List of Archived Posts
2003 Newsgroup Postings (2/2 - 2/16)
- Wanted: Weird Programming Language
- Wanted: Weird Programming Language
- Network separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- what is the difference between ALU & FPU
- Network separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- Networks separation using host w/multiple network interfaces
- Networks separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- Unused address bits
- diffence between itanium and alpha
- diffence between itanium and alpha
- Early attempts at console humor?
- diffence between itanium and alpha
- Early attempts at console humor?
- diffence between itanium and alpha
- diffence between itanium and alpha
- Network separation using host w/multiple network interfaces
- diffence between itanium and alpha
- diffence between itanium and alpha
- Network separation using host w/multiple network interfaces
- Network separation using host w/multiple network interfaces
- Crypto Standards Organizations
- diffence between itanium and alpha
- diffence between itanium and alpha
- Will Apple ever offer a newsreader?
- diffence between itanium and alpha
- diffence between itanium and alpha
- Early attempts at console humor?
- diffence between itanium and alpha
- diffence between itanium and alpha
- diffence between itanium and alpha
- "average" DASD Blocksize
- "average" DASD Blocksize
- "average" DASD Blocksize
- DOD 5200.28-STD capable OS?
- "average" DASD Blocksize
- How much overhead is "running another MVS LPAR" ?
- diffence between itanium and alpha
- Early attempts at console humor?
- Easiest possible PASV experiment
- Early attempts at console humor?
- diffence between itanium and alpha
- diffence between itanium and alpha
- "average" DASD Blocksize
- Filesystems
- Filesystems
- HASP assembly: What the heck is an MVT ABEND 422?
- diffence between itanium and alpha
- HASP assembly: What the heck is an MVT ABEND 422?
- "average" DASD Blocksize
- HASP assembly: What the heck is an MVT ABEND 422?
- Easiest possible PASV experiment
- Easiest possible PASV experiment
- "average" DASD Blocksize
- When/why did "programming" become "software development?"
- "average" DASD Blocksize
- RFC 3092
- Re : OT: One for the historians - 360/91
- Re : OT: One for the historians - 360/91
- Surprising discovery
- Dijkstra on "The End of Computing Science"
- FBA suggestion was Re: "average" DASD Blocksize
- Tubes in IBM 1620?
- Eliza
- OT: One for the historians - 360/91
- COMTEN- IBM networking boxes
- Tubes in IBM 1620?
- OT: One for the historians - 360/91
- iso 8583
- OT: One for the historians - 360/91
- The relational model and relational algebra - why did SQL become the industry standard?
- COMTEN- IBM networking boxes
- COMTEN- IBM networking boxes
- The relational model and relational algebra - why did SQL become the inudstry standard?
- COMTEN- IBM networking boxes
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Wanted: Weird Programming Language
Newsgroups: alt.folklore.computers
Date: Sun, 02 Feb 2003 23:06:13 GMT
space_cadet@gmx.de (Sebastian) writes:
Forgive my ignorance - I'm still young - but who is Jean Sammet?
somebody that worked on 3rd floor 545 tech sq for something called the
boston programming center ... until it was absorbed by the break-away
vm development group from the 4th floor cambridge science center. when
the vm development outgrew the 3rd floor and moved out to the old sbc
building in burlington mall (as part of some legal thing another sbc
being sold to cdc) ... nat rochester and a couple others were absorved
into the science center.
http://www.garlic.com/~lynn/subtopic.html#545tech
sometimes when i would come in on weekends, I would bring the kids in
and they could do stuff like play spacewar on the 1130/2250 ... but
they also had a tendency to chase each other up and down the
hall. when that happened, jean would come around to complain.
some past jean sammet refs:
http://www.garlic.com/~lynn/2000d.html#37 S/360 development burnout?
http://www.garlic.com/~lynn/2000f.html#66 360 Architecture, Multics, ... was (Re: X86 ultimate CISC? No.)
http://www.garlic.com/~lynn/2001m.html#47 TSS/360
http://www.garlic.com/~lynn/2002h.html#59 history of CMS
http://www.garlic.com/~lynn/2002j.html#17 CDC6600 - just how powerful a machine was it?
http://www.garlic.com/~lynn/2002o.html#76 (old) list of (old) books
http://www.garlic.com/~lynn/2002o.html#78 Newsgroup cliques?
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Wanted: Weird Programming Language
Newsgroups: alt.folklore.computers
Date: Sun, 02 Feb 2003 23:51:43 GMT
space_cadet@gmx.de (Sebastian) writes:
Forgive my ignorance - I'm still young - but who is Jean Sammet?
or try a search engine on jean sammet ... a few
http://perpugilliam.csclub.uwaterloo.ca/office/books/68028110__.html
http://www.acm.org/awards/fellows_citations_n-z/sammet.html
http://www.awc-hq.org/lovelace/1989.htm
http://laika.ed.csuohio.edu/fall98/edb567/Women_in_computers/JeanESammet.htm
http://www.mtholyoke.edu/offices/comm/csj/022500/techconference.html
http://tap.mills.edu/pioneers/
http://www.msci.memphis.edu/~hernandg/comp1900/inflangs.html
http://www.heuse.com/programming-c.htm
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Mon, 03 Feb 2003 13:00:11 GMT
Emil Naepflein writes:
Turn ipforwarding in your host off. Most OS have a way to control this.
ipforwarding off was somewhat the result of Interop '88 sunday night
meltdown, crash & burn ... with most hosts being connecting to two or
more of the four floor nets ... and everybody forwarding everything.
ref:
http://www.garlic.com/~lynn/rfcstd.htm#ietfstd
ipforwarding off default in rfc1122/std3 (pg 28 & 29):
Any host that forwards datagrams generated by another host is acting
as a gateway and MUST also meet the specifications laid out in the
gateway requirements RFC [INTRO:2]. An Internet host that includes
embedded gateway code MUST have a configuration switch to disable the
gateway function, and this switch MUST default to the non-gateway
mode. In this mode, a datagram arriving through one interface will
not be forwarded to another host or gateway (unless it is
source-routed), regardless of whether the host is single-homed or
multihomed. The host software MUST NOT automatically move into
gateway mode if the host has more than one interface, as the operator
of the machine may neither want to provide that service nor be
competent to do so.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Mon, 03 Feb 2003 22:37:40 GMT
vjs@calcite.rhyolite.com (Vernon Schryver) writes:
That is mistaken. There were many other instances of problems from
misconfigured routers before and after. Contrary to the sales brochures
from Dan Lynch &co, the InterOp trade shows were never very influental
in technical circles (as opposed to marketing) and did not do much
real testing. Most of the testing was of political gamesmanship,
such as fighting over which vendor should be thrown off which demo
net for which real or pretend sin. (I participated in more than one
or two InterOp "interoperability tests" and "technology demos," I
think including one at InterOp '88.) The first InterOp in Monterey
and perhaps the second were different and not mere network trade shows.
it wasn't misconfigured routers .... it was (nearly?) every
client/host (not the routers) that were constantly forwarding. the
point wasn't to do interoperability testing. the point was getting the
show up and running at all. it was crashing and burning until
something like 3-4 am monday morning ... when they started going
around and turning off ip-forwarding. interop '88 was the second
interop .... and it was held in san jose. there were four floor nets
inconnected with routers by a couple vendors. the four floor nets were
crashing and burning ... until they started going around and turning
off ip forwarding in all the booths.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Mon, 03 Feb 2003 23:00:53 GMT
vjs@calcite.rhyolite.com (Vernon Schryver) writes:
Moreover, my vague recollections of the problems that night was
that they were not simply (or rather falsely) due to hosts that
were acting as routers, but hosts acting as routers with the wrong
net mask and a very serious bug involving confusing MAC and IP
broadcasts as IP unicasts.
misconfiguring was possibly what gave rise to the last couple words of
the sentence that i quoted from 1122 ... aka
The host software MUST NOT automatically move into gateway mode if the
host has more than one interface, as the operator of the machine may
neither want to provide that service nor be competent to do so.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Mon, 03 Feb 2003 22:56:23 GMT
i believe the comment was that it was possibly the first time that a
large number of workstations/pcs had been simulataneously connected on
two or more independent lans. there were possibly 100(?) booths
... and each booth possibly had two or more workstations of some sort
... and large percentage of these workstations/pcs were connected to
two or more different nets. the forwarding problem wasn't at the
routers ... the forwarding problem was in all the workstations and
pcs. the forwarding would start a packet storm would build up and the
nets die. this started sometime sunday afternoon and lasted to the wee
hours of monday am. some people were getting concerned about whether
the problem would be found and identified before the show opened
monday morning.
other random stuff from the era ... long ago and far away
http://www.garlic.com/~lynn/2002k.html#12 nsfnet rfp announce
http://www.garlic.com/~lynn/2000e.html#10 nsfnet award
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Mon, 03 Feb 2003 23:26:47 GMT
other tidbits from that show ....
there was a number of OSI, x.400, etc stuff in various booths.
snmp still hadn't won ... and was even looked at askance by some
number of people.
case was officially in the booth kiddie corner from where I had a
couple of workstations ... one with a megapel display. got case to
come over and install snmp ... so could demo it on megapel display.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: what is the difference between ALU & FPU
Newsgroups: comp.arch
Date: Tue, 04 Feb 2003 01:57:14 GMT
"del cecchi" writes:
I worked with some guys that had been part of IBM's first imaginary
processor (no, not the 360/91) FS. Later I got to work on a whole
imaginary system with an imaginary operating system. Fort Knox w/Bach
and Beethoven
So there.
del cecchi
but one could claim that the current as/400 is fort knox re-incarnated
and s/38 was FS re-incarnated.
so you might be able to claim that you got to work on both the
imaginary versions and the real(?) versions.
random fs comments:
http://www.garlic.com/~lynn/subtopic.html#futuresys
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Tue, 04 Feb 2003 07:48:53 GMT
basically something acting as a gateway (indicated by ipforwarding on)
will answer to ARPs for any ip addresses on one network interface that
it has another network interface for.
i.e. a gateway has network interfaces a: 10.1.1.1/255.0.0.0
b: 168.1.1.1/255.255.0.0
any ARP request arriving on interface "b" for anything on 10-net, the
host will respond to with interface-b MAC address. then when it starts
to get incoming packets (on interface "b") for any address on 10-net
(with its interface-b MAC address), it will ARP on the 10-net
interface for those ip-addresses ... attempting to act as a gateway to
forward the packets.
similarly any ARP request arriving on interface "a" for anything on
the 168.1-net, the host will respond to with the interface-a MAC
address. when it starts to get incoming packets (on interface "a") for
any address on 168.1-net (with its interface-a MAC address), it will
ARP on the 168.1-net interface for those ip-address ... attempting to
act as a gateway to forward the packets.
if the gateway has ip-forwarding on and default route 0.0.0.0 to the
168.1.1.1 interface and any ARP request comes in on interface-a for
either its 10.1.1.1 ip-address or any non-10-net address (not just
168.1-net addresses), it will respond with its interface-a MAC
address.
it would be a malformed environment if the only way a multihomed host
can receive packets for one of its ip-addresses is to turn
ipforwarding one so that it will answer to ARP requests for that
ip-address on all of its other network interface(s).
so what are some of the possible faulty configurations and/or
implementations leading to a multihomed host having to turn on
ipforwarding.
One possible situation is
1) DNS is set-up with multiple A-records for each of the ip-addresses
for the host,
2) each of the multihomed network interfaces are on different physical
LANs,
3) there is no gateway installed that is configured for routing
packets between the different LANs, and
4) there are client applications that fail to implement multiple
A-record support
A client application w/o multiple A-record support that is configured
on one LAN/subnet and the first DNS A-record lists an ip-address on a
different LAN/subnet will attempt to ARP for the first ip-address and
not get a response. It will then give up w/o checking any of the other
A-record ip addresses.
There are actually two parts of gateway support:
1) answering to ARPs on a network interface for ip-addresses different
than what is assigned to that network interface
2) forwarding packets
Ideally, if there actually are lots of situations involving brain dead
client implementations (lacking multi-A-record support) in multihomed
server environments, then it would be preferrable to have a multihomed
option w/o having to turn on ipforwarding. Such a case would activate
a multihomed host to respond to all ARP requests for any of its
ip-addresses regardless of the interface they arrive on (w/o it
answering to ARP requests for ip-addresses not defined on the host,
also the ARP response would have to correspond to the network
interface that the ARP arrived on).
note such an option might be considered to violate network
architecture, but it is possibly preferrable (if there are a large
number of brain dead client applications in multihomed environments)
to turning on full ipforwarding.
draft from long ago and far away that never made it to RFC
COMMENTS SHOULD BE SENT TO: lekash@orville.nas.nasa.gov
Date: 26-Apr-88
Title: Multi-Homed Hosts in an IP Network
Authors: J. Lekashman (NASA Ames GE)
Host Behavior Working Group (retired) NASA Ames GE
IETF April 1988
Multi-Homed Hosts in an IP network
Status of This Memo
This is a request for commentary and ideas on how to handle multiple
interfaces in a networked environment. Several problems are raised,
and some solutions are presented. This is a draft. It is presented
in order to generate discussion and suggestions for improvement.
..snip..
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Tue, 04 Feb 2003 07:56:45 GMT
vjs@calcite.rhyolite.com (Vernon Schryver) writes:
I doubt that was true except in InterOp Inc.'s sales brochures.
The amazing parts about the InterOP networks were never their size
or complications. It was that each network was built from scratch
(less so in later years) within a week and that as far as those
who paid for tickets could tell, it worked.
i don't remember any interop sales brochures ... i just remember
sunday before interop '88 being very hectic and a lot of people on
monday ... lets say ... quite disgruntled ... about how well many of
the workstations/pc handled multiple network interfaces ... and just
because they had multiple network interfaces didn't mean that they
should be doing ip forwarding.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Networks separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Tue, 04 Feb 2003 08:30:18 GMT
vjs@calcite.rhyolite.com (Vernon Schryver) writes:
As I recall, the show net stablized before the network I was concerned
with.
i don't know if it makes any difference ... but there were four floor
nets for the interop '88 show ... and that the multihomed
workstation/pc configurations were causing lots of havoc sunday thru
early monday morning. i believe everybody putting up booths and
getting stuff working on the show floor were doing it sunday (and into
the wee hours of monday morning).
this was different than the interoperability testing ... which i
vaguely remember being in a 2nd floor room of the conference center
towards the back ... where there is a connection to the parking
garage.
i don't remember anybody starting to do booth/network setup on the
show floor that weren't there on sunday and were potentially part of
the network meltdowns. however the interoperability testing upstairs
went on all week.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Networks separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Tue, 04 Feb 2003 08:40:09 GMT
also ... one of the largest booths that year ... was the ibm booth,
done mostly by the person that obtained the 9-net for ibm. it was on
the other side of the show floor from the booth where i had some
workstations. i believe that year ibm also provided the network
connectivity between the show floor and the nsfnet backbone. which
then brings in the previous refs:
other random stuff from the era ... long ago and far away
http://www.garlic.com/~lynn/2002k.html#12 nsfnet rfp announce
http://www.garlic.com/~lynn/2000e.html#10 nsfnet award
interop '88 was almost a year after the above mentioned nsfnet
backbone award.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Tue, 04 Feb 2003 13:22:43 GMT
arielgont@softhome.net (Fernando Gont) writes:
Why wouldn't it send an ARP request for the IP address of the router
it has been configured to "use"?
the description wasn't sending an ARP request for the IP address of
the router it had been configured to use.
the scenario was DNS with multiple A-records for a multihomed
"server" and a client that didn't support multiple A-records.
http://www.garlic.com/~lynn/2003c.html#8 Network separation using host ..
An email was sent yesterday describing a situation where clients
weren't working unless a multihome host/server had ipforwarding turned
on ... and that it was convenient when multihome implementations
turned ipforwarding on automatically (in violation of rfc 1122).
A multihomed host/server has network interfaces on disjoint LANs,
there is no router configured interconnecting the disjoint LANs, there
are A-records for each of the multihomed host/server network
interfaces, there are clients on one or more of the disjoint network
interfaces running applications w/o multiple A-record support.
A client (w/o multiple A-record support) gets the DNS response and
does an ARP for the ip-address in the first A-record. If the ip
address is for the server's interface on a different LAN, the server
(w/o ipforwarding turned on) doesn't answer the ARP request. When
there is no answer to the ARP request, the client applications "punts"
... since it doesn't have support to try the ip-addresses in the
additional A-records.
By turning on ipforwarding in the multihomed server .... the
multihomed server will then act as a gateway and answer to ARP
requests for additional networks that it is connected to. In this
particular scenario, the multihome server will answer to an ARP
request for one of its own network interfaces on a different LAN. The
client gets back an ARP response and starts sending packets to the
"gateway" interface. However, the multihomed host/server doesn't
actually forward the packets since they are addressed to itself under
a different ip-address.
The scenario is one where (because of lack of multiple A-record
support in client applications) it was desirable to have multihomed
host/servers to answer to all ARP requests for any of its ip-addresses
regardless of the network interface the ARP request arrived on. There
was no real desire to have the multihomed host/server to actually act
as a router forwarding packets ... but it was necessary to turn on
ipforwarding just to get the multihome host/server network support to
do the multi-interface ARP response operation.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Unused address bits
Newsgroups: comp.arch
Date: Tue, 04 Feb 2003 13:35:10 GMT
nmm1@cus.cam.ac.uk (Nick Maclaren) writes:
There are actually dozens of successful changes, most recently with
the 32- to 64-bit Unix systems as done by SGI, Sun, IBM and HP. But
it was standard practice even earlier.
Where the IBM 370 range was unusual was that it supported mixed
addressing modes within the same 'executable' (though the concepts
were different). A lot of work was put in to make that transparent
to users!
and then if you really wanted to get complicated there was dual
address space mode ... that in part tried to address the issue of
running out of addressable space (pre 31-bit) and in part to address
the issue of allowing system services in disjoint address spaces to
(directly) access parameters passed by pointer (w/o requiring kernel
to move data). With XA-mode and 31-bit mode addressing, came access
registers and program call expanding the cross address space
addressability.
past dual address space, access regisgters, & program call posts:
http://www.garlic.com/~lynn/98.html#36 What is MVS/ESA?
http://www.garlic.com/~lynn/2000c.html#84 Is a VAX a mainframe?
http://www.garlic.com/~lynn/2000d.html#28 RS/6000 vs. System/390 architecture?
http://www.garlic.com/~lynn/2000e.html#58 Why not an IBM zSeries workstation?
http://www.garlic.com/~lynn/2001d.html#28 Very CISC Instuctions (Was: why the machine word size ...)
http://www.garlic.com/~lynn/2001h.html#73 Most complex instructions
http://www.garlic.com/~lynn/2001i.html#13 GETMAIN R/RU (was: An IEABRC Adventure)
http://www.garlic.com/~lynn/2001k.html#16 Minimalist design (was Re: Parity - why even or odd)
http://www.garlic.com/~lynn/2002d.html#51 Hardest Mistake in Comp Arch to Fix
http://www.garlic.com/~lynn/2002g.html#17 Black magic in POWER5
http://www.garlic.com/~lynn/2002g.html#18 Black magic in POWER5
http://www.garlic.com/~lynn/2002l.html#51 Handling variable page sizes?
http://www.garlic.com/~lynn/2002l.html#57 Handling variable page sizes?
http://www.garlic.com/~lynn/2002n.html#58 IBM S/370-168, 195, and 3033
http://www.garlic.com/~lynn/2002n.html#74 Everything you wanted to know about z900 from IBM
http://www.garlic.com/~lynn/2002q.html#1 Linux paging
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 05 Feb 2003 03:02:00 GMT
Robert Myers writes:
We agree that DEC made critical management errors when it was at the
height of its power in the eighties. We agree that avoiding those
errors would not have required super-human insight. I think we agree
that DEC needed a VAX replacement before it managed to get one on the
market and we agree that competitors were chopping away at DEC's
franchise with less expensive products. What is it that we disagree
about, except that I know that even the lowliest order taker has to
have some sales skills?
the mid-range was starting to be killed by workstations and the
growing power of PCs. any company with majority of the products in the
mid-range adapting to such a situation would become a totally
different company than what it had been (i.e. all the people start all
over again ... either under a different company name or under the
facade of the current company name).
past related discussions/threads:
http://www.garlic.com/~lynn/2001m.html#15 departmental servers
http://www.garlic.com/~lynn/2001n.html#23 Alpha vs. Itanic: facts vs. FUD
http://www.garlic.com/~lynn/2002.html#2 The demise of compaq
http://www.garlic.com/~lynn/2002.html#7 The demise of compaq
http://www.garlic.com/~lynn/2002d.html#4 IBM Mainframe at home
http://www.garlic.com/~lynn/2002j.html#34 ...killer PC's
http://www.garlic.com/~lynn/2002j.html#66 vm marketing (cross post)
including vax domestic and world wide shipments:
http://www.garlic.com/~lynn/2002f.html#0 Computers in Science Fiction
& refs:
http://www.garlic.com/~lynn/2002f.html#1 Blade architectures
http://www.garlic.com/~lynn/2002f.html#5 Blade architectures
http://www.garlic.com/~lynn/2002g.html#80 Pipelining in the past
http://www.garlic.com/~lynn/2002g.html#84 Questions on IBM Model 1630
http://www.garlic.com/~lynn/2002i.html#30 CDC6600 - just how powerful a machine was it?
http://www.garlic.com/~lynn/2002q.html#3 Vector display systems
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch
Date: Wed, 05 Feb 2003 03:13:34 GMT
Robert Myers writes:
'Twas obvious to me by 1985 what was coming, because at that point I
had a 32016 card plugged into the expansion bus of a PC running
programs that until then had been grinding away on a Cray. And, yes,
that's right, it means that DEC didn't close on their hot lead, and
neither did Prime or Perkin-Elmer. I was running Concurrent DOS and
would have just plugged in another card if I had needed more compute
power (although some projects still required a Cray). I moved on to
another job before I found such a step necessary, and that company
moved from a micro-VAX to an Alliant in the late eighties, there being
no high-end DEC product even to consider.
i've had certain foundness for perkin-elmer ... and before they bought
it interdata. i saw a perkin-elmer box operating as ibm mainframe
telecom controler in heavy production use as recently as 1997.
in the '60s as an undergraduate ... i was one of four people involved
in building a ibm telecommunication compatible controller starting
with interdata/3 ... and reverse engineering the ibm channel interface
... and getting some blame for originating the ibm pcm (plug
compatible manufactur) controller market. random refs:
http://www.garlic.com/~lynn/subtopic.html#360pcm
i didn't get to look underneath the covers of the most recent sighting
... but have been told by people that sold perkin-elmer boxes in the
80s ... that the channel board was still wire-wrap and possibly the
original design that we had done from the '60s.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Early attempts at console humor?
Newsgroups: alt.folklore.computers
Date: Wed, 05 Feb 2003 03:43:31 GMT
"GerardS" writes:
THe IBM system was VM/CMS. Under CMS (which ran under VM) would (as a
default on a 1050 or 2741 terminal) raise and lower the printing/typing
mechanism to signify that 2 CPU seconds (of processing time) had elapsed.
This was to give the user a type of warning that their program was
consuming CPU time. On other types of terminals, a "blip" message was
displayed (this could be modified/set by the CMS user). _______Gerard S.
originated on cp/67 .... and we (cambridge science center) ported
apl\360 to cp/67 cms ... and released it as cms/apl. palo alto science
center redid some of the stuff a couple years later .. they also did
the the apl m'code assist for the 370/145 and released as apl/cms
(many apl applications on 145 w/apl-mcode assist had thruput
compareable to running on 370/168-3 w/o assist)
earlier post in this thread:
http://www.garlic.com/~lynn/2003b.html#71 Early attempts at console humor
http://www.garlic.com/~lynn/2003b.html#72 Early attempts at console humor
lots of apl posts (intermixed with hone posts):
http://www.garlic.com/~lynn/subtopic.html#hone
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 05 Feb 2003 12:10:06 GMT
"del cecchi" writes:
I hate to dispute your theory, but the fact is that business
oriented machines in the midrange did quite well for IBM well into
the era in question, and even today, even with marketing that is no
better than DEC's was. Remember the "magic box" commercials? Or as
we called them "tragic box"?
The Rochester boxes of the era, the S/36 and S/38 which morphed into
AS400 and then into i-series made lots of money for IBM, and still do.
And do it without much marketing. Don't confuse the fact that no one
who had an alternative would buy a Low End 370 like 4321 or racetrack
with the idea that there was no midrange market.
i agree that the business market that the rochester products sold into
were much more resistent to the workstation & pc onslaught than many
of the other mid-range & mini markets. however i believe that the
sales/volume explosion of 4341s (370s) and vaxs (that occured in the
late 70s and early 80s) had significantly slowed by the mid-80s
... which the referenced world wide ship numbers for vaxs thru 1988
would tend to support:
http://www.garlic.com/~lynn/2002f.html#0 Computers in Science Fiction
in the late 70s and very early 80s there were a number of situations
were customers went in and ordered 4341s in the hundreds similar to
the usaf reference:
http://www.garlic.com/~lynn/2001m.html#15 departmental servers
one of the first was in the late '70s by a large chip firm that placed
a single order for 800 4341s in support of chip design process. inside
the corporation there was a similar phase .... STL went thru a period
where it converted one conference room on every floor in every tower
to a 4341 room. the marketing organization first got 4341s/148s in
every region ... which then trickled out to every large branch
installing its own machine.
by the mid-80s there was some expectations that those 4341s would be
upgraded to later models like the 4381s and it never really
materialized. starting in the mid-80s .... workload was instead
showing up on workstations and pcs ... and eventually the 4341s were
just discontinued ... rather than being upgraded. The vax numbers seem
to indicate possibly a similar fate in the market.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Early attempts at console humor?
Newsgroups: alt.folklore.computers
Date: Wed, 05 Feb 2003 12:19:15 GMT
"GerardS" writes:
THe IBM system was VM/CMS. Under CMS (which ran under VM) would (as a
default on a 1050 or 2741 terminal) raise and lower the printing/typing
mechanism to signify that 2 CPU seconds (of processing time) had elapsed.
This was to give the user a type of warning that their program was
consuming CPU time. On other types of terminals, a "blip" message was
displayed (this could be modified/set by the CMS user). _______Gerard S.
also note that by the time of apl\cms on vm/370 (instead of cms\apl on
cp/67) ... a large percentage of apl\cms would have been 3277 with
the apl characterset and keyboard ... instead of the cp "set apl on"
switching the terminal translate table to apl ... it would have change
to fullscreen 3270 apl operation. random apl translate table ref:
http://www.garlic.com/~lynn/2000f.html#6 History of ASCII (was Re: Why Not! Why not???)
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 05 Feb 2003 12:30:42 GMT
... aka not only didn't the sales explosion for the 4341/vax market
segment machines (from the late 70s & early 80s) not continue into the
mid & late 80s ... but even upgrade/replacement of existing installed
machines was being siphoned off into workstation & pc sales.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 05 Feb 2003 12:41:55 GMT
Robert Myers writes:
Ummm, as I recall, IBM, which had introduced the PC to the world,
was just as uncomprehending of its implications as was DEC. I may be
mistaken, but I think that IBM also suffered some reverses of fortune,
although they are still in business.
When I talked to an IBM salesman about the possibility of networking a
couple of PC's, he suggested that I really should buy a mini-computer.
End of discussion with IBM.
IBM was in a lot broader set of markets. as mentioned, the rochester
product line has been much more resistent to the workstation/pc
onslaught ... as has the higher end mainframes.
however, the mid-range 370 ... frequently selling head-to-head into
the same market as vax machines ... and saw even a larger explosion in
sales during the period ... went thru similar experience as vax.
as in the past ... i've claimed that the whole SAA strategy of the
late 80s and early 90s was something of an attempt to put the genie
back in the bottle. saa/genie refs:
http://www.garlic.com/~lynn/2002p.html#34 VSE (Was: Re: Refusal to change was Re: LE and COBOL)
http://www.garlic.com/~lynn/2002q.html#40 ibm time machine in new york times?
whole set of 3tier, middle layer, and saa posts:
http://www.garlic.com/~lynn/subnetwork.html#3tier
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Wed, 05 Feb 2003 13:36:00 GMT
Emil Naepflein writes:
I once had such a situation where someone connected a new cluster with a
local cluster interconnect and a global cluster interconnect the wrong
way and the consequence was that 40 high availability clusters stopped
communication because of an updated ARP entry.
yes ... my wife and i started looking at a lot of that in late '80s
when we started the ha/cmp project ... which eventually started
shipping products in the early '90s.
trivia from the period. another issue was ip-address take-over in a
high availability cluster. there was a bug in the reno/tahoe 4.3 code
that a large number of platforms were running. in theory, the ARP
cache has a time-out ... so that you would generally expect clients to
recognize a new MAC in a ip-address take-over scenario. However, the
4.3 code (instantiated all over the place) had a single hip-pocket MAC
in the ip code that called the ARP processing. If the IP-address was
the same as the ip-address in the last packet processed, it used a
saved MAC address (and didn't bother calling ARP routine). This value
never timed-out ... the only way it went away was when the client had
to communicate with a different IP-address. in a ha/cmp configuration
with a strongly oriented client/server activity ... this might
possibly be hours (or forever).
a work around was to have a sever (after an ip-address take-over in a
ha/cmp scenario) try and ping every client that it possibly knew
about.
random ha/cmp refs:
http://www.garlic.com/~lynn/95.html#13
http://www.garlic.com/~lynn/subtopic.html#hacmp
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 05 Feb 2003 13:39:48 GMT
Jan C. Vorbrüggen writes:
Just so. And at the time, one could even in academia still have made up
somewhat of a price difference because of DEC's much superior software,
both OS and networking. But the actual price - and, later, performance -
differential was so large that one immediately lost the argument. Not
- as in "not at all" - supporting TCP/IP was another important nail in
the coffin.
and possible some distraction with OSI ... some past decnet/osi threads:
http://www.garlic.com/~lynn/2001e.html#32 Blame it all on Microsoft
http://www.garlic.com/~lynn/2001e.html#34 Blame it all on Microsoft
http://www.garlic.com/~lynn/2001n.html#25 Unpacking my 15-year old office boxes generates memory refreshes
http://www.garlic.com/~lynn/2001n.html#27 Unpacking my 15-year old office boxes generates memory refreshes
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Wed, 05 Feb 2003 22:59:30 GMT
ref:
http://www.garlic.com/~lynn/2003c.html#17 diffence between itanium and alpha
http://www.garlic.com/~lynn/2003c.html#19 diffence between itanium and alpha
http://www.garlic.com/~lynn/2003c.html#20 diffence between itanium and alpha
http://www.garlic.com/~lynn/2003c.html#22 diffence between itanium and alpha
the other is that it seems that the most vulnerable of these
departmental mega-installs were online/terminal related environments
(aka the large population of installations were for supporting people
sitting at keyboards).
as the workstations, PCs, and networks evolved ... they frequently
started to address the requirements of people at keyboards better than
the host/terminal configurations (resulting in the bleeding off of the
applications from the host/terminal environment, leaving the hosts
somewhat stagnant and contributing to not seeing a lot of these
installations upgrading to next generation hosts ... aka
4341->4381s).
rochester wasn't totally immune. i believe that some of the early pc
business software (like peachtree) was competing in the same market as
many of the rochester s/3? whatever machines ... and started to make
some impact by the 386 timeframe. somewhat the difference was that it
was much easier for the interactive/online terminal use of shared host
bleed off into dedicated pc&workstation boxes .... and/or higher-end
workstations acting in place of shared interactive/online host. it was
somewhat longer cycle to see (frequently batch) business processes
show up on this class machines competing in major market segment
addressed by rochester products (and of course the current talk is of
really moving up the value chain into full fledged glass house data
processing).
then, of course, there is the whole dasd/disk farm issue.
PCs gained many of their early mega-sales in the business community by
offering a combination of host terminal emulation and local processing
(spreadsheet, games, etc) in a single keyboard format ... which was
both a desk-space issue as well as a cost issue (new PCs were
frequently priced less then the <original> price of aging 3270s
that they replaced).
In the middle '80s one of the high level technical people in the dasd
division gave a presentation at an internal world wide conference
making the claim that the executive in charge of the communication
divison was going to be responsible for the death of the dasd
division.
while terminal emulation provided a fast uptake market penetration for
PCs .... within a couple years it began to represent a bottleneck to
evolving mainframe host & their disk farms into vital role in the
emerging client/server (and later 3tier) environments. The bottleneck
of the terminal emulation paradigm contributed significantly to
wholesale migration of critical business data off mainframes to local
PC disk storage (and various horror stories about business continuity
issues with lack of backup).
The point of the dasd division claims were that numerous, innovative,
high-thruput, high-value SAN-like use of mainframes starting in the
mid-80s were torpedod by executives in charge of the communication
division (motivated by the adverse impact to the huge install base of
communication division hardware supporting the terminal emulation
paradigm).
In large respect, SAA was an attempt to restructure client/server
paradigm so that it inhibited major migration of business critical
data off of the mainframe disk farms ... w/o having to change the
terminal emulation paradigm (preserving the big install base of
communication division boxes, aka the communication division was
willing to sacrifice the whole corporate disk business as part of
minimizing any impact to the installed communication division boxes).
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Wed, 05 Feb 2003 22:49:32 GMT
arielgont@softhome.net (Fernando Gont) writes:
I don't understand why it was needed.
If multiple A records are returned, and all hosts listed in the A
records are up and you cannot get to the first one, regardless of the
"broken" client, you have a routing problem.
so i interpret your question as what kind of conjecture is there about
a installation having:
• two or more physical LANs
• clients distributed across the physical LANs
• each client has a unique ip-address
• single multihomed server with physical LAN adpaters for each of the
distinct physical LANs.
• each server LAN adapter has a unique ip-address
• the server provides one or more services for client use
• the clients all run the identical client software loaded with the
identical server host name
• the client software is crippled in that it lacks multi-A record
support
• because there is no need for inter-client communication and/or any
communication between the clients with anything other than the single
server; there is no need for router function across the multiple LANs
• the server provide DNS function for all LANs from a single DNS file
• the single DNS file defines the server with a single host name (that
is installed in all copies of the client software)
• the single DNS file defines a unique A-record with a unique
ip-address for each of the server's LAN interfaces
so why would somebody do such a thing? ... I don't know.
On the surface it seems not unreasonable for a small to medium
business to have deployed something of this nature. pure conjecture is
some business oriented company that has two or more 8-port e-net
hubs on each corridor ... with 6-7 offices/clients hung off each
hub and each hub having one port connected to the server.
So what kind of conjectures are why they might want to do this as a
peering operation rather than configuring the hubs hierarchical (so
everybody has LAN connectivity to everybody else). First might be why
not? Another might be they might have compartmentalized security
requirements. Maybe the hubs they have don't support hierarchical
configuration. I don't claim to know. It seems like a perfectly valid
configuration, at least until you get into some of the more
peculiar/wierd characteristics of (at least some) ip implementations.
... now in DNS one of the A-records has to be first ... and it will be
first regardless of the client requesting host->ip-address
resolution. If the clients asking for host->ip-address resolution are
on the same LAN as the LAN adapter with the ip-address in the first
record ... the ARP works.
So finally, we get to the situation described in previous postings ...
if the clients asking for host->ip-address resolution are on a LAN
different than the one indicated by the first A-record/ip-address then
it may not work ... as per previous discussion.
Possibly somebody can declare the described simple multi-LAN
environment as a violation of the ip architecture and tell customers
that they aren't allowed to create such enet configurations. Maybe
while they are at it they can declare client implementations w/o
multiple A-record support as illegal ip implementations.
We do know how to make it not work.
We can force the server to only answer ARP requests for the ip-address
that corresponds to the LAN adapter interface that the ARP request
arrived on.
And we know how to make it work.
We can allow the server to answer ARP requests for ip-addresses other
than the interface that the ARP request arrived on.
We could also declare that customers have to customize the client
applications based on which physical LAN they are connected to, so
that they utilize a LAN specific server host name.
Maybe somebody has a specialized DNS server ... that can serve out of
a different file based on the LAN interface the DNS request arrived
from? This possibly has a down-side of explaining to people why they
need a different DNS for each of their LANs.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Network separation using host w/multiple network interfaces
Newsgroups: comp.protocols.tcp-ip
Date: Thu, 06 Feb 2003 03:28:10 GMT
re:
http://www.garlic.com/~lynn/2003c.html#24
one of the solutions given to the multihome type of configuration
in the previous post from the 26apr88 multihome draft mentioned
in:
http://www.garlic.com/~lynn/2003c.html#8
was to have a unique ip-address for each LAN interface and a "host"
generic ip-address. Incoming ARPs would be answered for the ip-address
of the LAN interface that they arrived on ... and any ARP for the
generic host ip-address would be answered regardless of the LAN
interface it arrived.
the current ip architecture has an issue with regard to whether an
address is a network specific item (aka a LAN interface) ... while
there are paradigms that are just looking for a host address w/o the
complexity of worrying about network specific address issues. the work
around to achieving a generic host address is trying to hide behind
the facade of listing every possible network address (multiple
a-records) ... and pushing the problem out to the client application
code to try every possible network address as a way of discovering
some workable host address.
this dichotomy of only having network address support when some large
percentage of applications only want to deal with host/server address
is one of the issues with regard to multihome operation (as can be
seen with some number of client applications failing to implement
multiple A-record support).
the configuration problem described in the previous post can be
discounted as a routing problem ... aka the installation of router
magically eliminates the problem ... even tho there is direct LAN
connectivity between client and server ... the introduction of a
router "solves" the problem by making the traffic flow from one LAN to
another LAN in order to have communication between the client and the
server.
however one has got to suspect that any ip solution that requires that
packets must be routed across two different LANs in order to have
client/server communication (bypassing a direct physical LAN
connection between the client and the server) somehow is an indication
of a mismatch between the networking abstraction and the real world.
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Crypto Standards Organizations
Newsgroups: sci.crypt
Date: Thu, 06 Feb 2003 11:43:54 GMT
note that in the past, if NIST was picking up some standard from some
other body ... it typically rewrote it from scratch for FIPS. A couple
years ago, NIST said that in some cases they would start adopting
other organizations standards as is. One of those has been from X9
(actually X9F, crypto & security). You see that today with FIPS186-2
for digital signature standard ... the ecdsa part references X9.62
from X9F.
Moving up to ISO ... X9 is the sitting chair for TC68
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Thu, 06 Feb 2003 12:33:07 GMT
Terje Mathisen writes:
At the same time Novell's NCP stack ran 2X to 10X faster, with 3X to
10X less memory footprint.
minor historical note regarding beginnings of a pc server company in
provo ... also some reference to early beginnings of company called
sun.
http://www.garlic.com/~lynn/96.html#4a John Hartmann's Birthday Party
http://www.garlic.com/~lynn/2000g.html#40 No more innovation? Get serious
http://www.garlic.com/~lynn/2001d.html#42 IBM was/is: Imitation...
http://www.garlic.com/~lynn/2001n.html#24 Alpha vs. Itanic: facts vs. FUD
http://www.garlic.com/~lynn/2002g.html#79 Coulda, Woulda, Shoudda moments?
http://www.garlic.com/~lynn/2002o.html#33 Over-the-shoulder effect
--
Anne & Lynn Wheeler | lynn@garlic.com - http://www.garlic.com/~lynn/
Internet trivia, 20th anniv: http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Fri, 07 Feb 2003 15:07:56 GMT
"Charlie Gibbs" writes:
This changed later on, though. By the time 386 systems were
becoming common, Compaq had achieved the ironic state of being
more IBM-compatible [1] than IBM themselves. (Of course, IBM
shooting themselves in the foot by trying to use Micro Channel
to close up the architectural Pandora's box they had opened
didn't help much.)
ibm had gone thru the (plug) compatible (manufactures ... PCM) issue in
the late '60s and early '70s. it is a business that i got blamed for
helping originate when i was undergraduate ... aka pcm ref:
http://www.garlic.com/~lynn/subtopic.html#360pcm
in the late '60s, we had taken a interdata/3 (later to bought by
perkin/elmer) and reverse engineered the ibm mainframe channel
interface and built our own channel interface board ... programming
the interdata/3 to support terminals and emulate ibm mainframe
telecommunication controller.
there was some folklore what we did then motivated the whole SNA
architecture and the complexity of the pu4/pu5 interface (some
reference that this also in the wake of FS being killed ... which was
a much grander attempt to provide even higher level of integration).
misc. fs:
http://www.garlic.com/~lynn/subtopic.html#futuresys
and of course ... the SAA was an attempt at a higher level to address
the wholesale leakage of mainframe activities out thru the terminal
emulation interface into distributed computing environment:
http://www.garlic.com/~lynn/subnetwork.html#3tier
about 20 years later (in the mid-80s) ... i was involved in trying to
get an PU4 simulator (that had been running in s/1 boxes and was
mapped to high-performance packet network) ported to risk platform and
made standard offering ... misc. past ref to presentation I made to
the SNA architecture review board:
http://www.garlic.com/~lynn/2000.html#53 APPC vs TCP/IP
http://www.garlic.com/~lynn/2001i.html#31 3745 and SNI
http://www.garlic.com/~lynn/2002c.html#42 Beginning of the end for SNA?
ref to part of the actual presentation:
http://www.garlic.com/~lynn/internet.htm#17
http://www.garlic.com/~lynn/99.html#67 System/1 ?
http://www.garlic.com/~lynn/2001i.html#21 3745 and SNI
http://www.garlic.com/~lynn/2001k.html#21 OT: almost lost LBJ tapes; Dictabelt
imagine treaking a lion in its den ... an audience of approx. 60
people and half were openly hostile. if the communication division was
unhappy with strategies to migrate away from terminal emulation for
PC infrastructures ... i.e. previous post in this thread:
http://www.garlic.com/~lynn/2003c.html#23 diffence between itanium and alpha
they were also extremely hostile about the idea of eliminating most of
their straight bread&butter boxes.
also part of the referenced configuration leveraged HYPERChannel link
adapters for high-speed backbone. I had also done the mainframe
rfc1044 support and was getting something like 50 times
thruput/performance as the base (non-rfc1044) support. misc rfc 1044
refs:
http://www.garlic.com/~lynn/93.html#28 Log Structured filesystems -- think twice
http://www.garlic.com/~lynn/96.html#14 mainframe tcp/ip
http://www.garlic.com/~lynn/96.html#15 tcp/ip
http://www.garlic.com/~lynn/96.html#17 middle layer
http://www.garlic.com/~lynn/98.html#34 ... cics ... from posting from another list
http://www.garlic.com/~lynn/98.html#49 Edsger Dijkstra: the blackest week of his professional life
http://www.garlic.com/~lynn/98.html#50 Edsger Dijkstra: the blackest week of his professional life
http://www.garlic.com/~lynn/99.html#36 why is there an "@" key?
http://www.garlic.com/~lynn/2000.html#90 Ux's good points.
http://www.garlic.com/~lynn/2000c.html#59 Does the word "mainframe" still have a meaning?
http://www.garlic.com/~lynn/2000f.html#30 OT?
http://www.garlic.com/~lynn/2001d.html#63 Pentium 4 Prefetch engine?
http://www.garlic.com/~lynn/2001d.html#65 Pentium 4 Prefetch engine?
http://www.garlic.com/~lynn/2001g.html#33 Did AT&T offer Unix to Digital Equipment in the 70s?
http://www.garlic.com/~lynn/2001h.html#44 Wired News :The Grid: The Next-Gen Internet?
http://www.garlic.com/~lynn/2001j.html#20 OT - Internet Explorer V6.0
http://www.garlic.com/~lynn/2002.html#11 The demise of compaq
http://www.garlic.com/~lynn/2002i.html#43 CDC6600 - just how powerful a machine was it?
http://www.garlic.com/~lynn/2002i.html#45 CDC6600 - just how powerful a machine was it?
http://www.garlic.com/~lynn/2002j.html#67 Total Computing Power
http://www.garlic.com/~lynn/2002k.html#31 general networking is: DEC eNet: was Vnet : Unbelievable
http://www.garlic.com/~lynn/2002n.html#58 IBM S/370-168, 195, and 3033
http://www.garlic.com/~lynn/2002o.html#51 E-mail from the OS-390 ????
http://www.garlic.com/~lynn/2002q.html#27 Beyond 8+3
http://www.garlic.com/~lynn/2003.html#67 3745 & NCP Withdrawl?
http://www.garlic.com/~lynn/2003b.html#29 360/370 disk drives
http://www.garlic.com/~lynn/2003b.html#44 filesystem structure, was tape format (long post)
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Will Apple ever offer a newsreader?
Newsgroups: comp.sys.mac.comm,alt.folklore.computers
Date: Fri, 07 Feb 2003 15:23:40 GMT
Toby Thain writes:
Brewster Kahle, http://www.kahle.org/Brewster/
via http://www.google.com.au/search?q=brewster+kahle
i saw brewster in nov. ...he was running around in his book-on-demand
van ... with a satellite dish on top.
first time i actually met brewster was at the wais, inc "house" ...
for their offices ... they bought what seemed to be old 3-story house
... just east of el camino in menlo park ... near the palo alto
line. of course ... this was all before aol bought him out.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Fri, 07 Feb 2003 15:34:02 GMT
bdc@world.std.com (Brian 'Jarai' Chase) writes:
Actually, this is only true with DECnet Phase-V, because it was based on
OSI; it wasn't released until the late 1980s (or early 1990s?). The
previous version of DECnet only supported 63 areas and 1023 nodes per
area, giving the world a total of 64449 unique system addresses. So,
during the 1980s, TCP/IP had a significant advantage in terms of its
address space, scalability, portability, and cost. By the time DEC
realized and remedied their technical deficiences, TCP/IP had already
won the battle.
note ... an earlier post in this thread had several references to much
earlier threads that contained extracts from troubles that decnet was
having with the whole osi support stuff.
http://www.garlic.com/~lynn/2003c.html#22 diffence between itanium and alpha
sna had a similar 64k addressing limit .... they then invented
cross-domain in the mid-80s to support some of the customer
configurations running several hundred thousand "terminals" (for
instance some of the airline res systems ... or even several millions,
some cable system set-top boxes infrastructures).
the previously posted ref.
http://www.garlic.com/~lynn/2003c.html#28 diffence between itanium and alpha
about the sna simulation on a high-performance peer-to-peer packet
implementation ... leveraged the cross-domain support by telling the
the mainframe pu5/vtam that everything was cross-domain (and then
managing the configuration in a replicated, distributed manner
.... avoiding several of the single-point-of-failure issues in the
native implementation).
airline res topic drift:
http://www.garlic.com/~lynn/99.html#17 Old Computers
http://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
http://www.garlic.com/~lynn/99.html#103 IBM 9020 computers used by FAA (was Re: EPO stories (was: HELP IT'S HOT!!!!!))
http://www.garlic.com/~lynn/99.html#136a checks (was S/390 on PowerPC?)
http://www.garlic.com/~lynn/2000f.html#20 Competitors to SABRE?
http://www.garlic.com/~lynn/2001.html#26 Disk caching and file systems. Disk history...people forget
http://www.garlic.com/~lynn/2001c.html#37 database (or b-tree) page sizes
http://www.garlic.com/~lynn/2001d.html#69 Block oriented I/O over IP
http://www.garlic.com/~lynn/2001g.html#45 Did AT&T offer Unix to Digital Equipment in the 70s?
http://www.garlic.com/~lynn/2001g.html#49 Did AT&T offer Unix to Digital Equipment in the 70s?
http://www.garlic.com/~lynn/2001j.html#17 I hate Compaq
http://www.garlic.com/~lynn/2001n.html#0 TSS/360
http://www.garlic.com/~lynn/2001n.html#3 News IBM loses supercomputer crown
http://www.garlic.com/~lynn/2002g.html#2 Computers in Science Fiction
http://www.garlic.com/~lynn/2002g.html#3 Why are Mainframe Computers really still in use at all?
http://www.garlic.com/~lynn/2002h.html#12 Why did OSI fail compared with TCP-IP?
http://www.garlic.com/~lynn/2002h.html#43 IBM doing anything for 50th Anniv?
http://www.garlic.com/~lynn/2002i.html#83 HONE
http://www.garlic.com/~lynn/2002j.html#83 Summary: Robots of Doom
http://www.garlic.com/~lynn/2002m.html#67 Tweaking old computers?
http://www.garlic.com/~lynn/2003.html#48 InfiniBand Group Sharply, Evenly Divided
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Fri, 07 Feb 2003 15:37:34 GMT
oh, slight acorn topic drift:
http://www.garlic.com/~lynn/2002g.html#79 Coulda, Woulda, Shoudda moments?
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Early attempts at console humor?
Newsgroups: alt.folklore.computers
Date: Fri, 07 Feb 2003 17:25:57 GMT
bdc@world.std.com (Brian 'Jarai' Chase) writes:
Well... this isn't console or mainframe humor, but NEXTSTEP binaries use
some interesting values for their magic numbers. For the m68k based
hardware, the first two bytes of a binary spell out "feedface" in hex.
The fat binaries they distributed used the magic number of "cafebabe" and
I think there might have been a third variation for one the ports to
either SPARC, PA-RISC, or Intel.
It's also worth noting that Apple's current OS X, which is based on
NEXTSTEP, recycles this bit of humor.
and nextstep was based on mach
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers
Date: Fri, 07 Feb 2003 17:35:27 GMT
desk top real estate is still issue .... recent ref:
http://news.com.com/2100-1001-983783.html?tag=fd_top
part of the original issue was being able to collapse a terminal and a
desk top pc into a single box, single display, single keyboard (with
terminal emulation in the pc) ... i had arguments about this with some
of the apple people before the mac was released.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Fri, 07 Feb 2003 19:01:29 GMT
Dan.Pop@cern.ch (Dan Pop) writes:
With vi in command mode, each key is an "exceptional key". To have it
working satisfactorily, from the user's point of view, the fallback
timeout would have to be so short as to render the whole scheme
ineffective (each character is sent alone in a packet). And the local
echo would be rather annoying to the user, who expects something
completely different as output.
when the 3101 came out ... the guy responsible for PVM ... did special
version for the internal home terminal program .... using the 3101 in
block mode and doing 3270 terminal emulation ... did a bunch of stuff
to preserve as much data in the local 3101 buffer ... attempting to
re-arrange data already on the screen w/o having to retransmit it
(analogous to mpeg compression ... don't transmit data already in the
previous frame, if necessary just location move orders).
this was significantly enhanced for pc's home terminal emulator ...
where both the pc & the host side kept track of serveral screens worth
of data ... and if data was in each other's buffer .... just transmit
pointer & length to data already in buffer (in a sense analogous to
huffman dictionary compression).
things were still at 2400 baud.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Fri, 07 Feb 2003 19:39:54 GMT
oh & some past 3101 posts:
http://www.garlic.com/~lynn/99.html#69 System/1 ?
http://www.garlic.com/~lynn/2000g.html#17 IBM's mess (was: Re: What the hell is an MSX?)
http://www.garlic.com/~lynn/2001b.html#12 Now early Arpanet security
http://www.garlic.com/~lynn/2001b.html#13 Now early Arpanet security
http://www.garlic.com/~lynn/2001h.html#32 Wanted: pictures of green-screen text
http://www.garlic.com/~lynn/2001m.html#1 ASR33/35 Controls
http://www.garlic.com/~lynn/2001m.html#54 Author seeks help - net in 1981
slightly related terminal i/o topic drift posts:
http://www.garlic.com/~lynn/2001f.html#57 any 70's era supercomputers that ran as slow as today's supercomputers?
Somewhere in the vm/370 time-frame ... there was some code that went
into the 3270 simulation of virtual "line-mode" 1052/3215 .... to
collect the one line write at a time ... and package them into a
single 3270 screen refresh write. This reduced some aspects of the
real i/o overhead that is touched on in the above reference.
Earlier, I had done a hack on the cms 1052/3215 i/o write routine
(DMSCIT) that instead of doing a separate i/o for each line written,
to do multiple line i/o writes. My cms change had the added benefit of
reducing the frequency of crossing back & forth between cms and the cp
kernel (as well as various scheduling overhead issues) ... and in
addition it had the benefit of helping not just 3270 (real) terminals
... but all terminals.
From: wheeler
Date: 3/29/79 17:07:00
Subject: changes to DMSCIT to chain write stacked data with one SIO.
I have just completed modifications to DMSCIT which will attempt to
write all queued write request with one SIO. The number of lines
chained together is dependent on the size of the lines (CONSTACK is
large enuf to hold 320 bytes of data). This can have a significant
performance impact on the system since the virtual machine is not
being waked up after each output line to add another to the write
queue. Initial tests running on a lightly loaded system (i.e. no
flushing of pages while idle, no queue delay, no page i/os) for a
LISTF command of 109 files showed a reduction of virtual/total time
from .11/.87 to .07/.44 seconds. Only filename/filetype/filemode list
was requested so each line 21 bytes and CIT was blocking up to 12
lines per sio. With longer lines the blocking (and savings would be
somewhat less). On heavily loaded systems the savings would be much
greater since the number of queue drops/adds and page flush/brings
would be reduced by a factor of 12 (even with 80 character lines the
drops/adds are reduced by a factor of 4, and the corresponding paging
activity to page in and out is reduced by 75%).
Corresponding changes for CP have not yet been done. Even tho the
write CCWs have been chained together, it appears that VCN is calling
QCN with 1 line at a time. This is not significant on a line terminal,
but there can be significant savings driving full screen devices if CP
would also block the output.
.. Lynn W., K03/281, San Jose Res., 408-256-1783 (8-276)
... snip ... top of post, old email index
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: "average" DASD Blocksize
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Fri, 07 Feb 2003 19:54:30 GMT
norbert.friemel@KIVBF.DE (Friemel, Norbert) writes:
http://www.sdisw.com/dasd_capacity.html
note that 2301 & 2303 drums are missing.
trivia question ... why did 2305-1 only have half the capacity and
cylinders as the 2305-2 (hint it also had half the avg. rotational
delay).
also, could anybody help with the product code names in the
table listed in the following post?
http://www.garlic.com/~lynn/2003b.html#7 Disk drives as commodities. Was Re: Yamhill
2301, 2303, 2311, 2314, 2321, 3340, 3344, 3830, 3880-23, 3390
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: "average" DASD Blocksize
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Fri, 07 Feb 2003 21:17:16 GMT
Matthew.Stitt@ABCDISTRIBUTING.COM (Matthew Stitt) writes:
The box behind the lady looks like a 2302 Fixed Disk drive;
that honker in the background looks like a 360 of some flavor.
Doesn't look like any of the 360's I've worked with. Might be a 91 or a 195.
elsewhere on the site the list it as a 2301 drum ... with close-ups
the 2305s where fixed-head disks ... but the 2301 & 2303 were fixed
head drums. the 2301 was essentially a 2303 that read/wrote four heads
in parallel (and therefor had 4 times the transfer rate).
somewhat related:
http://www.garlic.com/~lynn/2003.html#70 Disk drives as commodities. Was Re: Yamhill
http://www.garlic.com/~lynn/2003b.html#6 Disk drives as commodities. Was Re: Yamhill
http://www.garlic.com/~lynn/2003b.html#7 Disk drives as commodities. Was Re: Yamhill
http://www.garlic.com/~lynn/2003b.html#9 Disk drives as commodities. Was Re: Yamhill
http://www.garlic.com/~lynn/2003b.html#10 Disk drives as commodities. Was Re: Yamhill
http://www.garlic.com/~lynn/2003b.html#18 Card Columns
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: "average" DASD Blocksize
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Fri, 07 Feb 2003 21:19:54 GMT
IBM-MAIN@ISHAM-RESEARCH.COM (Phil Payne) writes:
The reason, of course, was that the 3370 worked (as long as you
didn't switch it off over the weekend) and IBM knew the 3380 didn't.
It was nine months before IBM could produce 3380s in volume.
can you spell stiction?
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: DOD 5200.28-STD capable OS?
Newsgroups: comp.security.unix,comp.security.misc,comp.security
Date: Fri, 07 Feb 2003 21:39:29 GMT
Daniel Faigin writes:
The main reason you are not seeing "B2 and above" is that the
evaluation methodology has not been defined for EAL5 and above. Thus,
there is no mutually accepted way to evaluate these things. However,
even with that, I'm aware of efforts by Getronics (nee Wang Federal
nee HFSI nee Honeywell) to evaluate a high-assurance XTS-400 OS, which
would have a Linux-compatible interface.
i'm actually in the process of getting an eal4 +/high for
something. there is evaluation methodology for higher than eal4
... but there isn't any multinational concensus aggreement on it (aka
you can get an eal5-high in europe ... but a certificate in the US
would be "donwgradded" to eal4-high).
I've somewhat made off-the-cuff comments that common criteria is
orange book with a lot of protection profiles tailored to specialized
conditions ... a common method for creating protection profiles ...
but somewhat difficult to equate between different protection profiles
(aka in some sense protection profiles contain deviations from what
would have been orange book .... based on unique circumstances of the
thing being evaluated).
Recently there was a presentation that of the sixty common criteria
evaluated products ... all but four had gotten special case
exemptions. So if in some sense common criteria protection profiles
represent special case exemptions from the orange book .... it turns
out then that there are additional special case exemptions from the
protection profiles (which sort of are special case exemptions to
begin with).
and for some topic drift ... i've merged glossaries/taxonomies
from several of these sources:
aka:
http://www.garlic.com/~lynn/index.html#glosnote
Security
Terms merged from: AFSEC, AJP, CC1, CC2, CC21 (CC site), CIAO, FCv1,
FIPS140, IATF V3 (IATF site), IEEE610, ITSEC, Intel, JTC1/SC27 (sc27
site), KeyAll, MSC, NIST 800-37, NCSC/TG004, NIAP, NSA Intrusion,
NSTISSC/CNSS, online security study, RFC1983, RFC2504, RFC2647,
RFC2828, TCSEC, TDI, TNI, and misc. Updated 20021108 with terms from
CIAO. Updated 20021205 with terms from 800-37 glossary.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: "average" DASD Blocksize
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Sat, 08 Feb 2003 16:28:16 GMT
markan@US.IBM.COM (Mark Nelson) writes:
It looks an awful lot like a 75. See
http://www.ibm.com/ibm/history/exhibits/attic/attic_054.html
for a quick history and photo. My first computing job (late 70s) was
working as a weekend operator for a company that ran a large
commercial time sharing system on a pair of model 75s. I can still
remember how to toggle in the "clear storage" instructions.
http://www.ibm.com/news/ls/1999/07/articles/sidebar_16.phtml
has an article on the role of the Model 75 in the US Apollo space
program.
Mark Nelson, RACF Design and Development, IBM Poughkeepsie
when dick bayles (et al) came out from cambridge science center to the
university to install cp/67 in jan. '68 ... i was impressed with his
ability to "play" the the toggles on the front of the 360/67. there
was a certain rhythm to setting the toggles ... and then clearing them
in a sweeping motion of the finger under the row of toggles. took a
bit of practice to duplicate his ability on the toggles.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: How much overhead is "running another MVS LPAR" ?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Sat, 08 Feb 2003 16:36:00 GMT
Anton_Mvs@DOUGLAWSONANDMIDDLEOFTHEROAD.COM (Anton Britz) writes:
The question is "Has anybody calculated how much of your machine do you
lose by running another MVS " ?
Years a'go, in a computing centre environment , where we had to charge
out as much as possible of the MIPS as per the IBM sales brochure, we
found that the overhead, the part you can not charge out, comes to about
28% of the allocated MIPS.
Has anybody done a recent survey ?
there was a joke about how much of the machine do you loose just by
running mvs ... period.
it was part of the point that a lot of the attention on cp/67
"overhead" (and significant efforts to performance optimize cp/67) was
because it was possible to run some number of applications with and
w/o cp/67 ... and measure the difference. because most of the other
operating systems lacked the ability to measure with & w/o the
operating systems .... there was much less attention paid to critical
performance issues.
however, while it wasn't directly possible to run with & w/o mvs to
really get a feal for how much of total system resources were being
consumed by mvs. some simple studies compared running the same exact
workload under MVT and under MVS .... and getting how much additional
processor was lost to MVS. Also, for at least some workloads, doing
something similar with running the same workload under CMS ... and
measuring how much of the processor was lost to MVS.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Sat, 08 Feb 2003 20:04:54 GMT
Brian Inglis writes:
I know TCP/IP routes around transient and permanent network
problems; ARPAnet may have had some such capability too.
IMPs handled that ... as well as trying to do things like traffic load
balancing .... also out-of-order packet arrival (because of different
paths).
i have vaque recollections from late '70s of somebody commenting that
a significant percentage of the 56kbit link bandwidth was starting to
be inter-imp chatter exchanging information in support of routing
strategies.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Early attempts at console humor?
Newsgroups: alt.folklore.computers
Date: Sat, 08 Feb 2003 20:33:20 GMT
somewhat related mainframe humor .... pithy sayings printed on the
6670 "separater" page
http://www.garlic.com/~lynn/99.html#52 Enter fonts (was Re: Unix case-sensitivity: how did it originate?
other refs to the same:
http://www.garlic.com/~lynn/2001g.html#5 New IBM history book out
http://www.garlic.com/~lynn/2002h.html#7 disk write caching (was: ibm icecube -- return of
http://www.garlic.com/~lynn/2002o.html#24 IBM Selectric as printer
misc. refs to ibm jargon file:
http://www.garlic.com/~lynn/2001g.html#5 New IBM history book out
http://www.garlic.com/~lynn/2001n.html#79 a.f.c history checkup... (was What specifications will the standard year 2001 PC have?)
http://www.garlic.com/~lynn/2002e.html#45 REXX and its designer (was: IBM 7090 instruction set)
http://www.garlic.com/~lynn/2002h.html#7 disk write caching (was: ibm icecube -- return of
http://www.garlic.com/~lynn/2002k.html#39 Vnet : Unbelievable
http://www.garlic.com/~lynn/2003b.html#19 Card Columns
fake corporate directive on password rules ... printed with 6670 on
corporate letterhead and posted on corporate bulletin boards ... (and
some people took quite a while to realize it was joke). Resulting
investigation led to placing all blank corporate letterhead paper
under lock & key (and it wasn't me that put it up on corporate
bulletin boards):
http://www.garlic.com/~lynn/2001d.html#51 OT Re: A beautiful morning in AFM.
http://www.garlic.com/~lynn/2001d.html#52 OT Re: A beautiful morning in AFM.
http://www.garlic.com/~lynn/2001d.html#62 OT Re: A beautiful morning in AFM.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Easiest possible PASV experiment
Newsgroups: comp.os.linux.networking,comp.protocols.tcp-ip
Date: Sat, 08 Feb 2003 20:47:01 GMT
"Don Kelloway" writes:
We are getting way of topic. <grin>
The book is obviously not technical when compared to the others we
read. In fact the one paragraph cited represented the entire
section discussing FTP.
The intent was nothing more than to offer validation that FTP is the
preferred method of transferring files as opposed to using HTTP.
This of itself being off topic from the subject of whether FTP (at
the Application Layer) uses compression when transferring data.
it may be purely gut feel .... but it seems to be that (some?) ftp
implementations leverage TCP windowing (aka latency compensation) much
better than http implementations.
misc. recent discussion:
http://www.garlic.com/~lynn/2003.html#55 Cluster and I/O Interconnect: Infiniband, PCI-Express, Gibat
http://www.garlic.com/~lynn/2003.html#59 Cluster and I/O Interconnect: Infiniband, PCI-Express, Gibat
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Early attempts at console humor?
Newsgroups: alt.folklore.computers
Date: Sun, 09 Feb 2003 00:29:14 GMT
... next press release oct '88. note that nextstep (as opposed to
next) was made available on a number of other platforms.
http://www.garlic.com/~lynn/2003c.html#32 Early attempts at console humor
...snip...
Software as Part of the System NeXT includes an unparalled amount of
software in the price of every NeXT Computer System. The software
starts with Mach, an advanced multitasking operating system compatible
with 4.3BSD UNIX, which is the standard operating system in higher
education communities. In addition, the NeXT Computer System includes
NextStep, a complete software environment consisting of four
components: the Window Server, the Workspace Manager, the Application
Kit and the Interface Builder. The object-oriented environment was
developed with the Objective-C programming language, from the
Stepstone Corp. NextStep solves the two major problems with
UNIX-based systems: They are too complex and difficult for most
non-programmers to use, and they require developers to spend an
inordinate amount of time and expertise creating graphical, end-user
applications. For users, NextStep makes the power of UNIX available
by substituting a window-based, graphical and intuitive interface for
the traditional UNIX comand-line interface. For developers, NextStep
includes the Application Kit, a set of interacting software "objects"
for constructing applications. Also included in NextStep is Interface
Builder, a completely new kind of software development tool.
Interface Builder works graphically, letting the developer construct
an application by choosing from a palette of available objects and
using the mouse and keyboard to modify the objects as needed, define
the layout and establish connections between objects. This process
permits the rapid construction of graphical user interfaces and makes
application development accessible to a much larger community.
NextStep uses the Display PostScript system to ensure true WYSIWYG
(What You See Is What You Get) between the screen and the printer.
The Display PostScript system includes a high-performance
implementation of the PostScript language, the de facto imaging
standard for printing. It simplifies the programming of graphical
applications that support high-quality printing.
...snip...
posted to comp.os.mach 18aug89
...snip...
Information on Mach licensing and distribution and technical
reports can be obtained by writing to:
Mach Project
c/o Rick Rashid
School of Computer Science
Carnegie Mellon University
Pittsburgh, PA 15213
or
mach@cs.cmu.edu
Current commercial products I am aware of are from NeXT, Encore,
Evans&Sutherland and BBN.
Mt Xinu has announced that it will be distributing commercial versions
of Mach on several architectures beginning early next year (in much
the way it has distributed BSD Unix).
CMU distributes Mach for the VAX, Sun 3 and IBM RT architectures.
Information on specific models, etc. can be obtained with the general
Mach information packet. The current release is referred to as Mach
Release 2. We are about to begin general distribution of Mach Release
2.5 (a number of Universities and companies already have early
releases of 2.5). There is a license for Mach from CMU and you will
need a Berkeley license to get a tape from us. There is no
distribution or license fee paid to CMU, however.
At CMU Mach runs on VAXen (uni and multiprocessors), DEC 3100s,
Multimaxes, Sun 3s, Sun 4s, 386s, IBM RTs, IBM 370s and Macintosh IIs.
Ports have also been done to a number of other machines by groups
outside CMU. CMU has limited capacity to distribute software so we
don't distribute Mach for all systems. We are, however, willing to
provide Mach free to any manufacturer who is interested in
distributing it for their own machines. If you want Mach for a
machine for which there is a port but no current distributor, you
should talk to your salesman or corporate representative.
...snip...
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Sun, 09 Feb 2003 00:42:45 GMT
Morten Reistad writes:
When the IMPs went away there was no hardware move to build proper
routers, even though the design of the Internet was screaming for them.
IBM built some RS6000(?) based boxes that were serious kludges for the
NSF bid, and Cisco hit the market with their GS series of routers. This
was also a serious kludge.
ibm pc/rts with 440kbit adapter boards .... with idnx multiplexors
handling mapping into the telco/mci T1 (1.5mbit) links. the backbone
locations had room with several racks of pc/rts. misc. past
refs:
http://www.garlic.com/~lynn/99.html#146 Dispute about Internet's origins
http://www.garlic.com/~lynn/2000.html#49 IBM RT PC (was Re: What does AT stand for ?)
http://www.garlic.com/~lynn/2000c.html#78 Free RT monitors/keyboards
http://www.garlic.com/~lynn/2001.html#4 Sv: First video terminal?
http://www.garlic.com/~lynn/2001e.html#76 Stoopidest Hardware Repair Call?
at the time of netnet1 bid, my wife & I were operating T1 backbone
with NSC boxes, and were sort of red-team on both nsfnet1 (T1) and
nsfnet2 (T3) although we weren't allowed to bid (although nsf audit of
what we had running was that it was five years ahead of all bid proposals
to build something new). random ref:
http://www.garlic.com/~lynn/internet.htm#0 Internet and/or ARPANET
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: diffence between itanium and alpha
Newsgroups: comp.arch,alt.folklore.computers,comp.protocols.tcp-ip
Date: Sun, 09 Feb 2003 14:34:43 GMT
Patrick Scheible writes:
Yeah, but I'm pretty sure the routing was already done in software
under NCP. Much has been made about 1/1/83 being the "birth of the
internet", but it didn't seem like such a big deal at the time, the
same things got done in pretty much the same way from the user's point
of view. E-mail, file transfers, etc. were well-established years
before. Routing didn't work as well for several years after the
switchover as it did before, chances for failure at more points, and
slower propagation of information about new hosts.
in part because a majority of the bandwidth wasn't being consumed
with inter-imp chatter.
the other issue was that internet got the ip layer and real gateways
which removed one of the significant inhibitors to the growth in the
size of the internet. i've asserted that one of the reasons that the
internal network was larger than the arpanet/internet up thru sometime
in the 1985 time-frame was that the internal network effectively had
gateway-like function from the beginning. the other growth factor in
the internet with the advent of ip (and conversion from the IMPs) was
the availability of the support on workstations and PCs .... greatly
increasing the potential nodes. with the big increase in the number of
nodes ... the dedicated 56kbit links of the IMPs wouldn't have been
sufficient to carry the inter-imp chatter.
1983 internal network:
http://www.garlic.com/~lynn/internet.htm#22
http://www.garlic.com/~lynn/99.html#112
random internal network posts:
http://www.garlic.com/~lynn/2002.html#32 Buffer overflow
http://www.garlic.com/~lynn/2002b.html#53 Computer Naming Conventions
http://www.garlic.com/~lynn/2002b.html#54 Computer Naming Conventions
http://www.garlic.com/~lynn/2002b.html#56 Computer Naming Conventions
http://www.garlic.com/~lynn/2002b.html#57 Computer Naming Conventions
http://www.garlic.com/~lynn/2002b.html#58 ibm vnet : Computer Naming Conventions
http://www.garlic.com/~lynn/2002b.html#59 Computer Naming Conventions
http://www.garlic.com/~lynn/2002b.html#64 ... the need for a Museum of Computer Software
http://www.garlic.com/~lynn/2002c.html#39 VAX, M68K complex instructions (was Re: Did Intel Bite Off More Than It Can Chew?)
http://www.garlic.com/~lynn/2002d.html#9 Security Proportional to Risk (was: IBM Mainframe at home)
http://www.garlic.com/~lynn/2002d.html#11 Security Proportional to Risk (was: IBM Mainframe at home)
http://www.garlic.com/~lynn/2002d.html#14 Mainframers: Take back the light (spotlight, that is)
http://www.garlic.com/~lynn/2002d.html#33 LISTSERV(r) on mainframes
http://www.garlic.com/~lynn/2002e.html#47 Multics_Security
http://www.garlic.com/~lynn/2002g.html#35 Why did OSI fail compared with TCP-IP?
http://www.garlic.com/~lynn/2002g.html#71 Coulda, Woulda, Shoudda moments?
http://www.garlic.com/~lynn/2002h.html#5 Coulda, Woulda, Shoudda moments?
http://www.garlic.com/~lynn/2002h.html#11 Why did OSI fail compared with TCP-IP?
http://www.garlic.com/~lynn/2002h.html#22 Why did OSI fail compared with TCP-IP?
http://www.garlic.com/~lynn/2002h.html#48 Why did OSI fail compared with TCP-IP?
http://www.garlic.com/~lynn/2002h.html#64 history of CMS
http://www.garlic.com/~lynn/2002j.html#4 HONE, , misc
http://www.garlic.com/~lynn/2002j.html#22 Computer Terminal Design Over the Years
http://www.garlic.com/~lynn/2002j.html#52 "Slower is more secure"
http://www.garlic.com/~lynn/2002j.html#64 vm marketing (cross post)
http://www.garlic.com/~lynn/2002j.html#66 vm marketing (cross post)
http://www.garlic.com/~lynn/2002k.html#18 Unbelievable
http://www.garlic.com/~lynn/2002k.html#19 Vnet : Unbelievable
http://www.garlic.com/~lynn/2002k.html#20 Vnet : Unbelievable
http://www.garlic.com/~lynn/2002k.html#23 Vnet : Unbelievable
http://www.garlic.com/~lynn/2002k.html#42 MVS 3.8J and NJE via CTC
http://www.garlic.com/~lynn/2002k.html#48 MVS 3.8J and NJE via CTC
http://www.garlic.com/~lynn/2002l.html#53 10 choices that were critical to the Net's success
http://www.garlic.com/~lynn/2002n.html#35 VR vs. Portable Computing
http://www.garlic.com/~lynn/2002o.html#4 Mainframe Spreadsheets - 1980's History
http://www.garlic.com/~lynn/2002o.html#17 PLX
http://www.garlic.com/~lynn/2002o.html#54 XML, AI, Cyc, psych, and literature
http://www.garlic.com/~lynn/2002o.html#78 Newsgroup cliques?
http://www.garlic.com/~lynn/2002q.html#4 Vector display systems
http://www.garlic.com/~lynn/2002q.html#31 Collating on the S/360-2540 card reader?
http://www.garlic.com/~lynn/2002q.html#35 HASP:
http://www.garlic.com/~lynn/2003.html#10 Mainframe System Programmer/Administrator market demand?
http://www.garlic.com/~lynn/2003.html#68 3745 & NCP Withdrawl?
http://www.garlic.com/~lynn/2003b.html#44 filesystem structure, was tape format (long post)
http://www.garlic.com/~lynn/2003b.html#46 internal network drift (was filesystem structure)
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/ Internet trivia
20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: "average" DASD Blocksize
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Mon, 10 Feb 2003 00:52:46 GMT
SEYMOUR.J.METZ@CUSTOMS.TREAS.GOV (Shmuel Metz , Seymour J.) writes:
You address data on an FBA device by block, but the disks are still
physically organized by cylinder and track. And I still don't understand
why IBM is unwilling to do FBA support for MVS :-(
response i got from stl (long ago and far away) ... was that if i gave
it to them fully tested and integrated ... it would still cost $26m
for release
part of the discussion was the whole multi-track search rework ...
which changes the design point from memory constrained to i/o
constrained.
past mvs fba posts:
http://www.garlic.com/~lynn/97.html#16 Why Mainframes?
http://www.garlic.com/~lynn/97.html#29 IA64 Self Virtualizable?
http://www.garlic.com/~lynn/99.html#75 Read if over 40 and have Mainframe background
http://www.garlic.com/~lynn/2000.html#86 Ux's good points.
http://www.garlic.com/~lynn/2000f.html#18 OT?
http://www.garlic.com/~lynn/2000g.html#51 > 512 byte disk blocks (was: 4M pages are a bad idea)
http://www.garlic.com/~lynn/2001.html#54 FBA History Question (was: RE: What's the meaning of track overfl ow?)
http://www.garlic.com/~lynn/2001d.html#64 VTOC/VTOC INDEX/VVDS and performance (expansion of VTOC position)
http://www.garlic.com/~lynn/2001g.html#32 Did AT&T offer Unix to Digital Equipment in the 70s?
http://www.garlic.com/~lynn/2002.html#5 index searching
http://www.garlic.com/~lynn/2002.html#10 index searching
http://www.garlic.com/~lynn/2002g.html#13 Secure Device Drivers
http://www.garlic.com/~lynn/2002l.html#47 Do any architectures use instruction count instead of timer
http://www.garlic.com/~lynn/2003.html#15 vax6k.openecs.org rebirth
past multi-track search posts:
http://www.garlic.com/~lynn/93.html#29 Log Structured filesystems -- think twice
http://www.garlic.com/~lynn/94.html#35 mainframe CKD disks & PDS files (looong... warning)
http://www.garlic.com/~lynn/97.html#16 Why Mainframes?
http://www.garlic.com/~lynn/97.html#29 IA64 Self Virtualizable?
http://www.garlic.com/~lynn/99.html#75 Read if over 40 and have Mainframe background
http://www.garlic.com/~lynn/2000.html#86 Ux's good points.
http://www.garlic.com/~lynn/2000c.html#34 What level of computer is needed for a computer to Love?
http://www.garlic.com/~lynn/2000f.html#18 OT?
http://www.garlic.com/~lynn/2000f.html#19 OT?
http://www.garlic.com/~lynn/2000f.html#42 IBM 3340 help
http://www.garlic.com/~lynn/2000g.html#42 4M pages are a bad idea (was Re: AMD 64bit Hammer CPU and VM)
http://www.garlic.com/~lynn/2000g.html#51 > 512 byte disk blocks (was: 4M pages are a bad idea)
http://www.garlic.com/~lynn/2000g.html#52 > 512 byte disk blocks (was: 4M pages are a bad idea)
http://www.garlic.com/~lynn/2001.html#19 Disk caching and file systems. Disk history...people forget
http://www.garlic.com/~lynn/2001c.html#17 database (or b-tree) page sizes
http://www.garlic.com/~lynn/2001d.html#60 VTOC/VTOC INDEX/VVDS and performance (expansion of VTOC position)
http://www.garlic.com/~lynn/2001d.html#64 VTOC/VTOC INDEX/VVDS and performance (expansion of VTOC position)
http://www.garlic.com/~lynn/2001l.html#40 MVS History (all parts)
http://www.garlic.com/~lynn/2002.html#5 index searching
http://www.garlic.com/~lynn/2002.html#6 index searching
http://www.garlic.com/~lynn/2002.html#10 index searching
http://www.garlic.com/~lynn/2002c.html#48 Swapper was Re: History of Login Names
http://www.garlic.com/~lynn/2002d.html#22 DASD response times
http://www.garlic.com/~lynn/2002e.html#8 What are some impressive page rates?
http://www.garlic.com/~lynn/2002f.html#8 Is AMD doing an Intel?
http://www.garlic.com/~lynn/2002g.html#13 Secure Device Drivers
http://www.garlic.com/~lynn/2002l.html#47 Do any architectures use instruction count instead of timer
http://www.garlic.com/~lynn/2002l.html#49 Do any architectures use instruction count instead of timer
http://www.garlic.com/~lynn/2002n.html#50 EXCP
http://www.garlic.com/~lynn/2002o.html#46 Question about hard disk scheduling algorithms
http://www.garlic.com/~lynn/2003.html#15 vax6k.openecs.org rebirth
http://www.garlic.com/~lynn/2003b.html#22 360/370 disk drives
http://www.garlic.com/~lynn/2003b.html#33 dasd full cylinder transfer (long post warning)
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Filesystems
Newsgroups: comp.sys.cbm,comp.sys.apple2,rec.games.video.classic,comp.os.cpm,alt.folklore.computers
Date: Mon, 10 Feb 2003 01:08:46 GMT
Peter Flass writes:
[not topposted due to user request]
AIX has had a "journaled file system" (JFS) for years, and I'd have to
assume they weren't the first. It has recently been ported to OS/2.
i believe it may have been the first unix product filesystem. it was
originally done using the rios (aka power/6000) database memory
.... basically segment that had fine-grain modification turned on. the
filesystem metadata was mapped into the segment. at checkpoints
... could go around and find all the areas that had been modified and
log them (note that logs tends to be cyclical and overwrites after
things have been commited, journals tend to be long term archives of
changes).
as part of being able to have it run on other hardware, palo alto did
a more conventional version where explicit calls were made to log as
change happened. it turned out that the explicit calls were less
overhead than running around afterwards and finding all the modified
data. this was all before the AIXV3 with JFS actually shipped. Note
that palo alto had also done AOS (bsd) for the PC/RT as an alternative
to the AIXV2 (at&t) shipped from austin.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: Filesystems
Newsgroups: comp.sys.cbm,comp.sys.apple2,rec.games.video.classic,comp.os.cpm,alt.folklore.computers
Date: Mon, 10 Feb 2003 15:05:28 GMT
Christopher Browne writes:
Ah, but the merit of journalling filesystems is that they "play well"
even if you have a power outage or other 'forcible crash.'
Journalling provides little value if the system /doesn't/ crash.
when my wife and i started ha/cmp ... we were dependent on JFS for
fast take-over .... regardless of the failure-mode or outage.
one of the failure-modes was design for 1-800 number system. out the
back of the ss7 to the 1-800 number lookup was two T1s. They had been
going to a fault tolerant system running the database. The 1-800
number system had design point of something like five-nines ... about
5 minutes of outage per year. while the hardware was fault tolerant,
the system had to come down for software maint. .... which was on the
order of 30 mins or more (single maint. operation per year blew a
minimum of six years outage budget).
Now, it turns out that the ss7 was already fault tolerant and was
prepared to redrive the request down the alternate T1 in case it
didn't get an answer back on the initial query. We claimed that the
SS7 fault tolerant management could mask all sorts of outages
and a non-fault-tolerant ha/cmp backend was actually superior to the
fault-tolernat implementation ... since either backend could be out
for whatever reason and everything would still work ... aka an ha/cmp
configuration easily met the outage budget relying on the
fault-tolerant redrive logic in the SS7 to mask outages.
Of course it would have been possible to also replicate the
fault-tolerant boxes ... at significantly increased cost ... and then
both the ha/cmp implementation and the fault-tolerant implementation
would have met the outage budget.
random ha/cmp ref:
http://www.garlic.com/~lynn/95.html#13
lots more ha/cmp refs:
http://www.garlic.com/~lynn/subtopic.html#hacmp
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/
Internet trivia 20th anv http://www.garlic.com/~lynn/rfcietff.htm
Refed: **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **, - **
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: HASP assembly: What the heck is an MVT ABEND 422?
Newsgroups: bit.listserv.ibm-main,alt.folklore.computers
Date: Mon, 10 Feb 2003 15:38:09 GMT
dba@DUDA.COM (David Andrews) writes:
I vaguely recall (MVT 21.7 days) that my university experimented with an
initiator replacement named "Executor". It may have been associated
with WVU.
Our computer center management forced its use for certain student job
classes. They liked it because it was faster for exactly those
code-and-go ("cafeteria") jobs that Phil refers to above. But it wasn't
all that stable; it had bugs and didn't support all JCL constructs. I
don't think it would allocate permanent datasets, for example. In fact,
it may not have driven DADSM at all.
Anybody el