds makes me sick
https://www.garlic.com/~lynn/2009b.html#61 Passport RFIDs cloned wholesale by $250 eBay auction spree
https://www.garlic.com/~lynn/2009c.html#56 Why use RFID in personal documents & cards at all?
https://www.garlic.com/~lynn/2009e.html#21 ATMs At Risk
https://www.garlic.com/~lynn/2009h.html#53 64 Cores -- IBM is showing a prototype already
https://www.garlic.com/~lynn/2009h.html#54 64 Cores -- IBM is showing a prototype already
https://www.garlic.com/~lynn/2009j.html#58 Price Tag for End-to-End Encryption: $4.8 Billion, Mercator Says
https://www.garlic.com/~lynn/2009m.html#2 Does this count as 'computer' folklore?
https://www.garlic.com/~lynn/2009m.html#4 Hacker charges also an indictment on PCI, expert says
https://www.garlic.com/~lynn/2009m.html#10 Does this count as 'computer' folklore?
https://www.garlic.com/~lynn/2009m.html#14 The Art of Creating Strong Passwords
https://www.garlic.com/~lynn/2009m.html#20 Does this count as 'computer' folklore?
https://www.garlic.com/~lynn/2009m.html#24 Does this count as 'computer' folklore?
https://www.garlic.com/~lynn/2009m.html#28 PCI Council Releases Recommendations For Preventing Card-Skimming Attacks
https://www.garlic.com/~lynn/2009n.html#7 Some companies are selling the idea that you can use just a (prox) physical access badge (single factor) for logical access as acceptable
https://www.garlic.com/~lynn/2009o.html#66 Need for speedy cryptography
https://www.garlic.com/~lynn/2009p.html#72 Crypto dongles to secure online transactions
https://www.garlic.com/~lynn/2010c.html#11 PC history, was search engine history, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010e.html#66 z9 / z10 instruction speed(s)
https://www.garlic.com/~lynn/2010f.html#26 Should the USA Implement EMV?
https://www.garlic.com/~lynn/2010g.html#21 Should the USA Implement EMV?
https://www.garlic.com/~lynn/2010o.html#49 The Credit Card Criminals Are Getting Crafty
https://www.garlic.com/~lynn/2011b.html#11 Credit cards with a proximity wifi chip can be as safe as walking around with your credit card number on a poster
https://www.garlic.com/~lynn/2011e.html#24 The first personal computer (PC)
https://www.garlic.com/~lynn/2012b.html#71 Password shortcomings

--
virtualization experience starting Jan1968, online at home since Mar1970

390 vector instruction set reuse, was 8-bit bytes

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 390 vector instruction set reuse, was 8-bit bytes
Newsgroups: comp.arch
Date: Wed, 10 Oct 2012 16:58:04 -0400
Quadibloc <jsavard@ecn.ab.ca> writes:
But as we also know, FS didn't stay dead. Just as IBM recycled its Model 85 technology and made incremental improvements to it rather than doing extra work by starting from scratch, FS went on to become the AS/400, and today's iSystem.

re:
https://www.garlic.com/~lynn/2012n.html#28 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#31 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#32 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#33 390 vector instruction set reuse, was 8-bit bytes

A vastly simplified FS was done as s/38. One of the issues was that it treated all disk space as common allocation pool and pieces could be scatter allocated across all available drives. As a result, all disks had to be backed up as single filesystem image ... and any single disk failure (common failure mode) would require complete filesystem restore (of all disks) ... potentially taking a day.

Part of s/38 was single-level store from FS ... something that was at least some part from tss/360 (and in its simplified s/38 form was common disk allocation pool across all drives that would never scale). I had noticed how misearably slow tss/360 was as an undergraduate when working on virtual machine cp/67. I would do some benchmarks on cp67 and the IBM SE working on tss/360 would also do same benchmarks. Example was fortran source edit, compile and execute ... cp67/cms benchmark with 35 users outperformed and had better trivial response than tss/360 with 4 users (on same exact hardware).

later when I did page-mapped filesystem for CMS, ... I could show three times the thruput of the standard (non-page-mapped) CMS filesystem for same moderate disk I/O workloads. I claimed I had avoided much of the things I felt had been done wrong in tss/360 (and was being carried forward into FS and later s/38) ... and part of the reason that I would periodically ridicule the FS activities. misc. past posts mentioning FS
https://www.garlic.com/~lynn/submain.html#futuresys
misc. past posts mentioning paged-mapped filesystem
https://www.garlic.com/~lynn/submain.html#mmap

as/400 was follow-on for s/34, s/36, & s/38
https://en.wikipedia.org/wiki/IBM_System_i

as mentioned in above ... even with low-end market target for s/38, and being able to get faster processors ... s/38 was "sluggish performance" (i.e. in 360 case, applications that ran on 370/195 ... moved to the fast FS available would have 370/145 thruput) ... some big part of the slowness was capability addressing (from FS) ... removed in as/400 ... from above:

The AS/400 removed capability-based addressing.[3] The AS/400 added source compatibility with the System/36 combining the two primary computers manufactured by the IBM Rochester plant.

... snip ...

One of the analysis of the FS storage addressing was that there could be five-levels of indirection before fetching the value for a simple ADD.

s/34&s/36 was much, much larger market than s/38.
https://en.wikipedia.org/wiki/IBM_System/36

... 20,000 s/38 customers and 300,000 s/34&s/36 ... although it mentions customers with $20,000 s/36 had hard time justifying upgrading to $40,000 as/400.

--
virtualization experience starting Jan1968, online at home since Mar1970

390 vector instruction set reuse, was 8-bit bytes

Refed: **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: 390 vector instruction set reuse, was 8-bit bytes
Newsgroups: comp.arch
Date: Wed, 10 Oct 2012 19:46:17 -0400
Stephen Fuld <SFuld@alumni.cmu.edu.invalid> writes:
I thought the 370/168 was a 370/165 plus virtual memory (paging) and that the /165 was the successor to the 360/65.

re:
https://www.garlic.com/~lynn/2012n.html#28 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#31 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#32 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#33 390 vector instruction set reuse, was 8-bit bytes
https://www.garlic.com/~lynn/2012n.html#35 390 vector instruction set reuse, was 8-bit bytes

Stephen Fuld <SFuld@alumni.cmu.edu.invalid> writes:
I thought the 370/168 was a 370/165 plus virtual memory (paging) and that the /165 was the successor to the 360/65.

370/155 & 370/165 were shipped w/o virtual memory. when virtual memory was announced there was hardware upgrade to add virtual memory (370/155-II and 370/165-II) ... which was extra cost for customers with machines already in the field.

there was also problems with the development of the virtual memory hardware for 165, schedule was slipping and eventually decision was made to drop several features from 370 virtual memory in order to buy six month schedule. This required that all the other processors that already had the full 370 virtual memory implementation to also drop the full features ... and any already developed "virtual memory" software using the dropped features had to be redone.

this had a big hit to vm370/cms since cms had been structured to have shared segments (across all address spaces) that used "segment protect" ... one of the features dropped as part of the 165 schedule rework for virtual memory retrofit.

mvt had protected its kernel storage with 360 "storage protect" keys. in the transission from mvt to virtual memory OS/VS2 SVS ... it effectively laid out MVT in a single 16mbyte virtual address space ... continuing to rely on 360 "storage protect" keys. In the transition from SVS to MVS ... where each application got is own 16mbyte address space ... segment sharing was used to map the MVS kernel into 8mbytes of every application address space ... but the kernel continued to use 360 "storage protect" keys (didn't have the vm370/cms problem where virtual address space could do nearly anything it wanted to within its own address space ... and therefor "shared segments" needed the original 370 virtual memory segment protection feature ... which was among the features dropped in order for 165 to meet schedule)

155 & 165 both had processor caches with 2mic main memory. Big change from 165 to 168 was from 2mic memory to 320nsec main memory (as well as virtual memory came standard). The machines were horizontal microcode and the 3033 engineers that I did some work with ... mentioned in going from 165 to 168 they reduced the avg. machine cycle time per 370 instruction from avg of 2.1 machine cycles to 1.6 machine cycles (by reworking the horizontal microcode implementing 370). There was then a doubling of cache size in 168-1 to 168-3 from 16kbytes to 32kbytes ... the 168-3 was basis used for 3033.

There were a couple of us that wanted to do 16-way 370 multiprocessor and had gotton some of the processor engineers working on 3033 to work on the 16-way effort in their spare time (a lot more interesting than what they were doing on 3033). Everything was going along fine, many in the company thought it was great ... until somebody happen to mention to the head of POK (high-end 370 mainframe land), that it might be decades before the POK favorite-son operating system (MVS) would have 16-way SMP support. At that time, the head of POK invited some of us never to visit the POK location again ... and told the 3033 engineers to solely focus on 3033 and stop being distracted.

the processor engineers working on 3033 were concurrent with group working on 3081 ... some more here:
http://www.jfsowa.com/computer/memo125.htm

when the 3033 was out the door, the 3033 engineers started work on 3090.

165 reference ... 2mic main memory, 8k cache standard, 16k optional
http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_PP3165.html

168 reference ... 8k cache standard, 16k optional
http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_PP3168.html

3033 reference
http://www-03.ibm.com/ibm/history/exhibits/3033/3033_room.html

product announcements
http://ed-thelen.org/comp-hist/IBM-ProdAnn/index.html

processor cycle time, memory cycle time, etc, table here (for both 360 & 370):
http://www.beagle-ears.com/lars/engineer/comphist/model360.htm

there were 3031, 3032, and 3033 ... plus an external channel director box (six i/o channels). the 158-3 had "integrated" channels ... the same microcode engine used for 370 execution was also shared doing channel i/o functions. for the channel director box, they took a 158-3 w/o the 370 microcode, just the integrated channel microcode. A 3031 was a 158-3 engine with new covers and the integrated channel microcode removed with a 2nd 158-3 engine (as channel director) with just the channel microcode. A 3032 was 168-3 with new covers and reconfigured to use (158-3 integrated channel) channel director. The 3033 was 168-3 logic remapped to faster chips ... and one or more (158-3 integrated channel) channel directors.

--
virtualization experience starting Jan1968, online at home since Mar1970

PDP-10 and Vax, was System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: PDP-10 and Vax, was System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Thu, 11 Oct 2012 10:06:30 -0400
jmfbahciv <See.above@aol.com> writes:
It had to do with the philosophy of the OS. A real timesharing system would make the perceived response time for the users a higher priority (term not to be confused withe hard/software priority implementations) than the background bookkeeping processing. A task-based OS philosophy places finishing the current task above everything.

one of the things that early cp67 did was make any terminal i/o (regardless of read or write) indication of human response event and "high priority". this led to all sorts of anomolous activity ... including users discovering that if they performed gratuitous terminal writes during long running compute bound operation, it got better throughput (CMS "blip" had built-in terminal i/o that would "wiggle" the 2741 typeball every two-seconds of cpu use, part of gaming the system was to cut the interval for "blip").

when I rewrote resource management ... terminal i/o allowed smaller bursts at more frequent intervals ... but aggregate use didn't increase. using much less than "fair share" ... also allowed a few more frequent use bursts. the combination tended to give extremely good interactive response for "trivial" activity ... but response would drop off (in system with lots of contention) as resource use increased.

this recent post discusses sub-second and .2 second response for vm370
https://www.garlic.com/~lynn/2012m.html#37 Why File transfer through TSO IND$FILE is slower than TCP/IP

channel attached 3272/3277 controller/terminal had .086 "hardware terminal" response ... in order to provide .2 second "human" respose ... required system response to be .114 or less.

this old post with old hardware measures shows that the newer 3274/3278 terminals could never meet .2sec ... because terminal hardware tended to be 1/3rd sec to 1/2sec.
https://www.garlic.com/~lynn/2001m.html#19 3270 protocol

one of the main east coast internal datacenters had published report that included claims that they had the best vm370 online service in the company with quarter second avg trivial system response. I complained pointing out that I ran several (similar, hardware configuration & workload) systems that had .11sec 90th percentile trivial system response (achieving .2 second human response with channel attached 3272/3277 terminals). response I got back was that particular datacenter could claim anything they wanted to in their reports.

from ibm jargon (by author of rexx):

bad response - n. A delay in the response time to a trivial request of a computer that is longer than two tenths of one second. In the 1970s, IBM 3277 display terminals attached to quite small System/360 machines could service up to 19 interruptions every second from a user I measured it myself. Today, this kind of response time is considered impossible or unachievable, even though work by Doherty, Thadhani, and others has shown that human productivity and satisfaction are almost linearly inversely proportional to computer response time. It is hoped (but not expected) that the definition of Bad Response will drop below one tenth of a second by 1990.

... snip ...

other recent posts mentioning sub-second, .2second &/or "bad response"
https://www.garlic.com/~lynn/2012.html#12 Who originated the phrase "user-friendly"?
https://www.garlic.com/~lynn/2012.html#13 From Who originated the phrase "user-friendly"?
https://www.garlic.com/~lynn/2012b.html#2 The PC industry is heading for collapse
https://www.garlic.com/~lynn/2012d.html#19 Writing article on telework/telecommuting
https://www.garlic.com/~lynn/2012d.html#30 Inventor of e-mail honored by Smithsonian
https://www.garlic.com/~lynn/2012d.html#38 Invention of Email
https://www.garlic.com/~lynn/2012e.html#87 Just for a laugh... How to spot an old IBMer
https://www.garlic.com/~lynn/2012e.html#98 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012i.html#74 HELP WITH PCOM - PASTE OPTION NOT WORKING CORRECTLY
https://www.garlic.com/~lynn/2012k.html#3 printer history Languages influenced by PL/1
https://www.garlic.com/~lynn/2012m.html#15 cp67, vm370, etc

--
virtualization experience starting Jan1968, online at home since Mar1970

Jedi Knights

Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 11 Oct 2012
Subject: Jedi Knights
Blog: Boyd Strategy
re:
http://lnkd.in/xZigE
and
https://www.garlic.com/~lynn/2012n.html#10 Jedi Knights
https://www.garlic.com/~lynn/2012n.html#23 Jedi Knights
https://www.garlic.com/~lynn/2012n.html#29 Jedi Knights

from '89 ... although doesn't mention 35-ton load limit ... but otherwise trying to match Abrams to USMC mission profiles ... somewhat noting Abrams was designed for slug-fest with Soviet tanks.
http://www.globalsecurity.org/military/library/report/1989/GKJ.htm

and from
http://www.army-technology.com/projects/abrams/

The first M1 Abrams battle tanks were delivered to the US Army in 1980. In all 3,273 M1 tanks were produced for the US Army, 4,796 M1A1 tanks were built for the US Army, 221 for the US Marines and 880 co-produced with Egypt.

and from
http://www.nationaldefensemagazine.org/archive/2002/September/Pages/Army_Marine6671.aspx
The Corps has yet to decide how many of the new vehicles it will need to replace the existing fleet of 403 tanks. MEFFV is not designed as a one-for-one replacement program, Beal said.

As to whether the MEFFV will be tracked or wheeled, Beal said that still is up in the air. "We would like, for the 30-ton assault variant, to have 8x8 tracks-over-wheels. That gives us the flexibility to operate in a tracked or wheeled mode," he said. "For the lighter ariants, we'll use wheels."


... snip ...

... somewhat implies trying to meet 35-ton load limit mission profiles.

now back to (army) tank warfare vis-a-vis marine mission profile ... sort of like sides in above
https://web.archive.org/web/20160326155408/https://www.mca-marines.org/gazette/attritionist-letters-archives

I know the person that "originated" COTS for the federal gov. ... basically take something that possibly doesn't do what you exactly want ... but makes up for it by being able to have hundreds of times the number. In more recent years he has partially recanted in that there are lots of instances of getting something that doesn't do any one thing particular well (may actually do everything badly) *AND* they cost so much, that can only have a relative few (the worst case scenario, lots of hiding behind promise of COTS, but pure facade).

Boyd would talk about dog fights, one such involved us pilot up against five migs ... described it almost like chess game ... at the same moment, all six pilots realize that the us pilot was about to shoot down the five migs. Attention to observation was significant advantage ... there were significant degrees of freedom in dog fight ...better observation was significant advantage ... mentions f86 canopy visibility
http://zenpundit.com/?p=3621

There is quite a bit to say about surprise in combat ... shooting your opponent out of the air before they even see you ... but "40-second" Boyd is about observing and maneuvering ... described in the previously mentioned Boyd tribute by Chuck
http://radio-weblogs.com/0107127/stories/2002/12/23/genghisJohnChuckSpinneysBioOfJohnBoyd.html

There is lots of adaption from 3d maneuvering to 2d and the tempo is quite a bit different ... but there is lots of discussion that many in usmc ... while they may or may not have any understanding ... they just don't want to change
https://web.archive.org/web/20160326155408/https://www.mca-marines.org/gazette/attritionist-letters-archives

I recently finished slightly different view of how things unfolded for Kuwait ... Long Strange Journey: An Intelligence Memoir (Patrick Eddington)
https://www.amazon.com/Long-Strange-Journey-Intelligence-ebook/dp/B004NNV5H2

with no preparations for Kuwait were made until they started showing Iraq staging forces in Kuwait along the Saudi border .. loc2030-36
An unnamed Bush administration official actually went so far as to claim that the lack of Iraqi propaganda attacks on Saudi Arabia argued against an Iraqi invasion of the kingdom. Focusing solely on Iraqi statements was precisely the mistake made prior to the invasion of Kuwait. Saddam's public declaration that he desired a "dialogue" with Kuwait had been nothing more than a smoke screen behind which he'd hid his military preparations--just as Hitler had over 50 years before when he annexed Austria. I had a sick feeling that history was about to repeat itself.

... snip ...

--
virtualization experience starting Jan1968, online at home since Mar1970

PDP-10 and Vax, was System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: PDP-10 and Vax, was System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Thu, 11 Oct 2012 11:06:49 -0400
"Charles Richmond" <numerist@aquaporin4.com> writes:
"With computers, anything less than instantaneous is too slow..."

re:
https://www.garlic.com/~lynn/2012n.html#37 PDP-10 and Vax, was System/360--50 years--the future?

there was human factors study regarding human perception threshold ... when they could differentiate between instantaneous and something slightly longer. there was difference between individuals in the study that varied between about .09 seconds and .2seconds. there was some conjecture that the variability was accounted for difference between different humans in the time signals propagate in the brain (individuals in the study were all employees of high-end research facility).

another human factors study related to productivity was that human attention started to wander when response wasn't within their threshold ... and then when the response came ... the time it took to refocus was about the same duration as attention spent wandering ... i.e. "loss of productivity" was twice the elapsed time delay between the expected response and the actual response.

lots of browser web stuff now tends to be much longer than threshold. i've mentioned periodically that help drive me to using browser background tabs (going on nearly decade now) ... where I queue up a couple hundred web pages in browser background tabs. Time to switch browser tabs is now purely local issue.

misc. past posts mentioning tab use
https://www.garlic.com/~lynn/2003i.html#44 graceful recovery when runs out of paging?
https://www.garlic.com/~lynn/2004e.html#54 Is there a way to configure your web browser to use multiple
https://www.garlic.com/~lynn/2004h.html#25 Why are programs so large?
https://www.garlic.com/~lynn/2005e.html#48 Mozilla v Firefox
https://www.garlic.com/~lynn/2005e.html#49 Mozilla v Firefox
https://www.garlic.com/~lynn/2005e.html#50 Mozilla v Firefox
https://www.garlic.com/~lynn/2005e.html#55 Mozilla v Firefox
https://www.garlic.com/~lynn/2005e.html#60 Mozilla v Firefox
https://www.garlic.com/~lynn/2005e.html#66 Mozilla v Firefox
https://www.garlic.com/~lynn/2005n.html#8 big endian vs. little endian, why?
https://www.garlic.com/~lynn/2005n.html#15 1.8b2 / 1.7.11 tab performance
https://www.garlic.com/~lynn/2005n.html#41 Moz 1.8 performance dramatically improved
https://www.garlic.com/~lynn/2005n.html#42 Moz 1.8 performance dramatically improved
https://www.garlic.com/~lynn/2005o.html#13 RFC 2616 change proposal to increase speed
https://www.garlic.com/~lynn/2006k.html#39 PDP-1
https://www.garlic.com/~lynn/2006q.html#51 Intel abandons USEnet news
https://www.garlic.com/~lynn/2007l.html#30 tab browsing
https://www.garlic.com/~lynn/2007m.html#8 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007t.html#74 What do YOU call the # sign?
https://www.garlic.com/~lynn/2008b.html#32 Tap and faucet and spellcheckers
https://www.garlic.com/~lynn/2008b.html#35 Tap and faucet and spellcheckers
https://www.garlic.com/~lynn/2008d.html#24 Javascript disabled in Firefox
https://www.garlic.com/~lynn/2008h.html#10 What would be a future of technical blogs ? I am wondering what kind of services readers except to get from a technical blog in next 10 years
https://www.garlic.com/~lynn/2008h.html#71 Mainframe programming vs the Web
https://www.garlic.com/~lynn/2008h.html#85 Mainframe programming vs the Web
https://www.garlic.com/~lynn/2008i.html#85 Which of the latest browsers do you prefer and why?
https://www.garlic.com/~lynn/2008k.html#37 squirrels
https://www.garlic.com/~lynn/2008p.html#29 How were you using the internet 10 years ago and how does that differ from how you use it today?
https://www.garlic.com/~lynn/2009g.html#54 Windowed Interfaces 1981-2009
https://www.garlic.com/~lynn/2009m.html#66 What happened to computer architecture (and comp.arch?)
https://www.garlic.com/~lynn/2009q.html#72 Now is time for banks to replace core system according to Accenture
https://www.garlic.com/~lynn/2010b.html#48 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010d.html#22 OT: PC clock failure--CMOS battery?
https://www.garlic.com/~lynn/2010k.html#39 Unix systems and Serialization mechanism
https://www.garlic.com/~lynn/2011c.html#73 IBM and the Computer Revolution
https://www.garlic.com/~lynn/2011e.html#73 History--Early Bell System teletypes
https://www.garlic.com/~lynn/2011j.html#29 DG Fountainhead vs IBM Future Systems
https://www.garlic.com/~lynn/2011l.html#61 Agents
https://www.garlic.com/~lynn/2012.html#15 Who originated the phrase "user-friendly"?
https://www.garlic.com/~lynn/2012d.html#19 Writing article on telework/telecommuting
https://www.garlic.com/~lynn/2012e.html#90 Just for a laugh ... How to spot an old IBMer

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Fri, 12 Oct 2012 11:48:50 -0400
"Charles Richmond" <numerist@aquaporin4.com> writes:
Which was a resounding failure. ISTM that the iAPX432 was too ambitious for the "state of the art" chip development at the time.

432 architecture book intro references burroughs and system/38 ... part of intro in this old post:
https://www.garlic.com/~lynn/2000f.html#48 Famous Machines and Software that didn't

'81 acm sigops at asilomar there was presentation by some people from 432 group. besides capability there was lots of complex operating system dropped into silicon. lots of bugs and every fix required new silicon chip replacement.

1975 i had done design for 370/125 (never shipped) ... wasn't the 432 capability ... but lots of complex operating system stuff that 432 dropped into (silicon) machine ... i defined as part of "machine" ... 5-way smp, raised level i/o abstract ... but it was all "microcode" ... fixes were floppy disk.

past posts mentioning 432
https://www.garlic.com/~lynn/2000d.html#57 iAPX-432 (was: 36 to 32 bit transition
https://www.garlic.com/~lynn/2000d.html#62 iAPX-432 (was: 36 to 32 bit transition
https://www.garlic.com/~lynn/2001g.html#36 What was object oriented in iAPX432?
https://www.garlic.com/~lynn/2002d.html#27 iAPX432 today?
https://www.garlic.com/~lynn/2002d.html#46 IBM Mainframe at home
https://www.garlic.com/~lynn/2002l.html#19 Computer Architectures
https://www.garlic.com/~lynn/2002o.html#5 Anyone here ever use the iAPX432 ?
https://www.garlic.com/~lynn/2003e.html#54 Reviving Multics
https://www.garlic.com/~lynn/2003m.html#23 Intel iAPX 432
https://www.garlic.com/~lynn/2003m.html#24 Intel iAPX 432
https://www.garlic.com/~lynn/2003m.html#47 Intel 860 and 960, was iAPX 432
https://www.garlic.com/~lynn/2004q.html#64 Will multicore CPUs have identical cores?
https://www.garlic.com/~lynn/2005q.html#31 Intel strikes back with a parallel x86 design
https://www.garlic.com/~lynn/2006n.html#42 Why is zSeries so CPU poor?
https://www.garlic.com/~lynn/2006p.html#15 "25th Anniversary of the Personal Computer"
https://www.garlic.com/~lynn/2006t.html#7 32 or even 64 registers for x86-64?
https://www.garlic.com/~lynn/2009o.html#13 Microprocessors with Definable MIcrocode
https://www.garlic.com/~lynn/2009o.html#18 Microprocessors with Definable MIcrocode
https://www.garlic.com/~lynn/2010h.html#40 Faster image rotation
https://www.garlic.com/~lynn/2011l.html#2 68000 assembly language programming

--
virtualization experience starting Jan1968, online at home since Mar1970

System/360--50 years--the future?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: System/360--50 years--the future?
Newsgroups: alt.folklore.computers
Date: Fri, 12 Oct 2012 12:57:20 -0400
John Levine <johnl@iecc.com> writes:
I gather that PARS, the Sabre-like system built for Eastern Airlines was also substantially rewritten and is the ancestor of TPF.

On the other hand, they all still have the same basic design as the original SABRE, tightly constrained chunks of code executed in response to interrupts, saving their own context. There were some pretty smart dudes back in 1960.


acp (airline control program) started being used by some financial institutions in the 70s for large financial transaction networks, motivating name change to TPF.

there was major problem ... while acp had loosely-coupled multiprocessor but not tightly-coupled multiprocessor (shared memory). 3081 was going to be tightly-coupled multiprocessor only ... no single processor option ... there was potential that all TPF customers were going to migrate to clone processor vendors ... that were still offering single processor machines. until corporation came out with belated 3083 (3081 with one processor removed) ... there were some ugly things done to vm370 trying to optimize TPF running in vm370 virtual machine ... vm370 operating 3081 multiprocessor and providing single processor (TPF) virtual machine (changes included in standard vm370 product which degraded thruput for nearly every other vm370 customer running multiprocessor.

my wife did short stint as chief architect for amadeus (have original document in box someplace) ... eu res system based on eastern's system one. she supported going with x.25 ... which got her quickly replaced by the sna forces ... didn't do them much good because amadeus went with x.25 anyway.

as an aside ... part of nail in future system effort ... was the system/one example that ran on 370/195 ... an future system machine built from the fastest available hardware (370/195) running system/one, would have thruput of 370/145 (10 to 30 times degradation in thruput).

misc. past posts mentioning amadeus
https://www.garlic.com/~lynn/2001g.html#49 Did AT&T offer Unix to Digital Equipment in the 70s?
https://www.garlic.com/~lynn/2001g.html#50 Did AT&T offer Unix to Digital Equipment in the 70s?
https://www.garlic.com/~lynn/2001h.html#76 Other oddball IBM System 360's ?
https://www.garlic.com/~lynn/2003d.html#67 unix
https://www.garlic.com/~lynn/2003n.html#47 What makes a mainframe a mainframe?
https://www.garlic.com/~lynn/2004b.html#6 Mainframe not a good architecture for interactive workloads
https://www.garlic.com/~lynn/2004b.html#7 Mainframe not a good architecture for interactive workloads
https://www.garlic.com/~lynn/2004m.html#27 Shipwrecks
https://www.garlic.com/~lynn/2004o.html#23 Demo: Things in Hierarchies (w/o RM/SQL)
https://www.garlic.com/~lynn/2004o.html#29 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2005f.html#22 System/360; Hardwired vs. Microcoded
https://www.garlic.com/~lynn/2005p.html#8 EBCDIC to 6-bit and back
https://www.garlic.com/~lynn/2006o.html#4 How Many 360/195s and 370/195s were shipped?
https://www.garlic.com/~lynn/2006r.html#9 Was FORTRAN buggy?
https://www.garlic.com/~lynn/2006y.html#14 Why so little parallelism?
https://www.garlic.com/~lynn/2007d.html#19 Pennsylvania Railroad ticket fax service
https://www.garlic.com/~lynn/2007e.html#52 US Air computers delay psgrs
https://www.garlic.com/~lynn/2007h.html#12 The Perfect Computer - 36 bits?
https://www.garlic.com/~lynn/2007k.html#72 The top 10 dead (or dying) computer skills
https://www.garlic.com/~lynn/2007o.html#59 ACP/TPF
https://www.garlic.com/~lynn/2007p.html#45 64 gig memory
https://www.garlic.com/~lynn/2008c.html#53 Migration from Mainframe to othre platforms - the othe bell?
https://www.garlic.com/~lynn/2008i.html#19 American Airlines
https://www.garlic.com/~lynn/2008i.html#34 American Airlines
https://www.garlic.com/~lynn/2008p.html#41 Automation is still not accepted to streamline the business processes... why organizations are not accepting newer technologies?
https://www.garlic.com/~lynn/2009j.html#33 IBM touts encryption innovation
https://www.garlic.com/~lynn/2009l.html#55 IBM halves mainframe Linux engine prices
https://www.garlic.com/~lynn/2009r.html#59 "Portable" data centers
https://www.garlic.com/~lynn/2010.html#23 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010g.html#29 someone smarter than Dave Cutler
https://www.garlic.com/~lynn/2010n.html#16 Sabre Talk Information?
https://www.garlic.com/~lynn/2011.html#17 Looking for a real Fortran-66 compatible PC compiler (CP/M or DOS or Windows, doesn't matter)
https://www.garlic.com/~lynn/2011.html#41 Looking for a real Fortran-66 compatible PC compiler (CP/M or DOSor Windows
https://www.garlic.com/~lynn/2011d.html#14 Sabre; The First Online Reservation System
https://www.garlic.com/~lynn/2011d.html#43 Sabre; The First Online Reservation System
https://www.garlic.com/~lynn/2011d.html#74 Multiple Virtual Memory
https://www.garlic.com/~lynn/2011i.html#77 program coding pads
https://www.garlic.com/~lynn/2012c.html#8 The PC industry is heading for collapse
https://www.garlic.com/~lynn/2012c.html#9 The PC industry is heading for collapse
https://www.garlic.com/~lynn/2012h.html#52 How will mainframers retiring be different from Y2K?
https://www.garlic.com/~lynn/2012j.html#5 Interesting News Article

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Sun, 14 Oct 2012 09:36:33 -0400
Peter Flass <Peter_Flass@Yahoo.com> writes:
It's mostly the same OS, in the same way that Version 2 is mostly the same as Version 1. The big change was from real-memory OS/360 to virtual memory OS/VS. Most of the rest has been incremental changes.

re:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other

OS/VS2 SVS (single virtual storage) was very close to running MVT in 16mbyte virtual machine address space. MVT had to build its own virtual memory tables and be able to do page faults and setup page i/o reads&writes. However, possibly the biggest change was having EXCP ... perform SIO on the passed application channel program. Channel I/O requires real addresses in the CCWs ... and now the passed application channel program would have virtual addresses in the CCWs.

The initial implementation of "AOS" (what was to become OS/VS2 SVS) was being built and run on 360/67 (before 370s were available) ... with a copy of CP67's CCWTRANS crafted into EXCP. CCWTRANS was scan the passed channel program (in CP67 case, from the virtual machine virtual address space), duplicating each CCW, fixing the virtual pages involved in data transfers in real storage (so they wouldn't be selected for page replacement before the i/o had completed) and replacing the virtual addresses with real addresses (along with UNTRANS after the i/o completed, unfix the virtual pages involved in the i/o transfers and release the storage for the duplicate channel program). Note in CP67 case supporting "MINIDISKS" ... it also required duplicating the disk arm seek location and adjusting the value to reflect the location on real disks.

Transition from OS/VS2 SVS to MVS required a lot more work. It required building a unique virtual address space for every application ... and laying out the MVT kernel as an 8mbyte image that appeared in every application virtual address space. There was also an issue with the OS/360 extensive pointer-passing API paradigm. MVT "subsystems" that were outside the kernel now how their own virtual address space ... but were still required to access (fetch/store data) at the passed API location in the application virtual address space.

old post discussing historical justification for os/360 moving to virtual memory (mentioning future system, mft II/RASP, hasp, etc):
https://www.garlic.com/~lynn/2011d.html#73 Multiple Virtual Memory

misc. past posts mentioning future system
https://www.garlic.com/~lynn/submain.html#futuresys

misc. past posts mentioning hasp
https://www.garlic.com/~lynn/submain.html#hasp

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Sun, 14 Oct 2012 09:50:46 -0400
Peter Flass <Peter_Flass@Yahoo.com> writes:
Is there any solid-state disk on a real (non-PC) machine yet?

misc. reference to oracle/sun
http://www.oracle.com/us/products/servers-storage/storage/flash-storage/ssd/overview/index.html
http://www.storage-switzerland.com/Articles/Entries/2011/9/8_The_Storage_Challenges_To_Oracle_Exadata.html
http://www.storagesearch.com/oracle.html

circa 1980 there was 1655 internally ... which were emulated fixed-head 2305 disks ... for paging (volitile storage) ... they were actually from intel but ibm didn't want to advertise in was using non-IBM devices.

misc. past posts mentioning solid-state disks
https://www.garlic.com/~lynn/2001c.html#17 database (or b-tree) page sizes
https://www.garlic.com/~lynn/2001l.html#53 mainframe question
https://www.garlic.com/~lynn/2002.html#31 index searching
https://www.garlic.com/~lynn/2002d.html#55 Storage Virtualization
https://www.garlic.com/~lynn/2002i.html#17 AS/400 and MVS - clarification please
https://www.garlic.com/~lynn/2002l.html#40 Do any architectures use instruction count instead of timer
https://www.garlic.com/~lynn/2002n.html#4 Tweaking old computers?
https://www.garlic.com/~lynn/2003b.html#15 Disk drives as commodities. Was Re: Yamhill
https://www.garlic.com/~lynn/2003b.html#17 Disk drives as commodities. Was Re: Yamhill
https://www.garlic.com/~lynn/2003c.html#55 HASP assembly: What the heck is an MVT ABEND 422?
https://www.garlic.com/~lynn/2003m.html#39 S/360 undocumented instructions?
https://www.garlic.com/~lynn/2004d.html#73 DASD Architecture of the future
https://www.garlic.com/~lynn/2004e.html#3 Expanded Storage
https://www.garlic.com/~lynn/2005d.html#26 Latest news about mainframe
https://www.garlic.com/~lynn/2005e.html#5 He Who Thought He Knew Something About DASD
https://www.garlic.com/~lynn/2005r.html#51 winscape?
https://www.garlic.com/~lynn/2006.html#38 Is VIO mandatory?
https://www.garlic.com/~lynn/2006c.html#1 Multiple address spaces
https://www.garlic.com/~lynn/2006e.html#46 using 3390 mod-9s
https://www.garlic.com/~lynn/2006k.html#57 virtual memory
https://www.garlic.com/~lynn/2006r.html#36 REAL memory column in SDSF
https://www.garlic.com/~lynn/2006s.html#30 Why magnetic drums was/are worse than disks ?
https://www.garlic.com/~lynn/2007e.html#2 Securing financial transactions a high priority for 2007
https://www.garlic.com/~lynn/2007e.html#59 FBA rant
https://www.garlic.com/~lynn/2007o.html#26 Tom's Hdw review of SSDs
https://www.garlic.com/~lynn/2007s.html#9 Poster of computer hardware events?
https://www.garlic.com/~lynn/2007u.html#4 Remembering the CDC 6600
https://www.garlic.com/~lynn/2008b.html#15 Flash memory arrays
https://www.garlic.com/~lynn/2009m.html#54 August 7, 1944: today is the 65th Anniversary of the Birth of the Computer
https://www.garlic.com/~lynn/2010g.html#11 Mainframe Executive article on the death of tape
https://www.garlic.com/~lynn/2010g.html#22 Mainframe Executive article on the death of tape
https://www.garlic.com/~lynn/2010g.html#55 Mainframe Executive article on the death of tape
https://www.garlic.com/~lynn/2010g.html#82 [OT] What is the protocal for GMT offset in SMTP (e-mail) header time-stamp?
https://www.garlic.com/~lynn/2010h.html#78 Software that breaks computer hardware( was:IBM 029 service manual )
https://www.garlic.com/~lynn/2011e.html#75 I'd forgotten what a 2305 looked like
https://www.garlic.com/~lynn/2011j.html#9 program coding pads

--
virtualization experience starting Jan1968, online at home since Mar1970

Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 14 Oct 2012
Subject: Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Blog: Enterprise Systems
re:
http://lnkd.in/NBbbzr

There are large number of major failed "re-engineering" efforts ... some being retargeted to non-mainframe ... but also some not re-hosted ... just remaining on mainframe platform. Lots of re-engineering failures ... aren't because of targeted to a different platform ... but because of the nature of re-engineering ... independent of platform

There may be mis-impression about current generation of x86 processors. For decades RISC chips have had significant performance advantage over x86 ... i.e. for decades RISC chips have had out-of-ordered execution, branch prediction, speculative execution, etc. The past few x86 generations have actually been risc cores with hardware layer that translates x86 instructions into risc micro-ops (starting to significantly eliminate the throughput difference between x86 and risc).

One of the latest generation of x86 is e5-2600 blade that is two-chip 8-core/chip for 16-core (or processor) multiprocessor with a rating of 527BIPS or 33BIPS/processor and IBM base price of $1815/blade or $3.44/BIPS)

By comparison maxed out configurations for z10, z196, and ec12 are 30BIPS (64 processors), 50BIPS (80 processors), 75BIPS (101 processors) or 469MIPS/processor, 625MIPS/processor, and 743MIPS/processor. Half of the z10->z196 per processor improvement is adopting risc "out-of-order" execution ... and much of the z196->ec12 per processor is further improvements in the use of risc "out-of-order" execution. However, a single e5-2600 blade at 527BIPS (with 16 processors) is still the processing equivalent of seven fully configured ec12 machines (i.e. 527BIPS/75BIPS).

IBM has list price of $28M for fully configured z196 (at 50BIPS) and elsewhere in this group there is statement that IBM makes 4% of its revenue from selling mainframes but 25% when services, software, and storage are included. That results in IBM earning total of 6.25 dollars for every dollar in mainframe sales (i.e. 25/4). That translates into $175M cost for a $28M max. configured z196 or $3.5M/BIPS ($175M/50BIPS). That compares to ibm's base list price of $1815 e5-2600 ... which at 527BIPS works out to $3.44/BIPS (or one millionth the price/BIPS of z196).

this is recent discussion in ibm-main mailing list of FICON i/o capacity.
https://www.garlic.com/~lynn/2012m.html#4

FICON was layered ontop (in mid-90s) of the fibre-channel standard ... that significantly cut the underlying fibre-channel throughput (originally from 88-92 time-frame). The above references IBM documents with z196 benchmark at peak 2m I/O ops/sec (IOPS) with 104 FICON channels, 14 storage subsystems and 14 system assist processors. It mentions that 14 SAPs are capable of 2.2M SSCH/sec at 100% processor busy, however it recommends normal operation that SAPs be kept at 70% or less processor busy (1.5M SSCH/sec)

By comparison, it lists a recent emulex announcement of single fiber-channel for e5-2600 capable of over one million IOPS. A single $1815 e5-2600 blade has seven times the processor throughput of a max ec12 and nearly the same I/O thruput of a max. configured z196 that uses 104 FICON channels ... but only using a single fibre-channel standard (i.e. the FICON layer enormously restricts the thruput of the underlying fibre-channel standard)

Other IBM references show that the recent zHPF and TCW enhancements to FICON ... partially restores the throughput of the underlying fibre-channel standard to FICON operations. Part of the original fibre-channel standard ... going back to 1988 ... was downloading complete I/O program in one transfer. The initial FICON layer for fibre-channel dropped back to only downloading a single CCW at a time. The recent zHPF enhancements for FICON introduces TCW that will package multiple CCWs for a single transfer.

A few racks of e5-2600 blades has the potential of having more I/O capacity and processing throughput than the total aggregate of all mainframes in the world today.

There is estimate that there are total of 10,000 mainframes in the world today. Even if every one of those turned out to be a max configured 50BIP z196 ... that only works out to be an aggregate of 500TIPS or less than 1000 e5-2600 blades. More realistic most of those systems are older ... so more like aggregate of 5TIPS or 100 e5-2600

other recent posts mentioning z10, z196, ec12, e5-2600, etc:
https://www.garlic.com/~lynn/2012.html#20 21st Century Migrates Mainframe with Clerity
https://www.garlic.com/~lynn/2012.html#23 21st Century Migrates Mainframe with Clerity
https://www.garlic.com/~lynn/2012.html#24 21st Century Migrates Mainframe with Clerity
https://www.garlic.com/~lynn/2012.html#56 IBM researchers make 12-atom magnetic memory bit
https://www.garlic.com/~lynn/2012.html#59 IBM's z196 Article at RWT
https://www.garlic.com/~lynn/2012.html#78 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012.html#80 Article on IBM's z196 Mainframe Architecture
https://www.garlic.com/~lynn/2012.html#90 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012.html#91 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012b.html#28 New IBM mainframe instructions
https://www.garlic.com/~lynn/2012b.html#30 New IBM mainframe instructions
https://www.garlic.com/~lynn/2012b.html#41 Are rotating register files still a bad idea?
https://www.garlic.com/~lynn/2012b.html#57 Has anyone successfully migrated off mainframes?
https://www.garlic.com/~lynn/2012d.html#2 NASA unplugs their last mainframe
https://www.garlic.com/~lynn/2012d.html#35 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012d.html#41 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012d.html#50 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012d.html#64 Layer 8: NASA unplugs last mainframe
https://www.garlic.com/~lynn/2012e.html#3 NASA unplugs their last mainframe
https://www.garlic.com/~lynn/2012e.html#4 Memory versus processor speed
https://www.garlic.com/~lynn/2012e.html#94 Can Mainframes Be Part Of Cloud Computing?
https://www.garlic.com/~lynn/2012e.html#99 Can Mainframes Be Part Of Cloud Computing?
https://www.garlic.com/~lynn/2012e.html#105 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012f.html#0 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012f.html#4 Can Mainframes Be Part Of Cloud Computing?
https://www.garlic.com/~lynn/2012f.html#7 Burroughs B5000, B5500, B6500 videos
https://www.garlic.com/~lynn/2012g.html#36 Should IBM allow the use of Hercules as z system emulator?
https://www.garlic.com/~lynn/2012g.html#38 Should IBM allow the use of Hercules as z system emulator?
https://www.garlic.com/~lynn/2012h.html#4 Think You Know The Mainframe?
https://www.garlic.com/~lynn/2012h.html#20 Mainframes Warming Up to the Cloud
https://www.garlic.com/~lynn/2012h.html#35 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#52 How will mainframers retiring be different from Y2K?
https://www.garlic.com/~lynn/2012h.html#62 What are your experiences with Amdahl Computers and Plug-Compatibles?
https://www.garlic.com/~lynn/2012h.html#70 How many cost a cpu second?
https://www.garlic.com/~lynn/2012i.html#11 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#15 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#16 Think You Know The Mainframe?
https://www.garlic.com/~lynn/2012i.html#84 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#88 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#89 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#95 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#1 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#9 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#34 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#46 Word Length
https://www.garlic.com/~lynn/2012j.html#66 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012j.html#69 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012j.html#70 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012j.html#95 printer history Languages influenced by PL/1
https://www.garlic.com/~lynn/2012j.html#96 The older Hardware school
https://www.garlic.com/~lynn/2012k.html#41 Cloud Computing
https://www.garlic.com/~lynn/2012l.html#20 X86 server
https://www.garlic.com/~lynn/2012l.html#27 PDP-10 system calls, was 1132 printer history
https://www.garlic.com/~lynn/2012l.html#28 X86 server
https://www.garlic.com/~lynn/2012l.html#29 X86 server
https://www.garlic.com/~lynn/2012l.html#30 X86 server
https://www.garlic.com/~lynn/2012l.html#34 X86 server
https://www.garlic.com/~lynn/2012l.html#39 The IBM zEnterprise EC12 announcment
https://www.garlic.com/~lynn/2012l.html#41 The IBM zEnterprise EC12 announcment
https://www.garlic.com/~lynn/2012l.html#42 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012l.html#47 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012l.html#51 Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#57 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#67 Strings (hijacked from: The IBM zEnterprise EC12 announcment)
https://www.garlic.com/~lynn/2012l.html#71 Strings (hijacked from: The IBM zEnterprise EC12 announcment)
https://www.garlic.com/~lynn/2012l.html#72 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#74 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#77 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#79 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#81 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#82 zEC12, and previous generations, "why?" type question - GPU computing
https://www.garlic.com/~lynn/2012l.html#83 IBM's z12 mainframe engine makes each clock count
https://www.garlic.com/~lynn/2012l.html#87 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#88 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#90 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012l.html#100 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#5 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#11 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#13 Intel Confirms Decline of Server Giants HP, Dell, and IBM
https://www.garlic.com/~lynn/2012m.html#28 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012m.html#31 Still think the mainframe is going away soon: Think again. IBM mainframe computer sales are 4% of IBM's revenue; with software, services, and storage it's 25%
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#67 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#14 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#17 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#18 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#19 How to get a tape's DSCB

--
virtualization experience starting Jan1968, online at home since Mar1970

Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 14 Oct 2012
Subject: Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Blog: Enterprise Systems
re:
http://lnkd.in/NBbbzr
and
https://www.garlic.com/~lynn/2012n.html#44

recent ibm-main mailing list post
https://www.garlic.com/~lynn/2012l.html#59

some measurements ... but only goes up to 178BIPS for I7 (single chip, 8core)
https://en.wikipedia.org/wiki/Million_instructions_per_second

the dhrystone mips benchmark isn't direct calculation of instructions per second ... it is "dhrystone" mips ... the number of interations of the benchmark compared to the number of interations for a 1mip vax machine (vax instructions being comparable to 360/370) ... a machine that 4341 competed favorably against (and assumed to be roughly equivalent types of instruction & mip rate as vax).

machines might actually have radically different actual number of instructions to execute a dhrystone iteration ... but would be normalized to the same base ... the number of interations done per second

e5-2690 two socket, 8cores/chip
http://www.istorya.net/forums/computer-hardware-21/485176-intel-xeon-e5-2690-and-e5-2660-8-core-sandy-bridge-ep-review.html

Dhrystone
E5-2690 @2.9GHZ 527.55BIPS
E5-2660 @2.2GHZ 428.15BIPS
X5690 @3.45GHZ 307.49BIPS
i7-3690 @4.78GHZ 288BIPS
AMD 6274 @2.4GHZ 272.73BIPS

Whetstone
E5.2690 @2.9GHZ 315GFLOPS
E5-2660 @2.2GHZ 263.7GFLOPS
X5690 @3.4GHZ 227GFLOPS
i7-3690 @4.78GHZ 176GFLOPS
AMD 6274 @2.4GHZ 168.11GFLOPS

reference is that max EC12 is 75BIPS (for 101 processors)
http://www.theregister.co.uk/2012/09/05/ibm_z12_mainframe_engine/

that corresponds to other statements that max EC12 is 50% more than max z196 at 50BIPS (with 80 processors).

also from post
https://www.garlic.com/~lynn/2012j.html#1
in (linkedin) mainframe experts discussion
http://lnkd.in/QhQ73A

https://en.wikipedia.org/wiki/IBM_System_z9
https://en.wikipedia.org/wiki/IBM_System_z10

• 2005 Z9 17.8BIPS 54processor
• 2008 Z10 30BIPS 64processor
• 2010 Z196 50BIPS 80processor

and again from
https://en.wikipedia.org/wiki/Instructions_per_second

some historic numbers ... where in 8yrs (2003 to 2011) BIPS rate went from 9.7 to 178 (factor of 18times) ... claims that a large factor in that increase was the move to RISC cores and hardware layer translating i86 instructions to risc micro-ops ... as well as "clone" competition ... two different vendors competing in the same market

• 2003 INTEL 9.7BIPS
• 2005 AMD 14.5BIPS
• 2006 AMD 20BIPS
• 2006 INTEL 49BIPS
• 2008 INTEL 82BIPS
• 2009 AMD 78BIPS
• 2010 INTEL 147BIPS
• 2011 AMD 109BIPS
• 2011 INTEL 178BIPS

There is also a fair amount of difference between the "desktop" chip lines and the chip-line done for "servers".

wiki MIPS page lists both 158-3 and vax as 1MIPS (in terms of dhrystone iterations) ... which corresponds to statements of the time as to the instruction rates of the machines

In 1979, I did some benchmarking for the Endicott performance group for national lab on 4341 ... that was looking at getting computational farm of 70 4341s. The issue was that I was allowed to play disk engineer in bldgs. 14&15 ... in part because I provided them lots of software that significantly increased their productivity ... which also gave me lots of access&leeway in using their machines. They tended to get the 2nd or 3rd engineering model (370s both from endicott and POK) ... to start doing disk i/o testing. As a result I had much better access to 4341 than the Endicott performance group.

rain elapsed time
• 158 45.64 sec
• 3031 37.03 secs
• 4341 36.21 secs

rain4 elapsed time
• 158 43.9 secs
• 3031 36.61 secs
• 4341 36.13 secs

also times approx: 145: 145secs, 168-3: 9.1 secs, 360/91: 6.77 secs, cdc6600: 35.77 secs.

old email mentioning 4341
https://www.garlic.com/~lynn/lhwemail.html#4341

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Sun, 14 Oct 2012 17:33:20 -0400
Shmuel (Seymour J.) Metz <spamtrap@library.lspace.org.invalid> writes:
Actually, the massive changes were from OS/VS2 Release 1 (SVS) to OS/VS2 Release 2 (MVS) and from MVS/SP Version 1 to MVS/SP Version 2 (MVS/XA); the latter dropping the S/360 I/O instructions lock, stock and barrel.

re:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each

much of 370/xa architecture that was done specifically tailored to helping MVS shortcomings ... in much the same way that endicott did "E-architecture" for 43xx machines for DOS/VS & VS1 ... although in endicott's case, many of the 43xx machines continued to run in 370 mode as opposed to e-architecture mode. misc. old 43xx email ... where different models were periodically referred to some "E" machine model.
https://www.garlic.com/~lynn/lhwemail.html#43xx

MVS had enormous pathlength in interrupt handling and SIO driver ... and disk thruput was becoming increasingly limited by that duration ... as elapsed time for disk i/o decreased ... the "redrive" time for MVS ... to take the interrupt indicating previous disk i/o had finished, process the interrupt, dequeue the next pending disk i/o and redrive the device ... was becoming increasing percentage of elapsed time (increasingly limiting factor in i/o thruput per second).

the change for 370/xa defined SSCH to replace SIO ... which basically offloaded taking the interrupt and redriving the next queued request to dedicated processor.

these recent posts mentionings max configuration z196 doing 2m SSCH/sec with 104 FICON channels, 14 storage subsystems, and 14 "System Assist Processors". Theoritically 14 SAPs running at 100% busy are capable of 2.2M SSCH/sec ... but recommendation is to keep SAPs at 70% busy or less (1.5M SSCH/sec).
https://www.garlic.com/~lynn/2012n.html#4
https://www.garlic.com/~lynn/2012n.html#44

In any case while SSCH was being designed/architecture, I was interested in how quickly I could do I/O redrive in straight 370 (w/o offload dedicated processor).

I was rewriting the I/O supervisor for the disk engineering and product test labs (bldg. 14&15) ... to allow them to do concurrent development testing in operating system environment. They had previously tried MVS ... but found in had 15min MTBF (requiring manual re-ipl) with single device testing. As a result they dropped back to doing dedicated stand-alone testing, prescheduled, 7x24 around the clock. My objective with I/O supervisor rewrite was make it bullet proof and never fail, allowing concurrent, on-demand, anytime testing ... significantly improving development productivity.

While, I was doing the rewrite, I figured I could try and make the interrupt/redrive pathlength as short as possible ... w/o impacting integrity of system and I/O operation. I never made it to quite what a separate dedicated external processor ... but significantly improvement over the operating systems of the time ... especially compared to the enormous elapsed time it took MVS (indicating the significant performance improvement justifying SSCH was primarily based on the comparison to the enormous poor throughput in the MVS implementation).

misc. past posts mentioning getting to play disk engineer in bldgs. 14&15
https://www.garlic.com/~lynn/subtopic.html#disk

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Sun, 14 Oct 2012 23:23:26 -0400
Anne & Lynn Wheeler <lynn@garlic.com> writes:
I was rewriting the I/O supervisor for the disk engineering and product test labs (bldg. 14&15) ... to allow them to do concurrent development testing in operating system environment. They had previously tried MVS ... but found in had 15min MTBF (requiring manual re-ipl) with single device testing. As a result they dropped back to doing dedicated stand-alone testing, prescheduled, 7x24 around the clock. My objective with I/O supervisor rewrite was make it bullet proof and never fail, allowing concurrent, on-demand, anytime testing ... significantly improving development productivity.

re:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each

as an aside, I did an internal report on all the changes for the disk engineering and product test labs ... and happened to reference that they had previously tried MVS for concurrent testing and found it had 15min MTBF ... which turned out to bring the wrath of the MVS organization on my head ... I think they would have gotten me fired if they could ... but they managed to make the rest of my career at ibm as unpleasant as possible whenever they could.

discussion of bringing down wrath of the MVS group and possible intersection with change in IBM culture after FS failure to sycophancy and make no waves (i.e. the culture change after FS failure appeared to promote/allow such behavior as acceptable)
https://www.garlic.com/~lynn/2009o.html#31 Justice Department probing allegations of abuse by IBM in mainframe computer market

past posts mentioning allowed to play disk engineer in bldgs 14&15
https://www.garlic.com/~lynn/subtopic.html#disk

misc. other past posts mentioning including MVS 15min MTBF reference bringing the wrath of the MVS group down on my head:
https://www.garlic.com/~lynn/2009h.html#68 My Vintage Dream PC
https://www.garlic.com/~lynn/2009h.html#71 My Vintage Dream PC
https://www.garlic.com/~lynn/2009k.html#17 Bulletproof
https://www.garlic.com/~lynn/2009k.html#49 A Complete History Of Mainframe Computing
https://www.garlic.com/~lynn/2009k.html#62 Hercules; more information requested
https://www.garlic.com/~lynn/2009o.html#17 Broken hardware was Re: Broken Brancher
https://www.garlic.com/~lynn/2009p.html#0 big iron mainframe vs. x86 servers
https://www.garlic.com/~lynn/2009q.html#74 Now is time for banks to replace core system according to Accenture
https://www.garlic.com/~lynn/2010b.html#38 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#100 "The Naked Mainframe" (Forbes Security Article)
https://www.garlic.com/~lynn/2010c.html#28 Processes' memory
https://www.garlic.com/~lynn/2010d.html#45 What was old is new again (water chilled)
https://www.garlic.com/~lynn/2010d.html#59 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#30 SHAREWARE at Its Finest
https://www.garlic.com/~lynn/2010f.html#53 F.B.I. Faces New Setback in Computer Overhaul
https://www.garlic.com/~lynn/2010g.html#0 16:32 far pointers in OpenWatcom C/C++
https://www.garlic.com/~lynn/2010g.html#32 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#44 16:32 far pointers in OpenWatcom C/C++
https://www.garlic.com/~lynn/2010h.html#10 Far and near pointers on the 80286 and later
https://www.garlic.com/~lynn/2010h.html#17 LINUX on the MAINFRAME
https://www.garlic.com/~lynn/2011o.html#23 3270 archaeology
https://www.garlic.com/~lynn/2012e.html#83 Why are organizations sticking with mainframes?
https://www.garlic.com/~lynn/2012f.html#24 Time to competency for new software language?
https://www.garlic.com/~lynn/2012i.html#28 Top Ten Reasons Why Large Companies Fail To Keep Their Best Talent
https://www.garlic.com/~lynn/2012j.html#13 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012j.html#29 How smart do you need to be to be really good with Assembler?
https://www.garlic.com/~lynn/2012j.html#52 The dbdebunk revival
https://www.garlic.com/~lynn/2012m.html#6 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee

--
virtualization experience starting Jan1968, online at home since Mar1970

Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 15 Oct 2012
Subject: Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Blog: Enterprise Systems
re:
http://lnkd.in/NBbbzr
and
https://www.garlic.com/~lynn/2012n.html#44
https://www.garlic.com/~lynn/2012n.html#45

Note that running at 100% utilization and still providing instantaneous on-demand response takes a lot of design, work and care&feeding. The economics trade-off of providing that level of sophistication is quite different if the system cost is running $3.44/BIPS versus $3.5M/BIPS (a factor of million difference in system cost). It is lot easier to justify the sophisticated support, care&feeding when the system cost is million times more expensive.

There are numerous cloud mega-datacenters around the world ... everyone with more processing power than the aggregate of all mainframes in the world today. There have been several stories about organizations carving out an "on-demand" supercomputer from a cloud operation ... that the "supercomputer" is provisioned, spun-up, run for a couple hrs and then shutdown ... all without actually talking to anybody at the cloud operations ... just using the standard online facilities. And several of these "on-demand" supercomputers (available processing power subset of a cloud operation) represent more processing power than the aggregate of the processing power of all mainframes in the world today. Imagine an organization having large number of mainframes ... several times that of all mainframes in the world today .. all sitting idle ... so that they would be available for instantaneous, on-demand use. One of the reasons that these mega-datacenters are like aircraft hangers ... is because each one contain multiple times the processing of the aggregate of all mainframes in the world today.

There was recent article that the number of datacenters (in the world) are decreasing but the avg. size is increasing ... in part because of the growth of the cloud megadatcenters ... with organizations moving some amount of their processing to the cloud.

In the 70s&80s ... my custom vm370 kernels ran on large number of internal datacenters and were expected to run at 100% utilization with near instantaneous interactive response ... when the equivalent MVS systems were recommended to run at 60-70% utilization in order to provide "decent" TSO response (where the "decent" TSO response was considered horrible and unacceptable by vm370/cms users). This is somewhat analogous to the recent published benchmarks of z196 I/O capacity and recommendation that SAPs are kept to 70% cpu utilization for SSCH processing (i..e. 1.5M SSCH/sec instead of peak possible 2.2M SSCH/sec).

for other topic drift ... recent discussion in a.f.c. regarding original design for SSCH and justification for move from SIO to SSCH in 370/xa:
https://www.garlic.com/~lynn/2012n.html#46 .
https://www.garlic.com/~lynn/2012n.html#47

note as mentioned above, max configured z196 was 2M IOPS using 104 FICONs, 14 storage subsystems (and 14 SAPs) ... even though standard recommendation is to keep SAP operation at 1.5M SSCH or less per second. This compares to recent announcement for single fiber-channel for e5-2600 that is capable of over million IOPS ... aka FICON is a layer built on top of underlying fibre-channel that enormously cuts fibre-channel throughput ... aka two of the e5-2600 fibre-channel would more than match the throughput of the z196 104 FICONs

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Mon, 15 Oct 2012 10:25:40 -0400
Shmuel (Seymour J.) Metz <spamtrap@library.lspace.org.invalid> writes:
What is "E-architecture"? I would have taken that as an older name for ESA, but neither DOS/VS[1] nor OS/VS1 supported that. I know of assists for those systems, and DOS/VSE supported ECPS:VSE, so I'm guessing that you mean one or more ECPS features.

[1] Although a successor to DOS/VSE eventually supported ESA.


re:
https://www.garlic.com/~lynn/2012n.html#46

dos/vs & os/vs1 were single virtual address space ... similar to SVS.

part of "E-architecture" moves the single virtual address space into the hardware (aka microcode). the 370 virtual memory tables effectively now exist in the hardware ... and there are instructions that activate entries (i.e. create a mapping between specific virtual page and a real page) and de-activate entries.

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Mon, 15 Oct 2012 10:29:07 -0400
Shmuel (Seymour J.) Metz <spamtrap@library.lspace.org.invalid> writes:
Current technology makes it possible to produce relativ3ly inexpensive processors only when you produce them in bulk. As a result, commodity processors these days are approximately as fast as mainframe processors if you don't count the complicated instructions. Where the mainframes have an advantage is:

1. They're designed for higher I/O bandwidth 2. They are designed to support a lot of processors under a single instance of the OS. 3. They're designed for better Reliability, Availability and Servicability (RAS) 4. There are some tasks that can be done with a single instruction that would require a loop on a commodity processor.

How significant those are depends on the particular installation.


re:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#47 history of Programming language and CPU in relation to each

recent discussion of recent mainframes vis-a-vis e5-2600 in (linkedin) mainframe discussion group:
https://www.garlic.com/~lynn/2012n.html#44 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
https://www.garlic.com/~lynn/2012n.html#45 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
https://www.garlic.com/~lynn/2012n.html#48 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Mon, 15 Oct 2012 12:07:56 -0400
Dan Espen <despen@verizon.net> writes:
MVS systems shine when it comes to handling huge work loads but they just don't seem that fast. That may change with the z/EC12.

re:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#47 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#49 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#50 history of Programming language and CPU in relation to each

some discussion in this (linkedin) mainframe discussion
https://www.garlic.com/~lynn/2012n.html#44 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
https://www.garlic.com/~lynn/2012n.html#45 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
https://www.garlic.com/~lynn/2012n.html#48 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

from discussion

max z10 has 64 processors and rated at 30BIPS or 469MIPS/processor max z196 has 80 processors and rated at 50BIPS or 625MIPS/processor max ec12 has 101 processors and rating of 75BIPS or 743MIPS/processor

part of improvement in aggregate BIPS rate is more processors. improvement in per core-processor MIPS rate has been slowly increasing ghz. However, major improvement from z10 to z196 is introduction of out-of-order instruction processing (something that has been in risc processor for decades to help mask cache miss latency ... and recent generations of i86 processors have gone to risc cores with hardware layer that translates i86 instructions into risc micro-ops).

part of improvement from z196 to ec12 is further improvements in out-of-order execution support (branch prediction, speculative execution, etc).

for DBMS / i/o intensive ... ec12 is claiming only 30% improvement over z196 (while raw processing is 50%). much of this could be the additional processing power w/o any significant change in i/o throughput.

the linkedin/mainframe discussion goes into some detail about mainframe FICON and the inefficiencies introduced compared to underlying fibre-channel standard. IBM articles show that they appear to partial compensated for those inefficiencies with recent zHPF and TCW ... but also says that there has been no change going from z196 to ec12.

any potential for significant change appears to be financial and other incentives to have application utilize mainframe "specialty" processors (i.e. have been added to the environment ... that are other than the traditional "main" processors) that are increasingly being used in the mainframe environment.

this beginning to look more & more like this effort I had circa 1985 to configure arbitrary mix of mainframe and non-mainframe processors in the same racks ... discussed in this recent ibm-main mailing list post
https://www.garlic.com/~lynn/2012l.html#82 zEC12, and previous generations, "why?" type question - GPU computing

it was going on concurrently with working with NSF with what was to become the NSFNET backbone (operational precursor to modern internet) ... having to juggle the rack meetings with meetings with director of NSF ... a couple old email reference
https://www.garlic.com/~lynn/2011b.html#email850314
https://www.garlic.com/~lynn/2007d.html#email850315

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Mon, 15 Oct 2012 17:17:32 -0400
Shmuel (Seymour J.) Metz <spamtrap@library.lspace.org.invalid> writes:
OS/VS1[1] didn't exploit ECPS:VSE, although it did support VM handshaking.

[1] As shipped yo customers; I have no idea what might have been running in the labs.


re:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#47 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#49 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#50 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#51 history of Programming language and CPU in relation to each

after the demise of the future system effort ... there was mad rush to get products back into the 370 pipeline. POK kicked off 303x effort in parallel with 3081&370xa. Endicott kicked off 138/148 in parallel with e-machines (becoming 4331 & 4341). past posts mentioning FS
https://www.garlic.com/~lynn/submain.html#futuresys

138/148 were follow-on to 135/145. they had microcode space and had put some stuff in microcode for vs1 (ECPS) that they thot might improve thruput (but didn't actually have any hard&fast information/criteria).. I then got roped into helping them with doing something similar for vm370 ... I was told that there was 6kbytes left (after what was taken up by ECPS for VS1). For this I did several runs that did detailed analysis of where vm370 kernel was spending its time, organized by instruction paths ... and sorted by percent use. It was then "lopped" off at 6kbyte (kernel 370 instructions dropped into microcode on approx. byte-for-byte basis ... even tho microcode was completely different instruction set than 370). some of the results reproduced here ... that resulted in things selected for vm370 ECPS
https://www.garlic.com/~lynn/94.html#21

these machines avg. approx. ten microcode instructions per emulated 370 instruction ... so when dropped directly into native microcode, it ran approx. ten times faster. The 6kbyte threshhold accounted for approx. 80% of vm370 kernel execution time ... so dropped into microcode it was speeded up by a factor of ten times.

endicott also con'ed me into going around to the marketing and product forcasting groups as part of justifying 138/148. One of the things I found was that domestic US forcasting tended to follow hdqtrs strategic direction ... since forcasting didn't actually result in anything except showing that you were toeing what ever corporate direction. By comparison the non-US forcasting was significantly more reliable since countries ordered machines from manufacturing and they were shipped to the country ... mis-forcasts were responsibility of the forcasting group and country (by comparison in the US, machines shipped directly to customer and mis-forcasts had to be eaten by the manufacturing plant ... as a result, manufacturing always redid their own product forcasts for the US).

Non-us said that without something unique, the forcast for 138/148 was nearly zero ... because of clone competition outside of the US. By comparison, US forcasting was that there was a set formula for 138/148 sales regardless of the features. Endicott attempted to make every 138/148 sold a vm370 machine (installed at the factory) ... sort of an earlier version of the mainframe LPAR facility ... to help with clone competition. They were shot down badly by corporate ... which was in the process of killing the vm370 product ... having been convinced by POK that it was necessary to kill vm370, shutdown the group and move all the people to POK or otherwise POK wouldn't be able to ship MVS/XA on schedule.

As a result 138/148 had vm370 ship as optional feature ... along with vm370 ECPS ... but VS1 did ship with VM370 handshaking (i.e. on a virtual machine page fault, vm370 could reflect a psuedo page fault to VS1 allowing VS1 to perform task-switch ... and then later vm370 reflect a psuedo i/o interrupt indicating that the page was now available). There is a very gross analogy with out-of-order instruction execution (being able to switch to a different instruction while waiting for data for previous instruction because of cache miss).

Also, Endicott managed to save the vm370 product mission ... but had to reconsitute a vm370 development group from scratch.

4300s managed to show up in 1979 ... well before 3081 ... some old 4300 email
https://www.garlic.com/~lynn/lhwemail.html#43xx

It would run in either 370 mode or e-architecture mode (having been specifically designed for single address space dos and vs1). Also the ECPS microcode "assists" from 138/148 were carried forward to 4331/4341.

I have no idea how many machines ran in e-architecture mode ... or whether most of them continued running in straight 370 mode (all the vm370 43xx machines of course ran in 370 mode ... since a unique virtual address space was used for every virtual machine).

43xx machines sold into the same mid-range market as vax/vms and in similar numbers for orders of one or few machines. big difference for 43xx were the large corporate multi-hundred machine orders ... sort of the leading edge of the distributed computing tsunami. old post with decade of vax numbers sliced&diced in various ways.
https://www.garlic.com/~lynn/2002f.html#0

the 43xx sales did represent something of threat to the POK 3033 sales. distributed computing tsunami was siphoning a lot of computing from datacenter ... slowing datacenter growth. However, a 4341 cluster was better price/performance, larger aggregate processing, larger aggregate i/o, smaller space footprint, smaller environmental footprint, etc. This was at a time when many datacenters were feeling space/environmental crunch with the large POK machines ... with both distributed 4341 and 4341 clusters providing relief.

At one point, POK responded to the 4341 threat by getting corporate to cut allocation for critical 4341 manufacturing component in half.

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Mon, 15 Oct 2012 17:34:16 -0400
Ahem A Rivet's Shot <steveo@eircom.net> writes:
Mo I'm talking about distributed transactions where to commit a transaction requires several machines to commit it, handling this safely allowing for machine failures requires a two phased commit protocol and some very carefully designed restart logic. If some of the individual machine transactions turn out to be distributed themselves then that's when the protocols start iterating and it gets hard.

when I was doing ha/cmp distributed locking for cluster scale-up ... past posts here
https://www.garlic.com/~lynn/subtopic.html#hacmp
old post referencing early JAN92 meeting on cluster scale-up in ellison's conference room
https://www.garlic.com/~lynn/95.html#13

they had something called fast-commit where log record was written, but the actual RDBMS record hadn't yet been written back to disk. however, in loosely-coupled/distributed enviornment they would disable fast-commit ... if lock moved to different system ... the RDBMS record had to be first flushed to disk and the other system would read it from disk when it got the lock.

The lock could be transferred over communication link ... but the actual record could only move through disk. I did a bunch of stuff so that I could do memory-to-memory transfer of DBMS record that piggy-backed on the same transmission that moved lock ownership.

However, the DBMS vendors were very apprehensive about actually doing it. In the case where fast-commit was disabled in multi-system environment, recovery for any record required at most a single log. Where fast-commit was allowed for multi-system environment ... with direct memory to memory transfers w/o requiring disk write/read ... recovery would require mult-log merge ... and the log records had to merge in the same exact sequence as the original transaction order.

It wasn't until the past couple years ... that vendors have gotten sufficiently confident about correctly doing recovery with multi-log merge (it use to be peoples' eyes would tend to glaze over as you get deep into the sequence of correct multi-log merge).

some of this contributed to 2009 post:
https://www.garlic.com/~lynn/2009p.html#43 From The Annals of Release No Software Before Its Time

misc past posts mentioning distribted lock manager/DLM
https://www.garlic.com/~lynn/2001c.html#66 KI-10 vs. IBM at Rutgers
https://www.garlic.com/~lynn/2001e.html#2 Block oriented I/O over IP
https://www.garlic.com/~lynn/2001e.html#4 Block oriented I/O over IP
https://www.garlic.com/~lynn/2001j.html#47 OT - Internet Explorer V6.0
https://www.garlic.com/~lynn/2001k.html#5 OT - Internet Explorer V6.0
https://www.garlic.com/~lynn/2002e.html#67 Blade architectures
https://www.garlic.com/~lynn/2002e.html#71 Blade architectures
https://www.garlic.com/~lynn/2002f.html#1 Blade architectures
https://www.garlic.com/~lynn/2002f.html#4 Blade architectures
https://www.garlic.com/~lynn/2002f.html#5 Blade architectures
https://www.garlic.com/~lynn/2002f.html#6 Blade architectures
https://www.garlic.com/~lynn/2002f.html#7 Blade architectures
https://www.garlic.com/~lynn/2002k.html#8 Avoiding JCL Space Abends
https://www.garlic.com/~lynn/2004i.html#1 Hard disk architecture: are outer cylinders still faster than inner cylinders?
https://www.garlic.com/~lynn/2004i.html#2 New Method for Authenticated Public Key Exchange without Digital Certificates
https://www.garlic.com/~lynn/2004m.html#0 Specifying all biz rules in relational data
https://www.garlic.com/~lynn/2004m.html#5 Tera
https://www.garlic.com/~lynn/2004q.html#10 [Lit.] Buffer overruns
https://www.garlic.com/~lynn/2004q.html#70 CAS and LL/SC
https://www.garlic.com/~lynn/2004q.html#71 will there every be another commerically signficant new ISA?
https://www.garlic.com/~lynn/2005.html#40 clusters vs shared-memory (was: Re: CAS and LL/SC (was Re: High Level Assembler for MVS & VM & VSE))
https://www.garlic.com/~lynn/2005f.html#32 the relational model of data objects *and* program objects
https://www.garlic.com/~lynn/2005h.html#26 Crash detection by OS
https://www.garlic.com/~lynn/2005i.html#42 Development as Configuration
https://www.garlic.com/~lynn/2006c.html#8 IBM 610 workstation computer
https://www.garlic.com/~lynn/2006c.html#41 IBM 610 workstation computer
https://www.garlic.com/~lynn/2006j.html#20 virtual memory
https://www.garlic.com/~lynn/2006o.html#32 When Does Folklore Begin???
https://www.garlic.com/~lynn/2006o.html#33 When Does Folklore Begin???
https://www.garlic.com/~lynn/2006o.html#62 Greatest Software, System R
https://www.garlic.com/~lynn/2006x.html#3 Why so little parallelism?
https://www.garlic.com/~lynn/2007c.html#42 Keep VM 24X7 365 days
https://www.garlic.com/~lynn/2007i.html#61 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007l.html#19 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007l.html#24 Is Parallel Programming Just Too Hard?
https://www.garlic.com/~lynn/2007m.html#55 Capacity and Relational Database
https://www.garlic.com/~lynn/2007n.html#49 VLIW pre-history
https://www.garlic.com/~lynn/2007q.html#33 Google And IBM Take Aim At Shortage Of Distributed Computing Skills
https://www.garlic.com/~lynn/2007s.html#46 "Server" processors for numbercrunching?
https://www.garlic.com/~lynn/2007v.html#42 Newbie question about db normalization theory: redundant keys OK?
https://www.garlic.com/~lynn/2007v.html#43 distributed lock manager
https://www.garlic.com/~lynn/2007v.html#47 MTS memories
https://www.garlic.com/~lynn/2008b.html#69 How does ATTACH pass address of ECB to child?
https://www.garlic.com/~lynn/2008d.html#25 Remembering The Search For Jim Gray, A Year Later
https://www.garlic.com/~lynn/2008d.html#70 Time to rewrite DBMS, says Ingres founder
https://www.garlic.com/~lynn/2008g.html#56 performance of hardware dynamic scheduling
https://www.garlic.com/~lynn/2008h.html#91 Microsoft versus Digital Equipment Corporation
https://www.garlic.com/~lynn/2008i.html#18 Microsoft versus Digital Equipment Corporation
https://www.garlic.com/~lynn/2008k.html#63 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008r.html#71 Curiousity: largest parallel sysplex around?
https://www.garlic.com/~lynn/2009.html#3 Is SUN going to become x86'ed ??
https://www.garlic.com/~lynn/2009b.html#40 "Larrabee" GPU design question
https://www.garlic.com/~lynn/2009b.html#43 "Larrabee" GPU design question
https://www.garlic.com/~lynn/2009h.html#26 Natural keys vs Aritficial Keys
https://www.garlic.com/~lynn/2009k.html#36 Ingres claims massive database performance boost
https://www.garlic.com/~lynn/2009k.html#67 Disksize history question
https://www.garlic.com/~lynn/2009m.html#39 ACP, One of the Oldest Open Source Apps
https://www.garlic.com/~lynn/2009m.html#84 A Faster Way to the Cloud
https://www.garlic.com/~lynn/2009o.html#57 U.S. begins inquiry of IBM in mainframe market
https://www.garlic.com/~lynn/2009p.html#35 DB2 announces technology that trumps Oracle RAC and Exadata
https://www.garlic.com/~lynn/2010b.html#32 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010k.html#54 Unix systems and Serialization mechanism
https://www.garlic.com/~lynn/2010l.html#14 Age
https://www.garlic.com/~lynn/2010n.html#82 Hashing for DISTINCT or GROUP BY in SQL
https://www.garlic.com/~lynn/2011.html#23 zLinux OR Linux on zEnterprise Blade Extension???
https://www.garlic.com/~lynn/2011f.html#8 New job for mainframes: Cloud platform
https://www.garlic.com/~lynn/2012d.html#28 NASA unplugs their last mainframe

--
virtualization experience starting Jan1968, online at home since Mar1970

The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years

From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 15 Oct 2012
Subject: The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years.
Blog: Enterprise Systems
re:
http://lnkd.in/aTwr-m

1966, 709/1401 ... 1967 360/30, 1968 360/67 (mostly ran as 360/65) ... still undergraduate but by then I had responsibility for the care & feeding of the production systems used for both academic and univ. administration.

a little take-off on the "backbone" theme ... the internal network was larger than the arpanet/internet from just about the beginning until sometime late '85 or early '86. misc. past posts mentioning internal network
https://www.garlic.com/~lynn/subnetwork.html#internalnet

One of the things that the VM networking had was effectively a logical gateway in every node *and* no limitation on network size. This is something that arpanet/internet didn't get until the great conversion from imp/host protocol to internetworking protocol on 1JAN1983 ... at a time when it had about 100 IMP networking nodes and possibly 250 connected hosts ... by comparison the internal network was on the verge of passing 1000 nodes. Past post with some of the network node updates from 1983 and a list of all worldwide internal locations that had one or more network nodes added during 1983:
https://www.garlic.com/~lynn/2006k.html#8

The HASP/JES networking support had many of the similar limitations as the original arpanet/internet ... it had limitation on number of network node definitions ... way less than the number of nodes in the internal network ... and what is worse, HASP/JES would discard traffic when either the originating node and/or the destination node wasn't in the local table definitions ... as a result a MVS system couldn't be trusted as a backbone node ... since it would be guaranteed to arbitrarily discard traffic (MVS systems could only be trusted as boundary nodes where they there were limited to discarding traffic where they were the destination ... but the origin node wasn't defined in their local table). misc. past posts mentioning hasp &/or hasp/jes networking (nji/nje)
https://www.garlic.com/~lynn/submain.html#hasp

The internal network was also non-SNA ... at least until its great change-over in the late 80s ... at a point when it would have been enormously more cost-effective and productive to have moved over to TCP/IP ... something akin to what BITNET-II did. misc past posts mentioning bitnet/earn
https://www.garlic.com/~lynn/subnetwork.html#bitnet

--
virtualization experience starting Jan1968, online at home since Mar1970

U.S. Sues Wells Fargo, Accusing It of Lying About Mortgages

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 15 Oct, 2012
Subject: U.S. Sues Wells Fargo, Accusing It of Lying About Mortgages
Blog: Financial Crime Risk, Fraud and Security
U.S. Sues Wells Fargo, Accusing It of Lying About Mortgages
http://community.corporatecompliance.org/CORPORATECOMPLIANCE/Communities/DiscussionGroups/ViewThread/?GroupId=97&MID=15692

tv business news just now had segment ... pointing out that likely result would be settlement (again) with relatively small fine (compared to amounts involved) as "small" cost of doing business (w/o having to admit wrong doing and helping to wall off from further litigation) ... including reference to too-big-to-jail. Discussion was why didn't they just settle the fine as fast as possible since the longer it drags out, some really ugly things could come out in the press.

U.S. sues Wells Fargo for mortgage fraud - Oct. 9, 2012
http://money.cnn.com/2012/10/09/news/wells-fargo-fraud/index.html
Wells Fargo Committed Mortgage Fraud To Maximize Profits, U.S. Suit Says
http://www.forbes.com/sites/halahtouryalai/2012/10/09/wells-fargo-committed-mortgage-fraud-to-maximize-profits-us-attorney-lawsuit/
U.S. Files Civil Mortgage Fraud Suit Against Wells Fargo
http://www.bloomberg.com/news/2012-10-09/u-s-files-civil-mortgage-fraud-suit-against-wells-fargo.html

Government Sues Wells Fargo for 'Reckless' Lending; Suit charges the FHA picked up hundreds of millions of dollars worth of bad mortgages
http://www.consumeraffairs.com/wells-fargo-news

from above:
As also alleged, Wells Fargo's bonus incentive plan -- rewarding employees based on the sheer number of loans approved -- was an accelerant to a fire already burning, as quality repeatedly took a back seat to quantity. ... At issue are more than 100,000 FHA-backed loans. Wells Fargo said the loans met federal lending guidelines when half of them didn't, the government alleges.

... snip ...

Just finished Bair's "Bull by the Horns"
https://www.amazon.com/Bull-by-the-Horns-ebook/dp/B0061Q688A

... she goes into some of her perspective that OCC, Federal Reserve and Sec. of Treasury allowed whole slew of things to go on ... some of it interfering with things that should have been under control of FDIC ... but also allowing other things totally outside of the control of the FDIC ... i.e. loan originators by non-depository institutions ... being packaged as securitized instruments and sold through wallstreet ... all outside of traditional banking structure (i.e. not using deposits as source of funds for making the loans ... but using the immediate sell-off of the toxic CDOs as maintaining source of funds to keep the loans primed). In other cases ... traditional bankng operations may have been booking the loans ... but again they weren't relying on deposits as source of funds ... so the FDIC audits with regard to deposits being at risk didn't come come into play.

One of the suggestions was to eliminate OCC ... since during the bubble they were so aligned ("captured") by their financial institutions to not provide any effective regulation ... then give FDIC regulatory/audit responsibility for all traditional banking operations ... and if too-big-to-fail is to continue (and not bring back Glass-Steagall) ... give Federal Reserve responsibility for the non-depository parts of bank holding companies.

As mentioned numerous times before SEC was effectively doing little or nothing last decade (example of some number of repeated settlements with various institutions were they were fined, promised to stop doing the bad stuff ... which they continued to do anyway ... the basis for the comments that the fines were so small compared to the sums involved ... that it effectively became viewed as cost of doing business).

This is case were SEC appeared to not being doing stuff on what might have been illegal short activity:

The Man Who Beat The Shorts
http://www.forbes.com/forbes/2008/1117/114.html

from above:
Watsa's only sin was in being a little too early with his prediction that the era of credit expansion would end badly. This is what he said in Fairfax's 2003 annual report: "It seems to us that securitization eliminates the incentive for the originator of [a] loan to be credit sensitive. Prior to securitization, the dealer would be very concerned about who was given credit to buy an automobile. With securitization, the dealer (almost) does not care."

... snip ...

<< opps, just double check and the above forbes URL appears to have recently gone 404 >>

this reference that illegal shorting and other illegal activity may be common place & wide spread and not having to worry about SEC
http://nypost.com/2007/03/20/cramer-reveals-a-bit-too-much/

--
virtualization experience starting Jan1968, online at home since Mar1970

Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 16 Oct 2012
Subject: Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Blog: Enterprise Systems
re:
http://lnkd.in/NBbbzr
and
https://www.garlic.com/~lynn/2012n.html#44
https://www.garlic.com/~lynn/2012n.html#45
https://www.garlic.com/~lynn/2012n.html#48

At least the straight through processing efforts with parallelizing technologies for the "killer micros" ... I did the speeds&feeds analysis early on and predicted that it wouldn't work ... but they were so enamored of their toys and toy demos they all kept charging ahead (never even doing any kind of their own speeds&feeds ... even rough ballpark).

I then got roped into doing major performance analysis & improvement on nearly 500k statement cobol application that ran these overnight settlement on 40+ max configured mainframes (>$30m per, i.e. number required to finish within the overnight window) ... no machine over 18months old. It had very large performance group that had been caring for the application for decades ... but they got somewhat myopic in what they were examining. I applied some other kinds of analysis to come up with some major improvements that had grown up over the decades.

In the middle of the last decade to financial industry group, I came back with another approach to do the re-engineering & parallelization ... all based on RDBMS & SQL (platform agnostic) ... including business rule based front-end. Initially the industry group was very excited ... but then the effort got shutdown ... comments was that there were still extensive scars left over in the executive ranks from the failed efforts from the 90s ... and it would take another generation or two before they had appetite to try again.

for other topic drift ... some old email mentioning effort that I had from 1985 to support arbitrary mixes of 370 processors and other types of processors in large number of racks:
https://www.garlic.com/~lynn/2011b.html#email850314 .
https://www.garlic.com/~lynn/2007d.html#email850315

I had also been working with NSF on what was to become NSFNET backbone (operational precursor to modern internet) and had to chose between rack/cluster meeting along with getting somebody else to make my presentation to the director of NSF ... old email
https://www.garlic.com/~lynn/lhwemail.html#nsfnet

As mentioned upthread, one of the big boosts in last few generations of i86 thruput was the migration to risc core with hardware layer that translates i86 instructions to risc micro-ops. Something similar can be seen for z196 & zEC12 with increasing incorporation of out-of-order execution ... something that has been part of risc for decades.

Note the above calculation, I used IBM's base list price for e5-2600 blades. The big cloud vendors have been saying their scale-of-operation is such that they have started doing their own assembly with careful choice of components (price/performance, lifetime TCO, etc) ... and are doing it for 1/3rd the cost of brand-name systems. The above calculations for million times difference in cost/BIPS may be closer to 3 million times difference when just considering the large cloud mega-datacenter operations.

Any statements about death of PC is related to the consumer boxes ... with migration to much smaller personal devices connected to the large cloud mega-datacenters.

--
virtualization experience starting Jan1968, online at home since Mar1970

Bull by the Horns

Refed: **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 17 Oct 2012
Subject: Bull by the Horns
Blog: Google+
re:
https://plus.google.com/u/0/102794881687002297268/posts/jGu3pKgQVLa

Bull by the Horns
https://www.amazon.com/Bull-by-the-Horns-ebook/dp/B0061Q688A

Sheila Bair Talks About Her Relationship With Tim Geithner And Who She Wanted For Treasury Secretary
http://www.businessinsider.com/sheila-bair-geithner-2012-10

Confidence Men
https://www.amazon.com/Confidence-Men-Washington-Education-ebook/dp/B0089LOKKS/

talks about how the economic "A-team" helped get the president elected ... but were going to choose the Swedish solution (in Japan or Swedish) and were also going to hold accountable those responsible. The president instead appoints the "B-team" many who participated in the economic bubble, weren't going to hold accountable those responsible ... and chose the Japan "zombie bank" solution.

Citigroup CEO Vikram Pandit Steps Down, Is Well Paid For . . . What Exactly?
http://www.rollingstone.com/politics/blogs/taibblog/citigroup-ceo-vikram-pandit-steps-down-is-well-paid-for-what-exactly-20121016

... what company did Greespan give an exemption to for violating Glass-Steagall ... while washington was lobbied for repeal of Glass-Steagall ... and who resigned after Glass-Steagall was repealed and joined what company as co-CEO (and Bair makes reference to Sec. of Treasury being their protege)

misc. recent posts mentioning Glass-Steagall
https://www.garlic.com/~lynn/2012.html#0 Revolution Through Banking?
https://www.garlic.com/~lynn/2012.html#25 You may ask yourself, well, how did I get here?
https://www.garlic.com/~lynn/2012b.html#16 Interview of Mr. John Reed regarding banking fixing the game
https://www.garlic.com/~lynn/2012c.html#32 US real-estate has lost $7T in value
https://www.garlic.com/~lynn/2012c.html#52 PC industry is heading for more change
https://www.garlic.com/~lynn/2012c.html#54 PC industry is heading for more change
https://www.garlic.com/~lynn/2012d.html#5 PC industry is heading for more change
https://www.garlic.com/~lynn/2012d.html#68 Memory versus processor speed
https://www.garlic.com/~lynn/2012e.html#1 The Dallas Fed Is Calling For The Immediate Breakup Of Large Banks
https://www.garlic.com/~lynn/2012e.html#16 Wonder if they know how Boydian they are?
https://www.garlic.com/~lynn/2012e.html#30 Senators Who Voted Against Ending Big Oil Tax Breaks Received Millions From Big Oil
https://www.garlic.com/~lynn/2012e.html#41 Why Are the Fed and SEC Keeping Wall Street's Secrets?
https://www.garlic.com/~lynn/2012f.html#31 Rome speaks to us. Their example can inspire us to avoid their fate
https://www.garlic.com/~lynn/2012f.html#71 When Mobile Telecommunications Routes Become Banks
https://www.garlic.com/~lynn/2012f.html#75 Fed Report: Mortgage Mess NOT an Inside Job
https://www.garlic.com/~lynn/2012g.html#9 JPM LOSES $2 BILLION USD!
https://www.garlic.com/~lynn/2012g.html#12 JPM LOSES $2 BILLION USD!
https://www.garlic.com/~lynn/2012g.html#16 Psychology Of Fraud: Why Good People Do Bad Things
https://www.garlic.com/~lynn/2012g.html#20 Psychology Of Fraud: Why Good People Do Bad Things
https://www.garlic.com/~lynn/2012g.html#28 REPEAL OF GLASS-STEAGALL DID NOT CAUSE THE FINANCIAL CRISIS - WHAT DO YOU THINK?
https://www.garlic.com/~lynn/2012g.html#56 Why Hasn't The Government Prosecuted Anyone For The 2008 Financial recession?
https://www.garlic.com/~lynn/2012g.html#59 Why Hasn't The Government Prosecuted Anyone For The 2008 Financial recession?
https://www.garlic.com/~lynn/2012g.html#82 How do you feel about the fact that today India has more IBM employees than US?
https://www.garlic.com/~lynn/2012g.html#84 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012g.html#87 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#25 US economic update. Everything that follows is a result of what you see here
https://www.garlic.com/~lynn/2012h.html#36 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#45 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#55 The Invention of Email
https://www.garlic.com/~lynn/2012h.html#82 Interesting News Article
https://www.garlic.com/~lynn/2012i.html#25 Can anybody give me a clear idea about Cloud Computing in MAINFRAME ?
https://www.garlic.com/~lynn/2012i.html#30 US Senate proposes national data breach notification act
https://www.garlic.com/~lynn/2012i.html#35 US Senate proposes national data breach notification act
https://www.garlic.com/~lynn/2012i.html#86 Should the IBM approach be given a chance to fix the health care system?
https://www.garlic.com/~lynn/2012j.html#28 Why Asian companies struggle to manage global workers
https://www.garlic.com/~lynn/2012j.html#47 Yahoo Password Breach: 7 Lessons Learned - Security - Attacks/breaches - Informationweek
https://www.garlic.com/~lynn/2012j.html#56 Failing Gracefully
https://www.garlic.com/~lynn/2012k.html#7 Is there a connection between your strategic and tactical assertions?
https://www.garlic.com/~lynn/2012k.html#9 Sandy Weill's About-Face on Big Banks
https://www.garlic.com/~lynn/2012k.html#12 The Secret Consensus Among Economists
https://www.garlic.com/~lynn/2012k.html#13 Is there a connection between your strategic and tactical assertions?
https://www.garlic.com/~lynn/2012k.html#31 History--punched card transmission over telegraph lines
https://www.garlic.com/~lynn/2012k.html#75 What's the bigger risk, retiring too soon, or too late?
https://www.garlic.com/~lynn/2012l.html#60 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012l.html#62 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012l.html#63 Singer Cartons of Punch Cards
https://www.garlic.com/~lynn/2012m.html#12 Does the IBM System z Mainframe rely on Security by Obscurity or is it Secure by Design
https://www.garlic.com/~lynn/2012m.html#58 General Mills computer
https://www.garlic.com/~lynn/2012m.html#59 General Mills computer
https://www.garlic.com/~lynn/2012m.html#63 General Mills computer
https://www.garlic.com/~lynn/2012m.html#71 General Mills computer
https://www.garlic.com/~lynn/2012n.html#1 STOP PRESS! An Auditor has been brought to task for a failed bank!
https://www.garlic.com/~lynn/2012n.html#3 General Mills computer
https://www.garlic.com/~lynn/2012n.html#20 General Mills computer
https://www.garlic.com/~lynn/2012n.html#55 U.S. Sues Wells Fargo, Accusing It of Lying About Mortgages

--
virtualization experience starting Jan1968, online at home since Mar1970

2012 History Conference

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 17 Oct 2012
Subject: 2012 History Conference
Blog: Google+
re:
https://plus.google.com/u/0/102794881687002297268/posts/fVphqt42NRA

History Conference.
http://www.usni.org/events/2012-us-naval-history-conference

Mitnick at history conference
http://blog.usni.org/2012/10/16/social-engineering-kevin-mitnick-history-conference

I'm not sure why it was history(?) conference, even Mitnick had little about history ... mostly about his security/PEN business. I don't remember it being on "take-down" book ... but part of the tracking was there was lots of his stuff going on NETCOM which had recently converted from (10mbit) ethernet to (100mbit) FDDI and the existing tools for scanning traffic couldn't keep up with FDDI rate and had to be rewritten to handle FDDI speeds.

One of panel was FS-ISAC mentions commercial banking hacks ... another one yesterday. At the presidential critical infrastructure protection meetings one of the biggest concerns about ISAC was the info leaking to the public, not because the bad guys would use it, because the public would find out how bad it was. In 1995, the industry was saying that commercial banking would never move to the internet because of a long list of vulnerabilities. Not one of the exploits since then wasn't on the original list (& by 1999 there were countermeasures for everyone of the exploits, for various reasons not deployed)

The lack of action over data breaches and the public victims from the resulting fraudulent financial transactions was one of the major reasons behind the cal. state data breach notification act ... some hope that the publicity would motivate corrective action and countermeasures (approx. half of the federal data breach bills since then would mostly eliminate requirement for notification).

PCI-DSS was created after the data breach notification legislation and has been used in arguments that notification is no longer needed. One of the criteria for PCI-DSS certification is not having a breach ... aka after having a breach, certification is revoked.

I periodically pontificate that the process is severaly mis-aligned and even if the planet is buried under miles of information hiding encryption, it still couldn't stop leaks (in part because the info is needed in dozens of business processes at millions of locations around the world).

Disclaimer: in the mid-90s, I helped draft an industry standard that slightly tweaks the current paradigm and eliminates crooks ability to use leaked information for fraudulent financial transactions (and there-for eliminates the motivation for majority of current breaches) ... it also significantly levels/flattens/commoditizes the trust landscape ... however there are significant stakeholders with enormous vested interest in the current status quo. old random trusted computing reference
https://web.archive.org/web/20010801203303/http://www.intel94.com/idf/spr2001/sessiondescription.asp?id=stp%2bs13

misc. recent posts mentioning data breach notification:
https://www.garlic.com/~lynn/2012.html#35 Israel vows to hit back after credit cards hacked
https://www.garlic.com/~lynn/2012c.html#3 zSeries Manpower Sizing
https://www.garlic.com/~lynn/2012c.html#6 The 15 Worst Data Security Breaches of the 21st Century
https://www.garlic.com/~lynn/2012d.html#49 Do you know where all your sensitive data is located?
https://www.garlic.com/~lynn/2012d.html#63 Fans of Threat Modelling reach for their guns ... but can they afford the bullets?
https://www.garlic.com/~lynn/2012e.html#17 Data theft: Hacktivists 'steal more than criminals'
https://www.garlic.com/~lynn/2012e.html#29 Visa, MasterCard warn of 'massive' security breach
https://www.garlic.com/~lynn/2012e.html#37 The $30 billion Social Security hack
https://www.garlic.com/~lynn/2012e.html#79 What's the takeaway on Audit?
https://www.garlic.com/~lynn/2012e.html#82 Fighting Cyber Crime with Transparency
https://www.garlic.com/~lynn/2012i.html#30 US Senate proposes national data breach notification act
https://www.garlic.com/~lynn/2012i.html#31 US Senate proposes national data breach notification act
https://www.garlic.com/~lynn/2012i.html#35 US Senate proposes national data breach notification act
https://www.garlic.com/~lynn/2012i.html#80 Firms told to own up to cybercrime attacks
https://www.garlic.com/~lynn/2012i.html#86 Should the IBM approach be given a chance to fix the health care system?
https://www.garlic.com/~lynn/2012j.html#56 Failing Gracefully
https://www.garlic.com/~lynn/2012j.html#61 The Myth of Password Complexity & Frequent Change Rules
https://www.garlic.com/~lynn/2012j.html#64 The Myth of Password Complexity & Frequent Change Rules
https://www.garlic.com/~lynn/2012j.html#68 The Myth of Password Complexity & Frequent Change Rules
https://www.garlic.com/~lynn/2012k.html#19 SnOODAn: Boyd, Snowden, and Resilience
https://www.garlic.com/~lynn/2012l.html#10 Cultural attitudes towards failure
https://www.garlic.com/~lynn/2012l.html#14 The growing openness of an organization's infrastructure has greatly impacted security landscape
https://www.garlic.com/~lynn/2012l.html#47 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012m.html#10 Does the IBM System z Mainframe rely on Security by Obscurity or is it Secure by Design

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Thu, 18 Oct 2012 10:28:16 -0400
scott@slp53.sl.home (Scott Lurndal) writes:
A big portion of the more effective optimization work in modern compilers is adapting the generated code to Optimize pipeline utilitization.

Intel and AMD offer optimization guides that give guidelines for efficient code generation, and much empirical research[*] goes into further optimizations. This is much more difficult to do when hand-coding assembler, and it makes the resulting hand-coded assembler more difficult to read and follow.

[*] Measuring L1/L2/L3 cache hit rates, processor write-buffer, return buffers, branch-prediction units, TLB miss rates, etc.


I mentioned having been brought in to look at the ten impossible things in major airline reservation system ... went away and rewrote the application in C, totally changed the paradigm, rehosted to distributed rs/6000, did all ten "impossible" things and came back in two months and demo'ed it.

Initially got 20 times performance improvement ... and then re-arranged order of things for cache sensitivity and got another five times ... for 100 times overall. then it was possible to do ten impossible things ... which took additional processing ... resulting in only ten times faster overall ... in terms of peak transactions/sec. However, base system was archaic user interface ... I collapsed three transactions that had to be manually strung together ... into a single transaction ... so the transactions/sec wasn't apples-to-apples, aka I needed only 1/3rd the number of transactions to accomplish the same thing ... but was still able to do ten times the number of more complex transactions/sec than the base.

ten 580s, were sized to handle all transactions for all airlines for all routes in the world (instead of just a specific or limited number of carriers).

more recent thread/discussion was a XScale with processing rate of those ten 580s
https://www.garlic.com/~lynn/2010b.html#80 Happy DEC-10 Day

more detail in the thread:
https://www.garlic.com/~lynn/2010b.html#73 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#74 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#76 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#78 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#79 Happy DEC-10 Day

--
virtualization experience starting Jan1968, online at home since Mar1970

The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 19 Oct 2012
Subject: The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years.
Blog: Enterprise Systems
re:
http://lnkd.in/aTwr-m
and
https://www.garlic.com/~lynn/2012n.html#54 The IBM mainframe has been the backbone of most of the world's largest IT organizations for more than 48 years.

IBM Science center machine room was on 2nd floor of 545 tech sq ... 360/67 and five 8+1 drive 2314 banks and one 5 drive 2314 bank. IBM FE painted the panels on each 2314 bank different colors ... to make them easier to identify when performing mount requests (yellow, green, blue, purple, red, and ???)

I had sponsored Boyd's briefings at IBM starting in the early 80s, he did stint command of NKP/spook-base 69-70 which Coram's Boyd biography lists as $2.5B "windfall" for IBM ... nearly $20B in today's dollars ... would have helped underwrite the disastrous IBM's Future System effort (various quotes are that IBM was the only computer company large enough that could have survived such a monumental disaster). gone 404 but lives on at the wayback machine describes some of spookbase (also references some 2250s and 360/65s)
https://web.archive.org/web/20030212092342/http://home.att.net/~c.jeppeson/igloo_white.html

past posts mentioning future system
https://www.garlic.com/~lynn/submain.html#futuresys

About the time Boyd was at NKP, I was brought in to help setup Boeing Computer Services ... installation of 360/67 and CP67 online services at hdqtrs off Boeing field. I would also have to periodic visit Renton datacenter which I thought (at the time) was possibly the largest in the world ... claiming something like $300M (in 1969 dollars) of IBM equipment. Summer of 1969 there was always pieces of 2-3 360/65s being staged in the halls around the datacenter waiting to be installed (arriving faster than they could be put on the floor). There was also disaster scenario where Mt. Rainier heats up resulting in monstrous mudslide that takes out the Renton datacenter ... so it was in the process of being replicated at the new 747 plant in Everett (business case that a week w/o Renton datacenter would cost the company more than the cost of the Renton datacenter). Also summer of 1969, 747 #3 could be seen flying Seattle skies for FAA certification.

In 1990 commandant of Marine Corps leverages Boyd to remake marine corp and there are now yearly meetings at Marine Corps univ to discuss how things are going (just last week). Last year that commandant wandered into the meetings and spoke for two hours (hadn't been on the agenda, but nobody was going to stop him). cheney also credit's Boyd with battle plan for desert storm.

one of the bible's is the "small wars manual" written in the mid-30s based on the latin american experience.
https://en.wikipedia.org/wiki/Small_Wars_Manual
however written about the same time about the same experience
https://en.wikipedia.org/wiki/War_Is_a_Racket
and biography of the current version ... by somebody from boston (I kept thinking I was reading fiction)
https://www.amazon.com/Confessions-Economic-Hit-Man-ebook/dp/B001AFF266

past posts & URLs mentioning Boyd
https://www.garlic.com/~lynn/subboyd.html

--
virtualization experience starting Jan1968, online at home since Mar1970

Should you support or abandon the 3270 as a User Interface?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 19 Oct 2012
Subject: Should you support or abandon the 3270 as a User Interface?
Blog: Enterprise Systems
re:
http://lnkd.in/bMXsgv

generally gui vis-a-vis cli (command line interface) comes up somewhat orthogonal to 3270. lot of 3270 started out sort of like glass-tty ... line interface on screen. later additional fullscreen interfaces (like editors) and menu interfaces.

the menu interfaces have been characterized a much more context rich interface ... especially for non-professional dataprocessing users. PROFS collected a bunch of cli applications and wrapped menus around them. initially inside the company it exploded the use by factor of at least ten times (aka most of the corporation was non-professional dataprocessing users). A professional dataprocessing expert can frequently be much more efficient with cli than with gui/menu. The "fullscreen" operation on 3270 (like original fullscreen editors) ... has been orthogonal to menu-mode operation.

There are a lot of gui/cli issues on one axis .... that is different axis from the 3270/non-3270 axis. lots of gui/cli issues can apply to whether it is a 3270 or non-3270.

So this is recent ibm-main discussion about .2second response ... minimum requirement for interactive computing ... the transition from channel attached 3272/3277 to 3274/3278 made it impossible to have interactive computing (similarly MVS made any kind of 3270 impossible for interactive computing). Detailed complaints to 3274/3278 product group came back with comment that 3274/3278 weren't designed for interactive computing but for data entry.
https://www.garlic.com/~lynn/2012m.html#37

this is recent ibm-main post about circa 1980 STL bldg. overflowing and 300 people from the IMS group were moved to offsite bldg. They were offered "remote" 3270s and basically had a revolt. I got con'ed into implementing channel extender that put channel attached, local 3270 controllers at the remote bldg ... giving the IMS group vm370/cms interactive computing (they couldn't tell difference between response on real channel attached and channel-extender attached)
https://www.garlic.com/~lynn/2012m.html#11

now there are some number of applications that have similar characteristics to that of 3270-menu type systems (including some number of browser-based menu applications) that run on fancy gui terminals. there are also some number of 3270 simulators running in separate teminal emulator windows that run on fancy gui terminals.

Some number of old-time 3270-based applications have gone thru some degree of maturity for certain class of human interactions (and require 3270 simulation to continue running) ... and their human factors compare favorably with some more recent, less mature implementations. Theoretically it would be possible to exactly reproduce the layout and interactions of those 3270-based applications w/o requiring the associated 3270 terminal simulation (there are relatively few real 3270s still in existence).

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Fri, 19 Oct 2012 16:26:46 -0400
Shmuel (Seymour J.) Metz <spamtrap@library.lspace.org.invalid> writes:
Impossible, adj.:

1. I don't know how to do it.

2. I've been ordered not to do it.

3. I don't have time to explain why it's a bad idea.

4. Doing it would cause political problems.


re:
https://www.garlic.com/~lynn/2012n.html#59 history of Programming language and CPU in relation to each

I demo'ed it ... and then the hand wringing started. Eventually they said that they hadn't really expect me to fix the problem, they just wanted to hire us as consultants and be able to tell the parent board for the next five years that we were looking at it. As it happens, one of the people on the board of the parent company, had over 15yrs earlier been executive in charge of the IMS group (before going off to do other stuff) ... when I had done a bunch of stuff for them.

at least part of the "impossible" was that they had possibly 800 people doing manual care&support of their current paradigm ... which in large part contributed to changing the paradigm , allowing the ten impossible things (and also eliminated the need for those 800; aka the executive in charge would have lost their empire).

turns out this particular operation was organized so that actual air passenger traffic wasn't the profit maker ... while reservation/ticket selling made most of the money. The air passenger traffic (actually flying passengers) could loose money, but the parent company could still be profitable from the money on the reservation/ticket side.

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Sat, 20 Oct 2012 10:30:18 -0400
greymausg writes:
The Internet has `brought out' the previous instances of fraud. Politicians have always lied, and relied, but it only happens every few years, and that is `what politicians do.' The internet allows everyone to try fraud, and makes people more `aware' that fraud is being attempted, more aware of it, and that is a good thing.

people are being directly exposed to the exploits that result in fraud ... and can almost consider the press about public taking precautions from such activity as obfuscation and misdirection ... away from the majority of fraud is still result of exploits that the public has no control over.

eariler in the week, us naval institute had history conference on cyber ... but there as little or no history ... lots of focus on various kinds of current exploits. In their blog:
I'm not sure why it was history(?) conference, even Mitnick had little about history ... mostly about his security/PEN business. I don't remember it being on "take-down" book ... but part of the tracking was there was lots of his stuff going on NETCOM which had recently converted from (10mbit) ethernet to (100mbit) FDDI and the existing tools for scanning traffic couldn't keep up with FDDI rate and had to be rewritten to handle FDDI speeds.

One of panel was FS-ISAC mentions commercial banking hacks ... another one yesterday. At the presidential critical infrastructure protection meetings one of the biggest concerns about ISAC was the info leaking to the public, not because the bad guys would use it, because the public would find out how bad it was. In 1995, the industry was saying that commercial banking would never move to the internet because of a long list of vulnerabilities. Not one of the exploits since then wasn't on the original list (& by 1999 there were countermeasures for everyone of the exploits, for various reasons not deployed)

The lack of action over data breaches and the public victims from the resulting fraudulent financial transactions was one of the major reasons behind the cal. state data breach notification act ... some hope that the publicity would motivate corrective action and countermeasures (approx. half of the federal data breach bills since then would mostly eliminate requirement for notification).

PCI-DSS was created after the data breach notification legislation and has been used in arguments that notification is no longer needed. One of the criteria for PCI-DSS certification is not having a breach ... aka after having a breach, certification is revoked.

I periodically pontificate that the process is severaly mis-aligned and even if the planet is buried under miles of information hiding encryption, it still couldn't stop leaks (in part because the info is needed in dozens of business processes at millions of locations around the world).

Disclaimer: in the mid-90s, I helped draft an industry standard that slightly tweaks the current paradigm and eliminates crooks ability to use leaked information for fraudulent financial transactions (and there-for eliminates the motivation for majority of current breaches) ... it also significantly levels/flattens/commoditizes the trust landscape ... however there are significant stakeholders with enormous vested interest in the current status quo. random long ago reference from trusted computing ... gone 404 but lives on at the wayback
https://web.archive.org/web/20011109072807/http://www.intel94.com/idf/spr2001/sessiondescription.asp?id=stp%2bs13


... snip ...

--
virtualization experience starting Jan1968, online at home since Mar1970

Should you support or abandon the 3270 as a User Interface?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 20 Oct 2012
Subject: Should you support or abandon the 3270 as a User Interface?
Blog: Enterprise Systems
re:
http://lnkd.in/bMXsgv
and
https://www.garlic.com/~lynn/2012n.html#61 Should you support or abandon the 3270 as a User Interface?

The new latest thing is downturn in PCs because of the migration to portable personal devices with wireless connection to the cloud.

IBM had attempted something in the late 80s with SAA ... but didn't succeed. In that time frame, a senior disk engineer got a talk scheduled at the internal, world-wide, annual, communication group conference and opened with the statement that the communication group was going to be responsible for the demise of the disk division. The issue was that the communication group was desperately trying to preserve their terminal emulation install base, fighting off client/server, distributed computing, etc. The disk division was seeing enormous amounts of data fleeing the datacenters because of the stranglehold that the communication group had. The disk division had come up with several products to address the problem ... allowing data to remain in the datacenter (sort of the current cloud) ... but since the communication group "owned" strategic responsibility for everything that crossed the datacenter walls ... the communication group could constantly veto whatever the disk division came up with.

misc. past posts on the subject:
https://www.garlic.com/~lynn/subnetwork.html#emulation

here is a totally different TSO/ISPF comparison with vm/cms.

the 23jun69 unbundling announcement was very traumatic to many parts of the corporation. "application" code not only had to be sold ... but the price had to cover the cost. the idea of cost effective development was very foreign to many parts of the corporation. however, there was a loophole, a development organization could show that their software revenue had to cover their development costs ... which could be done with bookkeeping across multiple applications. misc. past posts mentioning unbundling
https://www.garlic.com/~lynn/submain.html#unbundle

The first case I'm aware of was JES2 NJI networking. Typically company did high, medium, and low price sale price ... and then see if the resulting forecast numbers (times the price) covered the development cost (i.e. lower price could significantly increase forecast meaning more aggregate revenue than a higher price). JES2 NJI networking had no price point that resulting in covering development costs. There was the internal vm370 networking that they were never going to announce (POK was in the process of convincing corporate to kill vm370 product, shutdown the development group and move all the people to POK or otherwise POK wouldn't be able to meet the MVS/XA ship schedule ... Endicott eventually managed to save the vm370 product mission, but had to recreate a development group from scratch). A price for vm370 networking of nearly zero would still cover all vm370 development costs. The JES2 NJI group cut a deal where they could have a "joint" announcement. A fairly high price for the "joint" product resulted in a combined forecast that covered the JES2 NJI development&support costs. misc. past posts mentioning HASP, JES2, NJI, networking, etc
https://www.garlic.com/~lynn/submain.html#hasp

A couple years later, the bookkeeping got more relaxed ... two products just had to be in the same development group. The VM370 performance products was earning about the same revenue as TSO/ISPF and only had 3 people supporting ... while TSO/ISPF had a couple hundred people. They merge the two products into the same development group and effectively cut nearly all work on VM370 performance products ... so all that revenue could go to underwriting TSO/ISPF.

Lots of people may have been exposed to CMS IOS3270 menu w/o realizing it. The 3090 service processor was actually a pair of 4361s running a highly customized version of VM370 Release 6 and all the service processor menu screens were written in CMS IOS3270.

There was also a simulated "green card" written in IOS3270 that was used extensively inside the company. I've recently done a rough conversion to HTML ... and put it up as
https://www.garlic.com/~lynn/gcard.html

--
virtualization experience starting Jan1968, online at home since Mar1970

Fred Brooks on OS/360 "JCL is the worst language ever"

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Fred Brooks on OS/360 "JCL is the worst language ever"
Newsgroups: alt.folklore.computers
Date: Sat, 20 Oct 2012 17:10:55 -0400
Jim Haynes <jhaynes@alumni.uark.edu> writes:
Seems like the underlying problem is the assumption that the user wants to control all those parameters when all he really wants to do is get a file to store things in. It seems like yet another case of IBM solving the previous generation's problems in the current generation. The previous generation's problem was very limited capacity of disk storage media, so the user was allowed to fit data as tightly as possible into the available space. Other computer companies just let the user call for some number of blocks of storage, tolerating some waste of space in the last block. I assume that's what IBM eventually discovered in "Fixed Block Architecture".

1960s CKD architecture allowed trade-off with putting lots of structure and disk and using search commands that burned i/o at savings of real storage. by mid-70s the search architecture trade-off had started to invert (real storage becoming plentiful and i/o capacity increasingly becoming bottleneck resource) ... and by the late 70s aspects of fixed block architecture was starting to outperform ... in competition of CKD. the next couple CKD generations were actually low-level fixed-block cells ... which can be seen in various "space" calculations of how much data can be placed on track given various record sizes (roundup required to multiple of cell size). after that all CKD started to be simulated on industry standard fixec block architecture (i.e. real CKD disks haven't been manufactured for decades).

lots of past posts about ckd, search operations, dasd vis-a-vis fixed-block
https://www.garlic.com/~lynn/submain.html#dasd

note that cp67/cms & later vm370/cms was all logical fixed-block ... even when using ckd ... so when real fixed-block was introduced it was trivial to support.

however, the favorite son mainframe operation system has never supported fixed-block, being intricately tied to CKD ... even after there has been no real CKD built for decades. in the early 80s, I once offered them fixed-block support and was told I needed a $26M business case to cover the cost of publications and education ... and wasn't able to use total life-time costs as part of the justification ... just incremental additional disk sales ... on the order of $300M ... for the profit margin to cover the $26M publications and education ... and by the way, customers were supposedly buying CKD as fast as it could be built and any support for FBA would just result in the same amount of FBA substituted for CKD ... stacked deck.

recent post mentioning (oft repeated story) of disk division doing (late 80s) presentation that the communication group was going to be responsible for the demise of the disk division
https://www.garlic.com/~lynn/2012n.html#64

--
virtualization experience starting Jan1968, online at home since Mar1970

Fred Brooks on OS/360 "JCL is the worst language ever"

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Fred Brooks on OS/360 "JCL is the worst language ever"
Newsgroups: alt.folklore.computers
Date: Sat, 20 Oct 2012 17:17:38 -0400
Peter Flass <Peter_Flass@Yahoo.com> writes:
FSVO "previous generation." It seems to me that up until about 10-15 years ago disk storage was still a limited resource, and now that I think about it We still had to migrate stuff out with HSM in order to have enough space up until I left, a couple of years ago now, and they probably still do.

late 70s, the system/r (original SQL RDBMS) were being hassled by the IMS group ... the IMS group claiming that the implicit RDBMS index doubled the disk storage requirement compared to the same data in IMS ... and could increase numbers of I/Os by 5-6 times for index processing (again compared to IMS). The system/r retort was that IMS data tended to be cast in concrete and required enormous manual care&feeding to do almost anything.

By the early 80s, the significant increase in real storage availability was allowing caching for indexes and the drop in disk costs was dropping significantly (mitigating the double disk storage space) ... and the same time people costs was starting to explode ... when they could be found. The significant lower people/manual care&feeding (compared to IMS) coupled with change in hardware economics started see big shift to RDBMS.

misc. past posts mentioning original relational/sql
https://www.garlic.com/~lynn/submain.html#systemr

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Sat, 20 Oct 2012 17:36:10 -0400
Uncle Steve <stevet810@gmail.com> writes:
Stopping "leaks" from occurring is not the problem you think it is. The crush of information pummeling everyone at all times provides natural cover for the unwanted release of sensitive information. The major news networks and mass-media publications are all well-behaved in this regard. War is peace, freedom is slavery, ignorance is strength; words that were never more true than they are today.

Sure, the disclosure of credit-card numbers and the personal information of card-holders is bad. But what of the organizations that hold and control that information? I have a SIN (Cdn SSN equivalent) number that was manufactured deliberately, as opposed to being random issue. The motive for doing so was in the aid of a social engineering experiment, and not in a good way.


re:
https://www.garlic.com/~lynn/2012n.html#63 history of Programming language and CPU in relation to each other

what x9.59 financial industry standard did ... some references
https://www.garlic.com/~lynn/x959.html#x959

was not anything regarding stopping leakage ... it just tweaked the paradigm so that if credit card numbers and/or information from previous financial transactions leaked ... the information couldn't be used by crooks to perform fraudulent transactions (information could be plastered all over public locations and it would be useless to the crooks).

we had been doing this stuff on High Availability Cluster Multi-Processing scale-up ... reference here about meeting in ellison's conference room early Jan1992
https://www.garlic.com/~lynn/95.html#13

other email about cluster scale-up
https://www.garlic.com/~lynn/lhwemail.html#medusa

within possibly hrs of the last email in the above, the cluster scalup was transferred and we were told we couldn't work on anything with more than four processors. a couple weeks later it is announced as supercomputer.
https://www.garlic.com/~lynn/2001n.html#6000clusters1 920217

we then decide to leave. now two of the other people in the meeting from ellison's conference room ... also leave and show up at small client/server startup responsible for something called commerce server. We get brought in as consultants because they want to do payment transactions on the server, the startup had also invented this technology called SLL they want to use; the result is frequently now called electronic commerce. We have to map SSL to the payment transaction business process as well as walk through of most of the vendors selling digital certificates. I have final authority on all details on webserver to payment gateway and did various requirements about the browser to webserver interface ... but not final authority. Several of the browser/webserver security requirements were almost immediately violated.

Somewhat as result of doing electronic commerce, in the mid-90s, we were invited to participate in the x9a10 financial standard working group which had been given the requirement to preserve the integrity of the financial infrastructure for *ALL* retail payments. We do detailed end-to-end threat&vulnerability studies of various retail payment mechanisms in different environments as part of coming up with a secure solution standard for *ALL* retail payments.

--
virtualization experience starting Jan1968, online at home since Mar1970

Should you support or abandon the 3270 as a User Interface?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 20 Oct 2012
Subject: Should you support or abandon the 3270 as a User Interface?
Blog: Enterprise Systems
re:
http://lnkd.in/bMXsgv
and
https://www.garlic.com/~lynn/2012n.html#61 Should you support or abandon the 3270 as a User Interface?
https://www.garlic.com/~lynn/2012n.html#64 Should you support or abandon the 3270 as a User Interface?

old email from the 3092 (service processor) group
https://www.garlic.com/~lynn/2010e.html#email861031
https://www.garlic.com/~lynn/2010e.html#email861223

I had much earlier done some stuff with the manager that originally kicked off the 3092 service processor implementation ... as an alternative to what had been done for the 3081.

I had also introduced adventure inside ibm ... it had migrated to Stanford DEC10 machine and then been brought over to TYMSHARE's DEC10 machine (tymshare was large online vm370/cms commercial service bureau) and then ported to their vm370/cms system. Folklore was at one point TYMSHARE's CEO ordered that all games/non-business apps be removed from TYMSHARE ... until somebody told him that 30% of TYMSHARE's revenue was people playing games.

TYMSHARE started hosting free online VMSHARE discussion groups in AUG1976 (for SHARE organization) and very early on, I got process setup to get periodic copies of all VMSHARE to make them available on the internal network. vmshare archives
http://vm.marist.edu/~vmshare
sometimes(?) "404" ... but also at wayback machine
http://vm.marist.edu/~vmshare/

One of the systems I put VMSHARE files on was the internal HONE online vm370 world-wide sales&marketing support systems. I got early questions about why would HONE do that for me. They didn't realize that one of my hobbies had been providing highly enhanced production operating systems for internal datacenters ... and HONE had been one of my longtime customers since cp67 days (when HONE clones started being put up around the world ... I got my first trips outside the country to do various of the installations for them).

... not that ... but the author of rexx around 1980 did a multi-user spacewar game that ran on vm370/cms 3270. client code ran under cms driving 3270 interface and interacting with the spacewar "server" to coordinate all players activity. clients used "spm" interface to talk to spacewar server ... which was supported by vm370 networking ... so that players could be anyplace on the internal network. Problem was that fairly early some people wrote automated (robot) players and came to quickly dominate the games ... being able to do moves & firing much faster than human players. eventually the spacewar server was changed to increase energy use inversely proportional to interval between a players move ... somewhat leveling the playing field between robot and human players.

ibm had convinced lots of places to order 360/67 (effecitvely 360/65 with virtual memory hardware) for running tss/360. When tss/360 ran into lots of difficulties various places did their own virtual memory systems. Univ. of Michigan did (virtual memory) MTS, Stanford did Orvyl (and wilbur ran under orvyl), science center had done cp40 on 360/40 with special hardware mods for virtual memory ... and then when 360/67 become available cp40 morphed into cp67 (and much later into vm370). At some pointer Wilbur was ported to MVS.

Note that the first webserver outside europe was on the (stanford) SLAC vm370 system
http://www.slac.stanford.edu/history/earlyweb/history.shtml

part of the issue was mainframe tcp/ip support. The original product was on vm370 implemented in pascal/vs ... had some performance issues only getting 44kbytes/sec using nearly full 3090 processor. I did the enhancements for rfc1044 support and in some tuning tests at cray research got channel thruput between 4341 and cray (possibly 500 times improvement in bytes moved per instruction executed). some past posts mentioning rfc1044 support:
https://www.garlic.com/~lynn/subnetwork.html#1044

The base vm implementation was ported to MVS by implementing simulation of some vm370 kernel functions. Later I was working on tcp/ip implementation that had 5000 instruction pathlength and five buffer copies. This was compared to VTAM lu6.2 at 160k instruction pathlengths and 15 buffer copies.

A tcp/ip implementation in vtam was subcontracted out ... initially it was faster than lu6.2. The contractor was then told that everybody "knows" a correct tcp/ip implementation is slower than lu6.2 and ibm was only going to pay for a correct implementation.

Part of the issue why internal development (even major mvs oriented components) preferred local direct channel attached 3270s vm370/cms systems ... even vm/vtam channel attached 3270s significantly slowed things down ... and mvs was enormously worse (as mentioned upthread about 300 from ims group getting moved to offsite building)

--
virtualization experience starting Jan1968, online at home since Mar1970

Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 20 Oct 2012
Subject: Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Blog: Enterprise Systems
re:
http://lnkd.in/NBbbzr
and
https://www.garlic.com/~lynn/2012n.html#44
https://www.garlic.com/~lynn/2012n.html#45
https://www.garlic.com/~lynn/2012n.html#48
https://www.garlic.com/~lynn/2012n.html#56

with regard to upthread comment about large cloud mega-datacenters are doing their own assemblies claiming 1/3rd cost of brand name systems. they have also been on the bleeding edge of a lot of the other datacenter efficiency efforts, cooling, power consumption, support, etc. .... with possible factor of million times or more reduction in hardware cost/BIPS ... then other datacenter costs begin to represent larger & larger percentage of TCO (and require major attention).

if you have been a long time resident of silicon valley ... most of this isn't a surprise ... information pretty readily known. However, there has been whole slew about "secrets" this week regarding google

Google's Top Five Data Center Secrets (That Are Still Secret)
http://www.wired.com/wiredenterprise/2012/10/google-data-center-secrets/
Google Throws Open Doors to Its Top-Secret Data Center
http://www.wired.com/wiredenterprise/2012/10/ff-inside-google-data-center/all/
Google opens window into secretive data centers
http://phys.org/news/2012-10-google-window-secretive-centers.html
Google opens window into secretive data centers
http://news.yahoo.com/google-opens-window-secretive-data-centers-115411808--finance.html
Google Opens Up Data Centers for Virtual Tours
http://www.datamation.com/news/google-opens-up-data-centers-for-virtual-tours.html
Google Employs Stormtrooper to Protect Data Centers [PIC]
http://news.yahoo.com/google-employs-stormtrooper-protect-data-centers-pic-213306290.html
Out-of-Work Stormtrooper Lands Job Guarding Google's Servers
http://techland.time.com/2012/10/18/out-of-work-stormtrooper-lands-job-guarding-googles-servers/
Google Reveals Data Center Secrets - Hardware - Data centers -
http://www.informationweek.com/google-reveals-data-center-secrets/d/d-id/1106908
Google turns its data centers into an art show
http://www.computerworld.com/s/article/9232518/Google_turns_its_data_centers_into_an_art_show
Google turns its data centers into an art show
http://www.networkworld.com/news/2012/122412-it-shifts-from-owning-to-265342.html
Google now lets you explore its data centers with Street View [video]
http://news.yahoo.com/google-now-lets-explore-data-centers-street-view-205645747.html
Google Data Center Walkthrough
http://www.businessinsider.com/google-data-center-walkthrough-2012-10
Google Pays Engineers More Than Any Other Tech Company
http://www.businessinsider.com/google-pays-engineers-more-than-any-other-tech-company-2012-10
Google's Data Centers
http://techland.time.com/2012/10/17/take-a-look-inside-googles-data-centers/
Data Management: Google Razzle Dazzles With Data Center Display
http://www.technewsworld.com/story/76411.html
Photo Tour of Google's Data Centers
http://tech.slashdot.org/story/12/10/17/1355202/photo-tour-of-googles-data-centers
Thursday: A peek inside Google, tracking diplomats, big data spending
http://fcw.com/articles/2012/10/18/thursday-news-google-diplomats.aspx
Google Stays Ahead of Silicon Valley Pack on Engineer Salaries
http://blogs.wsj.com/digits/2012/10/17/google-stays-ahead-of-silicon-valley-pack-on-engineer-salaries/

In the google URLs ... there is a reference that the latest generation of their (blade) servers handles 20 times the capacity of their previous generation. You can see that in the current e5-2600 527BIPS and 2006 era 20-30 BIPS processing.

--
virtualization experience starting Jan1968, online at home since Mar1970

Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 21 Oct 2012
Subject: Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
Blog: Enterprise Systems
re:
http://lnkd.in/NBbbzr
and
https://www.garlic.com/~lynn/2012n.html#44
https://www.garlic.com/~lynn/2012n.html#45
https://www.garlic.com/~lynn/2012n.html#48
https://www.garlic.com/~lynn/2012n.html#56
https://www.garlic.com/~lynn/2012n.html#69

Cross-over reference to post in this group discussion mentioning SAA
http://lnkd.in/bMXsgv
and
https://www.garlic.com/~lynn/2012n.html#64

middle of 90s, a discussion asked what was the origin of middleware
https://www.garlic.com/~lynn/96.html#16 .
https://www.garlic.com/~lynn/96.html#17

in the SAA time-frame, there was a big gov. agency request for large multi-campus dataprocessing environment. my wife was co-author of response where 3-layer network & middle-layer was introduced ... some of it specified what later is referred to as middleware. we were then out making customer executive presentations on ethernet, (2-layer network) client/server and (3-layer network) middle-layer ... part of that presentation from the 80s:
https://www.garlic.com/~lynn/99.html#202 .

we were really being savaged by the token-ring and SAA forces (I've frequently characterized as trying to help preserve the terminal emulation install base) for doing those presentations ... lots of FUD and obfuscation. Token-ring forces came out with presentation about 16mbit t/r compared to ethernet. Their ethernet profile must have come from the original research 3mbit ethernet before CSMA/CD. The "new" Almaden research building had recently come online had been heavily provisioned with CAT4 supposedly for 16mbit T/R ... but they found 10mbit ethernet lan segment had both higher throughput as well as lower latency than 16mbit T/R.

Aggravating the issue was that SNA has no networking layer ... typical 16mbit T/R had 100-300 emulated 3270s sharing the same 16mbit with bridges. The client/server would have ethernet and tcp/ip routers so each 10mbit lan segment was separate. Some more pieces of the 3-tier customer executive presentation ... showing cost comparison
https://www.garlic.com/~lynn/2002q.html#40

past posts mentioning originating 3-tier architecture
https://www.garlic.com/~lynn/subnetwork.html#3tier

Another example of the enormous pressure trying to preserve the terminal emulation install base was the implementation of the microchannel 16mbit T/R cards designed for terminal emulation and very low use per card. The workstation division PC/RT had (16bit) AT-bus and designed their own 4mbit T/R card. The RS/6000 had (32bit) microchannel bus and were forced to use PS2 cards (and corporate direction that they couldn't do their own). The PS2 microchannel 16mbit T/R card had lower per-card throughput than the PC/RT 4mbit T/R card (i.e. the severely restricted 16mbit T/R per card thruput helping limit doing much more than terminal emulation).

misc. past posts mentioning terminal emulation
https://www.garlic.com/~lynn/subnetwork.html#emulation

x-over from Mainframes are still the best platform for high volume transaction processing.
http://lnkd.in/RsuUyw
... repeat of some information referenced up thread:

max configured zEC12 briefs have its DBMS throughput at 30% more than max configured z196 ... even though the processing rate is 50% more than z196. Part of this can be interpreted as the I/O throughput hasn't changed significantly from z196 (DBMS throughput increase primarily from additional processing but possibly bottlenecked by I/O).

The original FICON implementation severely restricted I/O throughput of the underlying fibre-channel standard (going back to 1988). The zHPF and TCW descriptions has it partially restoring to FICON some of the underlying fibre-channel capability and throughput .... however the documentation so far says that it is the same on zEC12 as z196.

max configured z196 has 80 processors rated at 50BIPS (and goes for $28M). z196 peak benchmarks have 2m IOPS using 104 FICON channels, 14 storage subsystems and 14 system assist processors. However, SAP reference has peak SSCH capacity of 2.2M/sec with all 14 SAPs running at 100% utilization ... but recommendation is that SAPs are kept at 70% or lower utilization (i.e. peak 1.5M SSCH/sec).

A recent announcement has single fibre-channel for E5-2600 blade capable of over million IOPS (aka two E5-2600 blade fibre channel having more I/O thruput than 104 FICON channels). E5-2600 blade have 527BIPS rating and IBM base list price of $1815. An IBM blade chassis holds something like 12 blades ... and possible to get something like 3 or 4 chassis in single rack ... possibly 48 blades in single rack with aggregate processing rate of 25,296BIPS (and around $90K ... compared to 50BIPS and $28M for z196)

.... snip ...

Note that real CKD havn't been manufactured for decades ... being an extra overhead emulation layer built on top of industry standard "fixed-block" disks ... note that all other things being equal ... having the extra emulation layer would introduce additional throughput degradation ... compared to systems directly using the disks.

I've frequently pontificated about long ago and far away offering FBA support and was told that even if I gave them fully integrated and tested FBA support, I still needed a $26M business case to cover education, training and publications ... and I wasn't able to use total lifetime costs ... only incremental new sales. To cover the $26M, I would need something like $300M new sales ... but was told (at the time) disks were selling as fast as they could be made, and any FBA support would result in customers switching from CKD to same amount of FBA.

Elsewhere, there is the comment about company makes much more on mainframe sevices, software, and storage than on direct computer ... in effect $6.25 total for every dollar in direct computer sales ... for $28M z196 that comes to $175M large cloud operations

Man Busts Out of Google, Rebuilds Top-Secret Query Machine
http://www.wired.com/wiredenterprise/2012/10/cloudera-impala-hadoop/

Papers describing the Google File System and Google MapReduce spawned Hadoop, an open source platform that lets you spread data across thousands of dirt-cheap computer servers and then crunch it into something useful. Google BigTable gave rise to an army of "NoSQL" databases that can juggle unusually large amounts of information. Google Pregel delivered multiple "graph" databases that can map the many online relationships between people and things.

... snip ...

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Sun, 21 Oct 2012 10:45:31 -0400
Anne & Lynn Wheeler <lynn@garlic.com> writes:
we then decide to leave. now two of the other people in the meeting from ellison's conference room ... also leave and show up at small client/server startup responsible for something called commerce server. We get brought in as consultants because they want to do payment transactions on the server, the startup had also invented this technology called SLL they want to use; the result is frequently now called electronic commerce. We have to map SSL to the payment transaction business process as well as walk through of most of the vendors selling digital certificates. I have final authority on all details on webserver to payment gateway and did various requirements about the browser to webserver interface ... but not final authority. Several of the browser/webserver security requirements were almost immediately violated.

re:
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other

because of the way SSL is deployed and used with browsers ... we early on starting referring to the SSL digital certificates as comfort certificates ... giving people the comfort of security ... as opposed to real security ... some past posts
https://www.garlic.com/~lynn/subpubkey.html#sslcert

more recent: Planet SSL: mostly harmless:
http://financialcryptography.com/mt/archives/001391.html

from above:
So where do we rate SSL? Mostly harmless, a tax on the Internet world, or a barrier to a better system?

... snip ...

note that various business interests in the existing SSL status quo have help resist alternative more secure paradigms.

and if you visit the above ... might notice that the previous entry is: "It's now official - Central Banks are not working to save the economy"
http://financialcryptography.com/mt/archives/001390.html

--
virtualization experience starting Jan1968, online at home since Mar1970

Mainframes are still the best platform for high volume transaction processing

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 21 Oct 2012
Subject: Mainframes are still the best platform for high volume transaction processing.
Blog: Mainframe Experts
re:
http://lnkd.in/RsuUyw

max configured zEC12 briefs have its DBMS throughput at 30% more than max configured z196 ... even though the processing rate is 50% more than z196. Part of this can be interpreted as the I/O throughput hasn't changed significantly from z196 (DBMS throughput increase primarily from additional processing but possibly bottlenecked by I/O).

The original FICON implementation severely restricted I/O throughput of the underlying fibre-channel standard (going back to 1988). The zHPF and TCW descriptions has it partially restoring to FICON some of the underlying fibre-channel capability and throughput .... however the documentation so far says that it is the same on zEC12 as z196.

max configured z196 has 80 processors rated at 50BIPS (and goes for $28M). z196 peak benchmarks have 2m IOPS using 104 FICON channels, 14 storage subsystems and 14 system assist processors. However, SAP reference has peak SSCH capacity of 2.2M/sec with all 14 SAPs running at 100% utilization ... but recommendation is that SAPs are kept at 70% or lower utilization (i.e. peak 1.5M SSCH/sec).

A recent announcement has single fibre-channel for E5-2600 blade capable of over million IOPS (aka two E5-2600 blade fibre channel having more I/O thruput than 104 FICON channels). E5-2600 blade have 527BIPS rating and IBM base list price of $1815. An IBM blade chassis holds something like 12 blades ... and possible to get something like 3 or 4 chassis in single rack ... possibly 48 blades in single rack with aggregate processing rate of 25,296BIPS (and around $90K ... compared to 50BIPS and $28M for z196)

recent posts mentioning zHPF, SSCH, etc:
https://www.garlic.com/~lynn/2012d.html#75 megabytes per second
https://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#5 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#11 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#13 Intel Confirms Decline of Server Giants HP, Dell, and IBM
https://www.garlic.com/~lynn/2012m.html#28 I.B.M. Mainframe Evolves to Serve the Digital World
https://www.garlic.com/~lynn/2012m.html#43 Blades versus z was Re: Turn Off Another Light - Univ. of Tennessee
https://www.garlic.com/~lynn/2012m.html#67 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012n.html#9 How do you feel about the fact that today India has more IBM employees than any of the other countries in the world including the USA.?
https://www.garlic.com/~lynn/2012n.html#13 System/360--50 years--the future?
https://www.garlic.com/~lynn/2012n.html#19 How to get a tape's DSCB
https://www.garlic.com/~lynn/2012n.html#44 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#48 Under what circumstances would it be a mistake to migrate applications/workload off the mainframe?
https://www.garlic.com/~lynn/2012n.html#51 history of Programming language and CPU in relation to each

--
virtualization experience starting Jan1968, online at home since Mar1970

Fred Brooks on OS/360 "JCL is the worst language ever"

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Fred Brooks on OS/360 "JCL is the worst language ever"
Newsgroups: alt.folklore.computers
Date: Sun, 21 Oct 2012 12:31:01 -0400
"Charles Richmond" <numerist@aquaporin4.com> writes:
Another reason for JCL was that you could allocate a cylinder for your file. That was supposed to speed disk access by reducing the time used in track seeks. ISTM that disks used to spin at 1800 or 3600 rpm... instead of the 7200 rpm of 3 1/2 " disks, or 5400 rpm for 2 1/2" disks. Faster rotational speeds are available, but the two speeds I mentioned are the most common for Winchester disks.

another reason for JCL (and DCBs and other stuff) was batch processing paradigm ... "job" was submitted for execution and it might run hrs or days (in some cases weeks) later ... when the responsible/knowledgeable person wasn't around.

lots of interactive stuff has responsible person sitting there and any glitches can be handle in real time. in batch processing ... detailed JCL specification help increase the probability that application runs to completion ... for instance, even before starts (hopefully) all the resource requirements are specified in detail.

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each
Newsgroups: alt.folklore.computers
Date: Sun, 21 Oct 2012 16:01:35 -0400
Morten Reistad <first@last.name> writes:
Just curious; how much changed in lightweight transactions since cics 1.7 (ca 1985) ?

cics history (gone 404, but lives on at wayback machine):
https://web.archive.org/web/20080123061613/http://www.yelavich.com/history/toc.htm

web support 2003
https://web.archive.org/web/20080123085741/http://www.yelavich.com/history/ev200303.htm

multiprocessor support 2004
https://web.archive.org/web/20090107054344/http://www.yelavich.com/history/ev200402.htm

i.e. cics was setup as great big multithreaded process under os/360 ... but only ran on single processor. in the late 90s, I knew of large mainframe operations running over 100 copies of CICS ... in part to get multiprocessor utilization.

linkedin mainframe thread: "Mainframes are still the best platform for high volume transaction processing":
http://lnkd.in/RsuUyw

my post in thread:
https://www.garlic.com/~lynn/2012n.html#72

other past posts mentioning CICS &/or BDAM
https://www.garlic.com/~lynn/submain.html#cics

univ. library in the 60s got an ONR grant to do online catalog ... some of the money went for 2321 datacell. the effort was also selected to be betatest for the original cics "product" ... and I got tasked with supporting/debugging it at the univ.

past posts in this thread:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#47 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#49 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#50 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#51 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#52 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#53 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#59 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#62 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#63 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#71 history of Programming language and CPU in relation to each other

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Sun, 21 Oct 2012 16:39:25 -0400
Uncle Steve <stevet810@gmail.com> writes:
Did you arrive at a standard specification that manages the endpoint vulnerability typical of encryption schemes? That is without the requirement of a trusted Big Brother?

re:
https://www.garlic.com/~lynn/2012n.html#63 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other

in the late 90s, I would makes offhand comments at security presentations that I would take $500 milspec piece, aggresively cost reduce by 2-3 orders of magnitude while making it more secure.

it was a generalized authentication trust mechanisms across all environments. there was issue if all environments went to chip-based authentication (aka something you have, in lieu of passwords and other mechanisms) ... people would need unique chip/card for every current pin/password they currently carry (large scores if not hundreds).

so the next step was changing from institutional-centric paradigm (one per institution per person) to person-centric paradigm (one per person) ... achieving 2-3 order of magnitude reduction per chip and 2-3 order of magnitude reduction in number of chips (overall 4-6 order of magnitude).

the issue then becomes if I present a person chipcard to my financial institution, government agencies, corporate door entry, etc ... what is required for all these institutions to trust the person-presented chipcard.

bunch of patents ... all assigned (we have no interest) including what is necessary to do in order to get large institutions to trust a person presented authentication device (i.e. move from institutional-centric trust model to person-centric trust model)
https://www.garlic.com/~lynn/aadssummary.htm

the patents were interesting ... we were working with boutique patent firm and we had over 50 patents and they said there would be well over 100 before we were done. the executives looked at cost of filing so many patents and directed all claims be repackaged into nine patents. Later the patent office came back and said that they were tired of these humongous patents where the filing fee didn't even cover the cost of reading the claims ... and the nine patents had to be repackaged in at least 30 patents.

lots of past posts on the general 3-factor authentication model
https://www.garlic.com/~lynn/subintegrity.html#3factor

as previous mentioned the current payment paradigm requires hiding information (encryption) as countermeasure to fraud ... and what was adopted for "electronic commerce" ... with SSL. In the x9.59 model, the only thing required is strong authentication ... and eliminates any requirement to hide information. I've periodically commented that the largest use of SSL in the world today is "electronic commerce" ... and the x9.59 standard eliminates the need to use SSL for encrypting/hiding payment transaction information. misc. past posts mentioning x9.59
https://www.garlic.com/~lynn/subpubkey.html#x959

even with strong authentication (eliminating skimming & breaches as vulnerabilitity ... using information from previous transactions for other fraudulent financial transactions ... aka can be considered form of replay attack) ... there still can be compromised endpoints that leverage the present of valid authentication to perform a fraudulent transaction (asking the person to approve the displayed transaction, but executing a totally different transaction).

In the late 90s, EU created the FINREAD standard as countermeasure to compromised endpoints being able to do transactions other than the ones the person thought they were approving. There was some missteps in the 2000 timeframe that resulting in industry pulling back from deploying a number of strong security technologies. misc. past posts mentioning EU FINREAD standard
https://www.garlic.com/~lynn/subintegrity.html#finread

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Sun, 21 Oct 2012 16:57:36 -0400
Uncle Steve <stevet810@gmail.com> writes:
Sounds like the typical "management" interference in the engineering design process.

re:
https://www.garlic.com/~lynn/2012n.html#63 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#75 history of Programming language and CPU in relation to each other

they think it is unreasonable that I have final authority sign-off on the webserver to payment gateway (servers sitting on the internet that also interface to the industry payment networks). misc. past posts mentioning payment gateway
https://www.garlic.com/~lynn/subnetwork.html#gateway

for browser/webserver; trivial example is I'm teaching business critical tcp/ip to a room of employees working on browser technology at the startup ... just about everybody in the audience are "paper" millionaires (i'm just being paid as consultant).

one of my requirements is multiple A-record support when the browser is making connection. people in the room say that is too advanaced & too complex. I offer to show them that is standard in client code ... from nearly any client in the bsd 4.3 tahoe & reno distirbutions. they still say it is too advanced/complex. Later I make comment that if the didn't see it in the steven's tcpip text book ... they can't do it.

it takes another year to get multiple A-record implementation in their browser.

--
virtualization experience starting Jan1968, online at home since Mar1970

U.S. banks on high alert against cyberattacks

Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 21 Oct 2012
Subject: U.S. banks on high alert against cyberattacks
Blog: Financial Crime Risk, Fraud and Security
U.S. banks on high alert against cyberattacks www.computerworld.com/s/article/9231515/U.S._banks_on_high_alert_against_cyberattacks

The Financial Services Information Sharing and Analysis Center has put U.S. banks on high alert against cyberattackers seeking to steal employee network login credentials to conduct extensive wire transfer fraud.

Cyberattack Warning Throws US Banks' Security Into High Gear
http://www.ecommercetimes.com/story/76211.html

US Banks Unable to Mitigate DDOS Attacks Despite Being Warned
http://news.softpedia.com/news/US-Banks-Unable-to-Mitigate-DDOS-Attacks-Despite-Being-Warned-Video-295313.shtml

Hackers May Have Had Help With Attacks on U.S. Banks, Researchers Say
http://bits.blogs.nytimes.com/2012/09/27/hackers-may-have-had-help-with-attacks-on-u-s-banks-researchers-say/

US banks hit by more than a week of cyberattacks
http://news.yahoo.com/us-banks-hit-more-week-cyberattacks-193125386--finance.html

a lot of fraudulent financial activity tries not to be visible .... crooks, ngos, countries ... doesn't matter. crashing the system stops the flow of money. visibility might possibly be immature and/or misdirection

Person that had tried unsuccessfully for a decade to get the SEC to do something about Madoff testified in the congressional hearings that tips turn up 13times more fraud than audits ... and that SEC didn't have a tip hotline, but had a 1-800 line for corporations to complain about audits. He was also asked in the hearings if new regulations were required. He said that while new regulations might be required, much more important was transparency and visibility.

One of the things is that for ages, something like 70% of identity theft related financial compromises (account fraud) has involved insiders. The rise of the internet and public press about external hacks have been utilized by insiders for misdirection and obfuscation .... creating ambiguity whether it is insider and/or outsider.

As an aside, annual USNI history conference last week was on cyber ... but very little was history. one of the panels discussions had the head of FS-ISAC. I've pontificated in the USNI blog about the lack of historical related cyber at the conference

--
virtualization experience starting Jan1968, online at home since Mar1970

Beyond the 10,000 Hour Rule

Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 21 Oct 2012
Subject: Beyond the 10,000 Hour Rule
Blog: Boyd Strategy
re:
http://lnkd.in/fB_SZU

Beyond the 10,000 Hour Rule
http://calnewport.com/blog/2010/08/09/beyond-the-10000-hour-rule-richard-hamming-and-the-messy-art-of-becoming-great/

Since I've been making the flatlanders reference for a couple decades ... I can't object to the 2-D reference. Ambiguity can be lack of information as well as conflicting information and/or misdirection. Fingerspitzengefuhl & Coup d'oeil can also be viewed as filling in the whole from only trivial amount of information.

past posts mentioning fingerspitzengefuhl and coup d'oeil
https://www.garlic.com/~lynn/2012.html#30 Age of Greed: The Triumph of Finance and the Decline of America, 1970 to the Present
https://www.garlic.com/~lynn/2012.html#89 Where are all the old tech workers?
https://www.garlic.com/~lynn/2012.html#95 Can anyone offer some insight
https://www.garlic.com/~lynn/2012b.html#19 "Buffett Tax" and truth in numbers
https://www.garlic.com/~lynn/2012b.html#26 Strategy subsumes culture
https://www.garlic.com/~lynn/2012b.html#59 Original Thinking Is Hard, Where Good Ideas Come From
https://www.garlic.com/~lynn/2012b.html#84 A Conversation with Peter Thiel
https://www.garlic.com/~lynn/2012f.html#52 Does the Experiencing Self "Out-OODA" the Remembering Self?
https://www.garlic.com/~lynn/2012f.html#77 Vampire Squid
https://www.garlic.com/~lynn/2012g.html#9 JPM LOSES $2 BILLION USD!
https://www.garlic.com/~lynn/2012g.html#16 Psychology Of Fraud: Why Good People Do Bad Things
https://www.garlic.com/~lynn/2012g.html#82 How do you feel about the fact that today India has more IBM employees than US?
https://www.garlic.com/~lynn/2012g.html#87 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#45 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012h.html#48 Difference between fingerspitzengefuhl and Coup d'oeil?
https://www.garlic.com/~lynn/2012h.html#79 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012i.html#92 Naked emperors, holy cows and Libor
https://www.garlic.com/~lynn/2012j.html#63 Difference between fingerspitzengefuhl and Coup d'oeil?
https://www.garlic.com/~lynn/2012j.html#77 Monopoly/ Cartons of Punch Cards
https://www.garlic.com/~lynn/2012k.html#31 History--punched card transmission over telegraph lines
https://www.garlic.com/~lynn/2012l.html#91 Difference between fingerspitzengefuhl and Coup d'oeil?
https://www.garlic.com/~lynn/2012m.html#64 Guest Post: Beakley on Boyd, Aerial Combat and the OODA-Loop

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Sun, 21 Oct 2012 19:57:57 -0400
Uncle Steve <stevet810@gmail.com> writes:
Interesting stuff, but you're not considering the whole picture. Government and corporations cannot be considered to have automatic trust, which is currently the basis for the trickle-down trust models extant in the financial sector, and indeed in the business world in general. Information security is hard.

re:
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#75 history of Programming language and CPU in relation to each other

one of the issues for financial sector & payment transaction is the current paradigm attack surface is so enormous ... extremely target rich environment. the standards work would cut the attack surface back from tens of millions to tens of thousands

another random metaphor i've used for payment transactions is going out of the airlock in open space w/o space suit. Then there is all the stuff about needed armored encryption ... because the world is very hostile environment and the current paradigm payment transaction has none of its own protection.

posts and references to the naken transaction metaphor at financial cryptography blog
https://www.garlic.com/~lynn/subintegrity.html#payments

slightly related is obfuscation and misdirection regarding insiders versus outsiders ... in (linkedin) financial crime risk, fraud and security group
https://www.garlic.com/~lynn/2012n.html#77 U.S. banks on high alert against cyberattacks

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Mon, 22 Oct 2012 18:34:31 -0400
jmfbahciv <See.above@aol.com> writes:
All that does is spread the cost of the fraud to every customers of the card's company.

there are various articles about how fraud is profit for the financial institutions.

electronic payments have an "interchange fee" ... the merchant submits the transaction amount and gets back that amount minus the "interchange fee". for decades that fee has been prorated roughly proportional to the amount of fraud for the type of transactions ... with something like an order of magnitude difference between the lowest and highest fraud interchange fee.

US financial institutions have avg. nearly 40% of their bottom line from these fees. eliminating most current fraud and reducing interchange fee charged by order of magnitude could result in around 1/3rd cut to their bottom line.

this extends into secondary analysis. the current electronic payment paradigm is the low hanging fruit for crooks and is source of significant profit from interchange fees charged merchants. significantly changing the electronic payment security paradigm not only eliminates that significant profit ... it will also drive crooks to other areas for fraud (places where the financial institutions can't justify charging merchants a significant profit and need to shoulder all the burden)

past posts in this thread:
https://www.garlic.com/~lynn/2012n.html#40 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#42 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#43 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#46 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#47 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#49 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#50 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#51 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#52 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#53 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#59 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#62 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#63 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#71 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#74 history of Programming language and CPU in relation to each
https://www.garlic.com/~lynn/2012n.html#75 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#76 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#79 history of Programming language and CPU in relation to each other

--
virtualization experience starting Jan1968, online at home since Mar1970

history of Programming language and CPU in relation to each other

From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: history of Programming language and CPU in relation to each other
Newsgroups: alt.folklore.computers
Date: Mon, 22 Oct 2012 21:07:21 -0400
Uncle Steve <stevet810@gmail.com> writes:
Aparently the Chaum patents do everything the right way except for assuring the identity of the people involved in the transactions. I suppose it is no great surprise that they have not resulted in a digital cash system for the masses. I wish I had the time to do proper research on this subject, but there are already too many demands on my time.

re:
https://www.garlic.com/~lynn/2012n.html#63 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#67 history of Programming language and CPU in relation to each other
https://www.garlic.com/~lynn/2012n.html#75 history of Programming language and CPU in relation to each other

they were part of digicash ... when digicash was being liquidated, i was brought in do the evaulation of the patent portfolio. there was deployment scaling & integrity issues, others of the patents were related to tricks for minimum no. of write operations for a transaction ... important when chipcard technology had less than 10k lifetime write operations.
https://en.wikipedia.org/wiki/DigiCash

--
virtualization experience starting Jan1968, online at home since Mar1970

Beyond the 10,000 Hour Rule

From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 22 Oct 2012
Subject: Beyond the 10,000 Hour Rule
Blog: Boyd Strategy
re:
http://lnkd.in/fB_SZU
and
https://www.garlic.com/~lynn/2012n.html#78 Beyond the 10,000 Hour Rule

regarding 2d & higher dimensions ... orientate can also be organize ... taking the data/information/knowledge/wisdom hierarchy ... orient/organize data to get information, orient/organize information to get knowledge and orient/organize knowledge to get wisdom/understanding. Fingerspitzengefuhl/Coup d'oeil is genius that can go quickly from data to wisdom/understanding

past posts referencing data/informationk/knowledge/wisdom
https://www.garlic.com/~lynn/95.html#8aa 2nd wave?
https://www.garlic.com/~lynn/2006w.html#47 'Innovation' and other crimes
https://www.garlic.com/~lynn/2008d.html#38 outsourcing moving up value chain
https://www.garlic.com/~lynn/2008i.html#29 What is your definition of "Information"?
https://www.garlic.com/~lynn/2003h.html#51 employee motivation & executive compensation
https://www.garlic.com/~lynn/2009.html#54 Business Science
https://www.garlic.com/~lynn/2010o.html#70 Compressing the OODA-Loop - Removing the D (and maybe even an O)
https://www.garlic.com/~lynn/2012f.html#72 New Study Debunks Idea That Human Performance Fits on A Bell Curve
https://www.garlic.com/~lynn/2012f.html#96 Hard Disk Drive Construction
https://www.garlic.com/~lynn/2012k.html#35 Has anyone looked at ORIENTATION using the metaphor of 'An Adaptive Toolbox'?
https://www.garlic.com/~lynn/2012k.html#36 Race Against the Machine
https://www.garlic.com/~lynn/2012k.html#42 The IBM "Open Door" policy

--
virtualization experience starting Jan1968, online at home since Mar1970

Protected: R.I.P. Containment

Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Date: 23 Oct 2012
Subject: Protected: R.I.P. Containment
Blog: Boyd Strategy
re:
http://lnkd.in/a-HJgb

Protected: R.I.P. Containment
http://thediplomat.com/2012/09/r-i-p-containment/

and/or other reasons for rewriting history

small war manual about tactics a century ago:
https://en.wikipedia.org/wiki/Small_Wars_Manual
however written about the same time about the same experience makes it sound like objective was to support most corrupt as part of plundering the country
https://en.wikipedia.org/wiki/War_Is_a_Racket

"Chiang Kai-Shek: China's Generalissimo and the Nation He Lost" makes the period sound like skirmishes between warlord factions ... and we supported a loosing faction
https://www.amazon.com/Chiang-Kai-Shek-Chinas-Generalissimo-ebook/dp/B00AALBQ5C/

more variations on supporting the most corrupt as part of plundering a country:

Collapse: How Societies Choose to Fail or Succeed
https://www.amazon.com/Collapse-Societies-Succeed-Revised-ebook/dp/B004H0M8EA
Why Nations Fail: The Origins of Power, Prosperity, and Poverty
https://www.amazon.com/Why-Nations-Fail-Prosperity-ebook/dp/B0058Z4NR8
Confessions of an Economic Hit Man
https://www.amazon.com/Confessions-Economic-Hit-Man-ebook/dp/B001AFF266

disclaimer: after end of hostilities in Germany, my wife's father was posted as adviser to the Generalissimo and allowed to take his family with him to Nanking.

Interpretation that 1st Iraq conflict wasn't the invasion of Kuwait but when they started marshaling forces along Saudi border:

Long Strange Journey
https://www.amazon.com/Long-Strange-Journey-Intelligence-ebook/dp/B004NNV5H2

and 2nd Iraq conflict

EXTREME PREJUDICE-- The Terrifying Story of the Patriot Act and the Cover Ups of 9/11 and Iraq
https://www.amazon.com/EXTREME-PREJUDICE-Terrifying-Patriot-ebook/dp/B004HYHBK2

and MICC supporting, if for nothing else, perpetual war as part of keeping the funds flowing
http://chuckspinney.blogspot.com/p/domestic-roots-of-perpetual-war.html

--
virtualization experience starting Jan1968, online at home since Mar1970

OT: Tax breaks to Oracle debated

Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: OT:  Tax breaks to Oracle debated
Newsgroups: alt.folklore.computers
Date: Wed, 24 Oct 2012 07:39:28 -0400
Dave Garland <dave.garland@wizinfo.com> writes:
Unsurprising. Corporations have become accustomed to extorting money (in the form of tax breaks) from communities.

Speaking from a location that has gotten suckered into these deals before, I say it's important to have explicit and firm commitments as to how many jobs will be created, how long they will last, the minimum that they will pay, and who they will hire, firmly in advance. And penalties that will have senior standing in case of default, bankruptcy, or sale of the parent company. These days, nobody but a fool would trust corporate entities to actually deliver what the promise, without having their vital body parts in a vise beforehand. Even then, it's tricky.


also con'ing municipalities to issuing bonds for various parts of the endeavors

--
virtualization experience starting Jan1968, online at home since Mar1970


previous, next, index - home