From: lynn@garlic.com (Lynn Wheeler) Date: 25 May, 2010 Subject: Wal-Mart to support smartcard payments Blog: Payment Systems Networkre:
another reference to interchange revenue
Wal-Mart Claims Issuers Block Progress of EMV Cards in U.S.
http://www.americanbanker.com/issues/178_55/bernanke-we-will-do-what-it-takes-to-end-too-big-to-fail-1057723-1.html
from above
Most U.S. issuers are loath to disrupt the existing model because,
according to Wal-Mart, if the EMV Integrated Circuit Card
Specifications were to become commonplace in this country, they would
lose significant revenue on signature-based credit and debit card
interchange fees.
... snip ...
with respect to x9.59 financial standard and the above referenced
rollout work for walmart a decade ago ... it would do both contact &
contactless ... as well as be able to perform a (contactless) secure
transaction within the distance, power, and elapsed time requirement
for transit turnstile. it would work with single-factor authentication
as well as mult-factor authentication (selectable based on risk and/or
transaction value)
https://www.garlic.com/~lynn/subintegrity.html#3factor
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: History: Mark-sense cards vs. plain keypunching? Newsgroups: alt.folklore.computers Date: Tue, 25 May 2010 20:59:04 -0400Peter Flass <Peter_Flass@Yahoo.com> writes:
i contributed to paper that killed 801 for the 4381 ... basically cisc chips was getting to the point that majority of 370 instructions could be implemented directly in silicon ... rather than implemented as microcode. use of 801/risc for as/400 also ran into trouble and a cisc chip was done for that also.
misc. posts mentioning 801, iliad, romp, rios, power, somerset,
power/pc, etc.
https://www.garlic.com/~lynn/subtopic.html#801
endicott had suckered me into helping with the virtual machine microcode
assist for 138/148 (precursor to 4331/4341) ... old post with some of
the studies that went into design of ecps for 138/148
https://www.garlic.com/~lynn/94.html#21 370 ECPS VM microcode assist
then they had me running around the world off and on for a year or so helping with business justifications for the machines with the various country business groups. the issue was that 370 microcode was about ten native instructions per 370 instructions ... a lot of ECPS basically moved high-use portions of kernel instructions into microcode on 1:1 basis.
in parallel with this future system was being killed
https://www.garlic.com/~lynn/submain.html#futuresys
during the height of future system, 370 efforts were terminated and 370
(software & hardware) product pipelines were allowed to go dry ... also
reference here
http://www.jfsowa.com/computer/memo125.htm
then after future system was killed, there was mad rush to get products back into the 370 pipeline. the favorite son operating system made the case that in order to make mvs/xa schedule, the virtual machine product needed to be killed, the development group in burlington mall shutdown and all the people moved to POK. somewhat because of the expanding virtual machine midrange product base ... endicott eventually managed to make the case for them to take-over the virtual machine product mission ... but they effectively had to reconstitute a development group from scratch.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Significant Bits Newsgroups: bit.listserv.ibm-main Date: 26 May 2010 07:00:47 -0700m42tom-ibmmain@YAHOO.COM (Tom Marchant) writes:
801/risc from early beginnings was single processor w/o cache
consistency support. I've periodically claimed that whole 801 was
adverse reaction to failed future system effort
https://www.garlic.com/~lynn/submain.html#futuresys
and to do the exact opposite in terms of hardware complexity. Also
along the way there were periodic observations of not wanting to pay the
significant hit on performance and thruput that standard 370
(multiprocessor) cache consistency cost. some old email mentioning 801
https://www.garlic.com/~lynn/lhwemail.html#801
and other posts mentioning 801
https://www.garlic.com/~lynn/subtopic.html#801
We did ha/cmp (High Availability Cluster Multi-Processing) cluster
scalup for rios/power because w/o cache consistency ... that was about
the only available scale-up option offered (couldn't hook up rios chips
to SCI ... since rios/power design didn't have any provisions for cache
consistency operation).
https://www.garlic.com/~lynn/subtopic.html#hacmp
Somerset was then the joint IBM, Motorola, apple, etc ... effort to
achieve a number of things
https://en.wikipedia.org/wiki/PowerPC
https://en.wikipedia.org/wiki/PowerPC_600
https://en.wikipedia.org/wiki/PowerPC_e600
https://en.wikipedia.org/wiki/PowerPC_G4
.. including single chip processor and support cache consistency for
SMP. Motorola did have a (non-801) RISC, the 88k that supported cache
consistency and had somewhat scalable cache consistency bus (and some of
somerset could be described as adapting 88k cache consistency to 801)
... and also not wanting to pay the significant hit on machine
performance that standard 370 (multiprocessor) cache consistency cost
https://en.wikipedia.org/wiki/Motorola_88000
The executive we were reporting too at the time (when we started ha/cmp) went over to headup somerset (he had also previously worked at motorola before coming to ibm).
misc. posts mentioning power & rios
https://en.wikipedia.org/wiki/IBM_POWER
The above refers to somerset as the "AIM alliance" (Apple, IBM, Motorola) ... also reference that Motorola had original tried to get Apple to upgrade their then use of 68k to Motorola's 88k risc processors.
the lore about more recent move of apple to intel chips was that power/pc work was falling way behind in doing low-power chips for laptops.
recent reference to 801 in the late 70s converging the large number of
different corporate microprocessors to 801:
https://www.garlic.com/~lynn/2010j.html#1
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Significant Bits Newsgroups: bit.listserv.ibm-main Date: 26 May 2010 13:08:01 -0700ibm-main@TPG.COM.AU (Shane Ginnane) writes:
so this was somewhat the motivation for the univ. to start the clone controller effort ... reverse engineering mainframe channel interface and building a channel interface board for an interdata/3 ... programmed to emulate 2702 ... with the addition of how "software" line-scanner on the ports which would dynamically determine bit rate (110 or 134.+).
the first bug was channel interface board wasn't releasing channel frequently enough and the 360/67 timer would redlight (i.e. it tic'ed approx 13mics and each tic had to update location 80 ... if there was still a pending location 80 update when the next timer tic came around ... it would machine check).
the next bug was overlooking 2702 line-scanner which placed leading bit in low-order bit position of byte ... filling byte in reverse order (i.e. 2702 terminal bytes were bit reverse). default ascii/tty would put bits into corresponding bit position as it came off the line. as long as things were always bit-reversed terminal world ... things were ok ... just have a fixed translate table to handle both ascii<->ebcidic translation and bit reversal. things got little more of problem when there was ascii over lan coming in whole byte (w/o bit reversal) and terminal with bit-reversed ascii bytes.
there was also some writeup blaming four of us at the univ. for clone
controller business. the interdata/3 involved into a cluster with
interdata/4 dedicated to the channel interface and one or more
interdata/3s dedicated to line-scanner function. then perkin-elmer
bought interdata ... and the box continued to live on and be sold
under the perkin-elmer brand. misc. past posts mentioning clone
controller business
https://www.garlic.com/~lynn/submain.html#360pcm
this writeup ... blames motivation for future system effort on the clone
controller business.
https://www.ecole.org/en/session/49-the-rise-and-fall-of-ibm
https://www.ecole.org/en/session/49-the-rise-and-fall-of-ibm
quote from above:
IBM tried to react by launching a major project called the 'Future
System' (FS) in the early 1970's. The idea was to get so far ahead
that the competition would never be able to keep up, and to have such
a high level of integration that it would be impossible for
competitors to follow a compatible niche strategy. However, the
project failed because the objectives were too ambitious for the
available technology. Many of the ideas that were developed were
nevertheless adapted for later generations. Once IBM had acknowledged
this failure, it launched its 'box strategy', which called for
competitiveness with all the different types of compatible
sub-systems. But this proved to be difficult because of IBM's cost
structure and its R&D spending, and the strategy only resulted in
a partial narrowing of the price gap between IBM and its rivals.
... snip ...
other past posts mentioning future system
https://www.garlic.com/~lynn/submain.html#futuresys
and then there have been a number of references that the distraction of
future system ... and policy of letting 370 product pipelines to got dry
... allowed clone processors to gain market foothold.
http://www.jfsowa.com/computer/memo125.htm
then there is reference in fergus & morris book about what happened to
corporate culture after the FS failure ... part of quote here:
https://www.garlic.com/~lynn/2001f.html#33
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Thu, 27 May 2010 08:25:28 -0400Peter Flass <Peter_Flass@Yahoo.com> writes:
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Problem Statement on the Cross-Realm Operation of Kerberos Newsgroups: alt.folklore.computers Date: Thu, 27 May 2010 16:22:27 -0400RFC 5868
Problem Statement on the Cross-Realm Operation of Kerberos
http://www.rfc-editor.org/rfc/rfc5868.txt
from above:
This document provides background information regarding large-scale
Kerberos deployments in the industrial sector, with the aim of
identifying issues in the current Kerberos cross-realm authentication
model as defined in RFC 4120.
... snip ...
Long ago we would periodically go by project athena to see what was going on (i.e. DEC & IBM both funded athena to tune of $25M and each had assist. directors of the project ... and various corporate people would also have review visits). One of the visits (at the time, the IBM assist director was person that I had worked with at the science center and had invented the compare&swap instruction) was watching the cross-realm message flows being worked on blackboard in real-time.
A few years ago I sat thru presentation on one of the first large-scale SAML cross-relm (coalition forces) deployments ... and had opportunity to comment that their messages looked identical to kerberos (modulo the bit arrangements being different).
athena wiki page
https://en.wikipedia.org/wiki/Project_Athena
kerberos wiki page
https://en.wikipedia.org/wiki/Kerberos_%28protocol%29
I was somewhat involved in the certificate-less flavor of kerberos
public-key authentication (pk-init) ... some past posts
https://www.garlic.com/~lynn/subpubkey.html#kerberos
then there was pressure to add PKI/certificate-based specification to the specification. Somewhat later, the person claiming responsibility for pushing PKI into Kerberos pk-init specification ... apologized ... finally realizing that it was redundant and superfluous in the kerberos environment.
saml wiki page
https://en.wikipedia.org/wiki/SAML_2.0
misc. past posts mentioning science center
https://www.garlic.com/~lynn/subtopic.html#545tech
and misc. posts mentioning SMP &/or compare&swap instruction
https://www.garlic.com/~lynn/subtopic.html#smp
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@garlic.com (Lynn Wheeler) Date: 27 May, 2010 Subject: Wal-Mart to support smartcard payments Blog: Payment Systems Networkre:
latest on the interchange revenue front (aka, above article reference "Wal-Mart Claims Issuers Block Progress of EMV Cards in U.S.")
Visa, MasterCard May Face Antitrust Concern on Fees, Durbin Says
http://www.businessweek.com/news/2010-05-27/visa-mastercard-may-face-antitrust-concern-on-fees-durbin-says.html
& walmart has been battling this for some time ... past reference
MasterCard Puts the 13-Year-Old Wal-Mart Case in the Rear-View Mirror
http://www.digitaltransactions.net/newsstory.cfm?newsid=2256
There was an article a couple years ago that the payment related revenue avgs. 40% of the bottom line for US institutions (some big issuers, 60%) compared to under 10% for European institutions. The scenario was that big component of payment fees are proportional to fraud ... and could mean factor of ten reduction in those fees if fraud was significantly reduced (theme was that US institutions have a lot more at stake if those fees are significantly reduced).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Seeking *Specific* Implementation of Star Trek Game Newsgroups: alt.folklore.computers Date: Fri, 28 May 2010 08:04:02 -0400Bernd Felsche <berfel@innovative.iinet.net.au> writes:
a decade ago, we were invited in to help word-smith the cal. electronic signature legislation. some of the parties involved were heavily involved in privacy issues and done detailed consumer privacy studies/survesy. the number one issue was identity theft, specifically the form of account fraud involving fraudulent transactions as result of various kinds of data breaches. very little corrective action (or even publicity) seemed to be taken ... and they seemed to think that the publicity from breach notifications might motivate corrective actions and countermeasures.
they were also working on "opt-in" privacy sharing legislation when GLBA passed with "opt-out" provision ... sort of federal pre-emption of their "opt-in" legislation. GLBA was also the bill that repealed Glass-Steagall and did some number of other things. The rhetoric on the floor of congress was major purpose of GLBA was that if you were already a bank ... you got to stay a bank, but if you weren't already a bank, you couldn't become a bank (specifically calling out wal-mart and m'soft).
even with the "opt-out" provisions in GLBA ... a few years ago at annual privacy conference in washington ... there was a panel of the FTC commissioners ... and during the discussion, somebody got up in the back of the room and said they were involved in providing call-center infrastructure for many of the financial institutions ... and they knew that the people answering the "opt-out" lines had no mechanism for recording information (aka there was no trace of people calling in requesting "opt-out" of privacy sharing). He asked if the FTC commissioners were planning on ever enforcing any of the privacy sharing legislation.
recent posts mentioning "opt-in"/"opt-out":
https://www.garlic.com/~lynn/2010b.html#82 Oldest Instruction Set still in daily use?
https://www.garlic.com/~lynn/2010b.html#94 Oldest Instruction Set still in daily use?
https://www.garlic.com/~lynn/2010b.html#95 Oldest Instruction Set still in daily use?
https://www.garlic.com/~lynn/2010h.html#73 Our Pecora Moment
https://www.garlic.com/~lynn/2010h.html#74 Idiotic programming style edicts
https://www.garlic.com/~lynn/2010h.html#75 Idiotic programming style edicts
https://www.garlic.com/~lynn/2010h.html#82 Costs Of Data Breaches Much Higher In U.S. Than In Other Countries, Study Says
https://www.garlic.com/~lynn/2010i.html#62 blasts from the past -- old predictions come true
https://www.garlic.com/~lynn/2010i.html#63 Wal-Mart to support smartcard payments
https://www.garlic.com/~lynn/2010i.html#77 Favourite computer history books?
https://www.garlic.com/~lynn/2010i.html#83 The Evolution of the Extended Enterprise: Security Stategies for Forward Thinking Organizations
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: A programming language history page. Newsgroups: alt.folklore.computers Date: Fri, 28 May 2010 10:51:51 -0400greymausg writes:
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Windows, Linux propel Q1 server sales, Unix boxes, mainframes stalled Newsgroups: bit.listserv.ibm-main Date: 28 May 2010 08:21:47 -0700Windows, Linux propel Q1 server sales, Unix boxes, mainframes stalled
from above:
... all had products in transition during the first quarter, which hurt
sales, and IBM's mainframes are expected to be upgraded in the second
half of this year, which damped sales of this legacy box.
... snip ...
related thread from early this year ... including reference to IBM
announcing new nehalem servers (being caught between rock & hard place):
https://www.garlic.com/~lynn/2010g.html#25 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#27 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#28 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#32 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#35 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#77 IBM responds to Oracle's Exadata with new systems
aka ...
IBM goes elephant with Nehalem-EX iron; Massive memory for racks and
blades
http://www.theregister.co.uk/2010/04/01/ibm_xeon_7500_servers/
from above:
With so much of its money and profits coming from big Power and
mainframe servers, you can bet that IBM is not exactly enthusiastic
about the advent of the eight-core "Nehalem-EX" Xeon 7500 processors
from Intel and their ability to link up to eight sockets together in a
single system image. But IBM can't let other server makers own this
space either, so it had to make some tough choices.
... snip ...
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@garlic.com (Lynn Wheeler) Date: 27 May, 2010 Subject: Wal-Mart to support smartcard payments Blog: Payment Systems Networkre:
Besides the rather large (POS chipcard) pilot in the US the early part of the century ... that seemed to flounder (and disappear w/o a trace) because of the yes card vulnerability (as opposed to the cost and/or fees) ... there were a number of "secure internet payment" products about the same time. They had gotten agreement from the 100 or so largest internet merchants (transactions quite skewed with over 70% of all transactions), with the merchants anticipating fees being dropped to POS PIN-debit rates. These products appeared to flounder when instead of lowered fees ... the merchants were told there would essentially be a surcharge on-top of what they were already paying (some cognitive dissonance after decades of fees proportional to fraud ... having the whole interchange fee structure thing turned on its head).
misc. past references to yes card
https://www.garlic.com/~lynn/subintegrity.html#yescard
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Information on obscure text editors wanted Newsgroups: alt.folklore.computers Date: Wed, 02 Jun 2010 08:06:30 -0400ArarghMail005NOSPAM writes:
Date: 03/14/83 08:02:33
From: wheeler
re: pc2; what do i need to do for an IBM order for two copies of the
following software for the pc2
6024002 IBM Macro Assembler 100 50
6024003 IBM BASIC Compiler 300 175
6024010 IBM Pascal Compiler 300 150
6024035 CP/M-86 Operating System
by Digital Research, Inc. 240 140
6024046 BASIC Programming Development System 130 78
6024048 Professional Editor 130 78
6024050 Diskette Librarian 45 27
6024051 Personal Editor 100 60
6024053 BASIC Primer 60 36
6024061 DOS 2.0 60 40
C 6025000 Guide to Operations Manual for the
IBM Personal Computer 49.50 29.75
6025005 Technical Reference Manual for the
IBM Personal Computer 36 21.50
... snip ... top of post, old email index
I actually didn't make much use of Professional Editor ... but made quite a bit of use of Personal Editor. Later on AT with more memory, started using KEDIT ... a PC clone of the mainframe XEDIT.
this claims personal editor lives on as PE32:
http://www.pe32.com/
and KEDIT still lives.
http://www.kedit.com/
The original CMS editor tooks lot of its syntax from CTSS ... and the implementation stream data from original file, thru memory and out to working file on disk ... then subsequent operations streamed back and forth between two working files ... until saved; when newest working file was renamed the original file.
As undergraduate in 68, this was the editor that I hacked the CMS 2250 display fortran library (from lincoln labs) as editor display/output driver ... to get early full-screen editor.
An enhanced CMS editor was then done that would load file into virtual memory (assuming it would fit, the "old" editor was renamed cedit and continued to be used for files that wouldn't fit in virtual memory).
At the univ. I had added ascii/tty terminal support to the existing
2741 & 1052 support in cp67. I've mentioned before that I tried to
get 2702 controller to do something that it couldn't quite do ... which
was motivation to start clone controller effort ... some past posts
https://www.garlic.com/~lynn/submain.html#360pcm
I also modified a version of HASP on MVT release18 ... removing the 2780 RJE support code (to limit program image size) and replacing it with 2741 & tty terminal support ... along with an editor that implementated the CMS editor syntax (I implemented it from scratch since the HASP execution environment was quite different from CMS). This was used for crje operation ... predating TSO.
The standard CMS editor was initially adopted to 3270 as full-screen display ... but still had command line and couldn't move around screen to modify data. EDGAR was the first wide-spread 3270 CMS editor that support full-screen operations.
Internally there was RED that saw quite a bit of use ... before XEDIT
came along (I participated in trying to get RED released as product in
place of XEDIT ... because at the time, it was much more mature and had
more function). A few past posts mentioning RED
https://www.garlic.com/~lynn/2002j.html#4 HONE, ****, misc
https://www.garlic.com/~lynn/2002p.html#39 20th anniversary of the internet (fwd)
https://www.garlic.com/~lynn/2006n.html#55 The very first text editor
https://www.garlic.com/~lynn/2006u.html#26 Assembler question
https://www.garlic.com/~lynn/2006u.html#61 Why these original FORTRAN quirks?
https://www.garlic.com/~lynn/2006y.html#4 Why so little parallelism?
with old email
https://www.garlic.com/~lynn/2006u.html#email781103
https://www.garlic.com/~lynn/2006u.html#email790523
https://www.garlic.com/~lynn/2006u.html#email800311
https://www.garlic.com/~lynn/2006y.html#email800311
https://www.garlic.com/~lynn/2006u.html#email800312
https://www.garlic.com/~lynn/2006y.html#email800312
https://www.garlic.com/~lynn/2006u.html#email800429
https://www.garlic.com/~lynn/2006n.html#email810531
https://www.garlic.com/~lynn/2002p.html#email821122
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@garlic.com (Lynn Wheeler) Date: 2 June, 2010 Subject: Warren Buffett faces hearing over ratings agencies Blog: Financial Crime Risk, Fraud and SecurityWarren Buffett faces hearing over ratings agencies; Berkshire Hathaway, which Warren Buffett leads as chairman and chief executive, is Moody's largest shareholder.
from above:
Billionaire investor Warren Buffett and the CEO of credit rating
agency Moody's are scheduled to face questions Wednesday from a
bipartisan panel probing the roots of the financial crisis.
... snip ...
So far there seems to be little reference to the congressional hearings from fall of 2008 into the rating agencies ... when the testimony was that the rating agencies were being paid for triple-A ratings on toxic CDOs when both the sellers and the agencies knew they weren't worth triple-A ratings. Also that the seeds for conflict of interest and misaligned business process were sown in the early 70s when the agencies switched from buyers paying for the ratings to sellers paying for the ratings
archived recent news item thread mentioning rating ageinces
https://www.garlic.com/~lynn/2010i.html#48 Fraud & Stupidity Look a Lot Alike
at the time of the 2008 congressional hearings, there was news items that the rating agencies were probably immune to criminal prosecution as one of the primary players in the financial mess ... because of the risk that they could lower the govs. credit rating.
Answers on Credit Ratings Long Overdue
http://www.nytimes.com/2010/06/01/business/01sorkin.html?ref=business
from above:
One of the enduring questions of the financial crisis is how the
credit ratings establishment got so much so wrong for so very
long. How could century-old institutions like the Moody's Investors
Service give their triple-A blessings to subprime junk?
... snip ...
old nyt reference to fall 2008 hearings
Rating Agencies Draw Fire on Capitol Hill
http://dealbook.nytimes.com//2008/10/22/rating-agencies-draw-fire-capitol-hill/
from above:
"Total revenues for the three firms doubled from $3 billion in 2002 to
over $6 billion in 2007," Mr. Waxman said. "Moody's had the highest
profit margin of any company in the S&P 500 for five years in row."
... snip ...
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: A "portable" hard disk Newsgroups: alt.folklore.computers Date: Wed, 02 Jun 2010 12:35:56 -0400Morten Reistad <first@last.name> writes:
there is a concret lip along the bottom of the spillway ... to keep the falling water from undercutting the ground under the dam ... especially before the 3rd powerhouse was built (in theory, nearly all the water can now go thru turbines and the spillway isn't really needed much any more) ... that concete splash guard had to be be repaired every couple years.
they use to have overhead lines from the generators to switch yard above the dam (out of view on hill to the right in the picture in wiki). in the 60s, the president's wife was having a beautification effort ... like eliminating billboards from interstates. Part of that involved eliminating the overhead transmission lines from the powerhouses to the switchyard. Somebody put together plan for buried pipes that contained the power lines (carefully located in the center of the pipes). Engineers protested that the (current) technology wouldn't work ... that gravity would result in the mass of the transmission cores "slumping" ... and touching the side of the pipewalls ... resulting in short/arc'ing. The engineers were overruled because it was mandated by the beautification efforts. A couple years later ... when there was big fire with the wires in the pipes slumping ... the engineers that objected ... were the ones blamed (wasn't PC to blame the president's wife).
a little more on turbines at the dam:
http://users.owt.com/chubbard/gcdam/html/hydro.html
the above mentions the reversable pumps ... that started out pumping water into the coulee for irrigation purposes ... but now also be used to pump water during off-peak into the coulee ... and then reverse the flow for power generation during peak hrs.
a couple past posts mentioning grand coulee/hydro-electric power
https://www.garlic.com/~lynn/2002n.html#43 VR vs. Portable Computing
https://www.garlic.com/~lynn/2003d.html#32 IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM
https://www.garlic.com/~lynn/2007o.html#14 Geothermal was: VLIW pre-history
https://www.garlic.com/~lynn/2008e.html#7 was: 1975 movie "Three Days of the Condor" tech stuff
https://www.garlic.com/~lynn/2008r.html#56 IBM drops Power7 drain in 'Blue Waters'
https://www.garlic.com/~lynn/2008n.html#68 VMware Chief Says the OS Is History
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Multiprise 3k for personal Use? To: <ibm-main@bama.ua.edu> Date: Thu, 03 Jun 2010 12:38:52 -0400dba@LISTS.DUDA.COM (David Andrews) writes:
she had done Peer-Coupled Shared Data architecture
https://www.garlic.com/~lynn/submain.html#shareddata
which saw little uptake (except for ims hot-standby) until sysplex. other battles that contributed to her not staying long in the position was SNA camp trying to force all loosely-coupled operations thru VTAM.
prior to taking the position in POK ... she had been in the JES group working on merged JES2/JES3 (figuring out what were the missing things in one ... that the customers of the other couldn't live w/o) ... JES Ultimate System.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Sat, 05 Jun 2010 11:02:52 -0400jmfbahciv <See.above@aol.com> writes:
some of the stand-alone evolved into small business lans ... again w/o defenses and/or countermeasures ... with lots of enhanced applications with scripting and sharing.
multi-user/timesharing systems tended to have defenses and
countermeasures between the different parties ... especially
when operating as commercial service bureaus. some number of
the virtual machine based service bureaus had moved upstream
into financial information ... and had clients from competitive
organizations ... where security and preventive measures became
extremely important. some past posts
https://www.garlic.com/~lynn/submain.html#timeshare
there is an issue now where the world-wide internet is an extremely anonomous hostile environment (requiring significant defenses and countermeasures) ... totally different from the early relatively safe, small, closed business lans.
there is some similarity in the technologies. the problem is taking the stand-alone desktop and small, closed, safe business lans ... and providing internet connectivity ... w/o building in fundamental defenses and countermeasures (including never allowing arbitrary executiion of things arriving over the network interface ... something that had become default in the safe, closed, small business lan environment).
there is significant difference starting with a platform that assumes defenses and countermeasures as part of original design ... as opposed to starting with a platform that makes no assumptions about defenses and countermeasures ... and then attempting to adapt it to hostile environment (say starting with a family automobile and using it for interplanetary transport).
I had above discussion a couple times ... when Jim badgered me into
interviewing for position of chief security architect in redmond ...
some past references
https://www.garlic.com/~lynn/2007o.html#7 Hypervisors May Replace Operating Systems As King Of The Data Center
https://www.garlic.com/~lynn/2008b.html#5 folklore indeed
https://www.garlic.com/~lynn/2008b.html#37 Tap and faucet and spellcheckers
https://www.garlic.com/~lynn/2008p.html#80 Making tea
https://www.garlic.com/~lynn/2009.html#60 The 25 Most Dangerous Programming Errors
https://www.garlic.com/~lynn/2009g.html#18 Top 10 Cybersecurity Threats for 2009, will they cause creation of highly-secure Corporate-wide Intranets?
https://www.garlic.com/~lynn/2009h.html#28 Computer virus strikes US Marshals, FBI affected
https://www.garlic.com/~lynn/2009i.html#22 My Vintage Dream PC
https://www.garlic.com/~lynn/2009l.html#20 Cyber attackers empty business accounts in minutes
https://www.garlic.com/~lynn/2010g.html#66 What is the protocal for GMT offset in SMTP (e-mail) header
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: What is the protocal for GMT offset in SMTP (e-mail) header Newsgroups: comp.mail.misc, alt.folklore.computers, comp.os.linux.advocacy Date: Sat, 05 Jun 2010 17:02:16 -0400John Levine <johnl@iecc.com> writes:
for organization that had large budget for already justified 3270s, it was a relative no-brainer to switch from real 3270s to PC with 3270 emulation; the costs were about the same, and for that (in a single desktop footprint) ... got 3270 terminal and some local personal computing (much simpler than going for business justification for incremental budget for PC based purely on emerging personal computing ... and having both a PC and a 3270 on the same desk).
misc. past posts mentioning 3270 emulation
https://www.garlic.com/~lynn/subnetwork.html#emulation
in the early days ... 3270 emulation help with PC uptake .... later on the communication business unit was trying to preserve the terminal emulation install base ... attempting to fend off all kinds of enhanced PC operation ... like trying to prevent client/server taking hold.
In that later period, we had come up with 3-tier architecture and were
out pitching to customer executives ... and taking barbs from the
communication interests (part of trying to preserve the terminal
emulation install base).
https://www.garlic.com/~lynn/subnetwork.html#3tier
Also in that period, one of the senior mainframe disk engineers had gotten a presentation scheduled at the world-wide annual internal communication conference ... and started out their talk with the statement that the communication group was going to be responsible for the demise of the disk division. The scenario was that several products that the disk division had attempted to announce ... supporting advanced use of mainframe storage in distributed environments ... were being blocked by the communication business unit (part of preserving their terminal emulation install base). The disk division showed that w/o such products ... their products would continue to show annual declines.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main Date: 6 Jun 2010 20:23:49 -0700eamacneil@YAHOO.CA (Ted MacNEIL) writes:
misc. past posts mentioning the subject:
https://www.garlic.com/~lynn/99.html#58 When did IBM go object only
https://www.garlic.com/~lynn/2002q.html#32 Collating on the S/360-2540 card reader?
https://www.garlic.com/~lynn/2002q.html#48 myths about Multics
https://www.garlic.com/~lynn/2002q.html#49 myths about Multics
by comparison cp67 and then vm370 not only shipped source ... but provided source maintenance ... fixes were shipped as source from which customer did new system rebuilds (there was significant retrenching with the OCO-wars).
cp67 was used for commercial timesharing service bureaus ... some of
which moved up the value stream providing financial information ...
getting customers from different, competitive large wall street firms
... where critical competitive information was frequently involved.
misc. references to virtual machines based timesharing (bears some
similarities with current cloud computing):
https://www.garlic.com/~lynn/submain.html#timeshare
science center also got into provide timesharing services ... both internally as well as to educational institutions in the cambridge area (students and other non-employees). The science center has also done a port of apl\360 to cms for cms\apl ... opening up workspace size to virtual memory limits (from typical 16k-32k bytes) and adding APIs for system facilities (like read/write files) ... which allowed using APL for large real-world applications. One of the early internal customers for the services was the business planning people in Armonk ... which loaded the most valuable of all corporate information on the system (for their APL business planning models). This assumed that there was significant security given the most valuable of all corporate resources on the same system with a lot of univ. students.
a couple recent posts mentioning providing services for armonk business
planners:
https://www.garlic.com/~lynn/2010e.html#24 Unbundling & HONE
https://www.garlic.com/~lynn/2010i.html#66 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
also slightly related ...
https://web.archive.org/web/20090117083033/http://www.nsa.gov/research/selinux/list-archive/0409/8362.shtml
i didn't hear about the above customers until much later. other recent
posts mentioning the above url:
https://www.garlic.com/~lynn/2010b.html#63 Source code for s/360 [PUBLIC]
https://www.garlic.com/~lynn/2010b.html#97 "The Naked Mainframe" (Forbes Security Article)
https://www.garlic.com/~lynn/2010d.html#62 LPARs: More or Less?
https://www.garlic.com/~lynn/2010f.html#59 More calumny: "Secret Service Uses 1980s Mainframe"
https://www.garlic.com/~lynn/2010f.html#74 Is Security a Curse for the Cloud Computing Industry?
https://www.garlic.com/~lynn/2010g.html#9 Far and near pointers on the 80286 and later
https://www.garlic.com/~lynn/2010g.html#40 someone smarter than Dave Cutler
https://www.garlic.com/~lynn/2010g.html#53 Far and near pointers on the 80286 and later
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main Date: 7 Jun 2010 06:35:30 -0700cfmpublic@NS.SYMPATICO.CA (Clark Morris) writes:
the question of security thru obscurity comes up periodically in relation to provable cryptography ... usely related to terms like snake-oil.
open infrastructures tend to have faults identified and corrected much earlier ... there can be a painful period if attempting to transition from obscurity to open ... since all sorts of hidden infections could be found festering ... taking some period to clean out.
we were tangentially involved in the cal. state data breach
notification legislation. we had been brought in to help wordsmith the
electronic signature legislation and some of the parties were also
heavily involved in privacy issues. they had done, detailed consumer
surveys on privacy. the number one issue was identity theft, a major
component being account fraud ... where skimmed/breached information
was being used for fraudulent financial transactions. there appeared
to be little or nothing being done about breaches ... and so it seemed
that they felt the resulting publicity from data breach notifications
would provide motivation to taking corrective action and
countermeasures.
https://www.garlic.com/~lynn/subpubkey.html#signature
at the time, they were also working on opt-in privacy sharing legislation, but then GLBA (possibly better known for repeal of Glass-Steagall) came out with opt-out privacy sharing ... sort of federal pre-emption of the cal. work in progress (the difference between opt-in/opt-out privacy sharing has been in the news recently).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main, alt.folklore.computers Date: Mon, 07 Jun 2010 17:01:41 -0400re:
and for some related topic drift (in a.f.c):
https://www.garlic.com/~lynn/2010j.html#15 Idiotic programming style edicts
late 90s, there was folklore that ceo of large financial institution let their CSO go with comment that fraud was much more cost effectively handled out of the public relations department (than the security department).
some of this could be considered short-term horizon associated with quarterly financial filings of public companies (minimize infrastructure investment and maintenance ... moving the funds into the profit column and used for executive bonuses ... planing on being long gone by the time the infrastructures begin collapsing).
maintaining the facade also showed up (at least) in financial sector
infrastructure protection meetings
https://en.wikipedia.org/wiki/Critical_infrastructure_protection
https://en.wikipedia.org/wiki/National_Information_Infrastructure_Protection_Act
at the time a lot of the infrastructure protection meetings were with
regard to y2k remediation ... however, there were also issue of
information sharing (ISAC) databases ... capturing exploits, threats
and vulnerabilities. the initial push-back was fear that ISACs would
be subject to FOIA ... when the FOIA issues were addressed ... there
was still pushback that information about exploits, threats and
vulnerabilities (at least in much of the financial industry) was
viewed as competitive advantage (who had them, who didn't, who knew
about them, and what were the countermeasures; even when the
information was not available publicly ... limited to industry
insiders and appropriate law enforcement agencies).
http://www.fsisac.com/
http://www.isaccouncil.org/
the bread&butter of the financial sector ... possibly much more than the other sectors, has been trust ... with lots of concern about publicity damaging their reputation. unfortunately this can result in coming to believe that managing information about exploits can replace/substitute actually doing something about exploits.
misc. past posts mentioning ISAC/FOIA
https://www.garlic.com/~lynn/2007i.html#17 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007u.html#0 folklore indeed
https://www.garlic.com/~lynn/2008g.html#11 Hannaford case exposes holes in law, some say
https://www.garlic.com/~lynn/2008m.html#82 Data sharing among Industry players about frauds
https://www.garlic.com/~lynn/2009f.html#48 Bankers as Partners In Crime Stopping
https://www.garlic.com/~lynn/2009n.html#11 Banks should share cyber crime information
https://www.garlic.com/~lynn/2009p.html#27 FBI: National data-breach law would help fight cybercrime
https://www.garlic.com/~lynn/2009p.html#45 ATM machines are increasingly attractive to hackers
scenario regarding OCO-wars (aka Object-Code-Only wars) ... sort of goes starting to charge for (23jun69 unbundling) application software in response to various litigation ... but making case that system/kernel software was still free; Future System project started (a least partially motivated by clone controllers); Future System distraction is credited with allowing clone processors to gain market foothold; change starting to charge for system/kernel software (at least partially motivated by clone processors gaining market foothold) ... then followed by OCO.
misc. past posts mentioning unbundling ... first 23jun69 announcement
for application software ... then later change starting to charge for
system/kernel software (my resource manager was selected as guinea pig
for starting to charge for system/kernel software)
https://www.garlic.com/~lynn/submain.html#unbundle
misc. past posts mentioning Future System
https://www.garlic.com/~lynn/submain.html#futuresys
misc. past posts mentioning clone controller
https://www.garlic.com/~lynn/submain.html#360pcm
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main, alt.folklore.computers Date: Tue, 08 Jun 2010 08:54:34 -0400cfmpublic@NS.SYMPATICO.CA (Clark Morris) writes:
aka, during the OCO-wars ... in the transition from freely available source to object-code-only ... I don't remember being able to hide threats and vulnerabilities being an argument ... it was about protecting corporate property (i.e. source) in a competitive environment with clone processors.
starting to charge for application software (23jun69 unbundling
announcement) was about various litigation ... but case had been made
that kernel/system software would still be free. later decision to start
charging for kernel software was in period when clone processors had
gained market foothold (during FS distraction, and my resource manager
was initial guinea pig for kernel software charging);
https://www.garlic.com/~lynn/submain.html#unbundle
OCO could be construed as further market inhibitors (in addition to software no longer free).
sometimes (in OCO-wars) there were issues raised about protecting customers from themselves ... that freely available source encourages customer programmers to make modifications ... which would cause problems/delays in moving to newer releases (things like newer source was incompatible with older source). customer source modifications could also result in delays in replacing existing machines with newer machines (that might have various kinds of differences).
there was case where AT&T had gotten a highly modified versions of early
csc/vm system (w/o multiprocessor support) ... old csc/vm email
reference (long before OCO-wars, still when vm370 shipped with full
source maintenance):
https://www.garlic.com/~lynn/2006v.html#email731212
https://www.garlic.com/~lynn/2006w.html#email750430
AT&T then made a large number of their own source modifications (things like virtual device support that ran over network connections ... aka being able to run application at one AT&T facility ... thinking it was doing i/o to local tape drive ... but tape drive was actually connected to system at another AT&T facility) ... which was widely distributed/used within AT&T.
Nearly a decade later, the national account manager for AT&T tracked me down looking for help in moving AT&T off that csc/vm system to a more current vm370. This was related to 3081 ... which was only going to be available in multiprocessor configuration ... and there was not going to be a non-multiprocessor (although this was later modified to ship 3083 ... in large part because ACP/TPF didn't have multiprocessor support). Since that particular csc/vm system (w/o multiprocessor support) was so entrenched in AT&T ... they were going to be forced to going to clone processor vendor that was selling newer uniprocessor machines (early csc/vm systems didn't have multiprocessor support until after the version that had escaped to AT&T; except for version that escaped to AT&T ... my csc/vm systems were limited to large number of internal installations ... which I could keep current).
misc. recent posts mentioning 3083
https://www.garlic.com/~lynn/2010.html#1 DEC-10 SOS Editor Intra-Line Editing
https://www.garlic.com/~lynn/2010.html#21 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010d.html#14 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010d.html#79 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#23 Item on TPF
https://www.garlic.com/~lynn/2010i.html#24 Program Work Method Question
https://www.garlic.com/~lynn/2010i.html#78 IBM to announce new MF's this year
other reference to 3081 (& future system)
http://www.jfsowa.com/computer/memo125.htm
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Tue, 08 Jun 2010 09:55:53 -0400jmfbahciv <See.above@aol.com> writes:
at least part of closing was supposedly related to mass. delegation always voting against military appropriations (if the end of cold war needed to have base closing/consolidation ... and if choice of two bases came down to all other issues being the same ... then that was as good a tie-breaker as any).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main, alt.folklore.computers Date: Tue, 08 Jun 2010 10:09:16 -0400Anne & Lynn Wheeler <lynn@garlic.com> writes:
also csc/vm email
https://www.garlic.com/~lynn/2006w.html#email750102
jan75, a couple engineers from POK came up to science center to talk about doing a 5-way SMP skunkworks effort.
in the morph from cp67 to vm370 ... there was a lot of simplification and dropping of code ... which accounted for large part of the effort to move the cp67 csc/vm system to a vm370 base. I did get a bunch of fastpath stuff put back in (that I had originally done as undergraduate on cp67 in 1968) which shipped in vm370 release 1plc9 (aka vm370 had monthly source maintenance mini-releases that were called plc or program level change).
in any case, spring of '75, they roped me into helping with 5-way SMP
skunkworks effort called VAMPS ... which was eventually killed w/o even
being announced ... some past posts
https://www.garlic.com/~lynn/submain.html#bounce
I got to do a lot of microcode/machine design ... queued i/o and queued i/o termination (something similar showed up later in "811" ... internal codename for 370xa). I also got to do multiprocessor dispatching interface ... somewhat similar to what showed up later in intel432 (but in microcode rather than silicon ... the i432 group gave a talk about one of the things that help kill i432 was putting really complex stuff into silicon ... and then difficulty in shipping fixes/patches).
after VAMPS was killed ... one or two of the people from VAMPS helped form another smp skunkworks effort for 16-way smp. this got killed and some people invited to never appear in POK again, when the head of POK was told that it might be decades before the POK favorite son operating system had (effective) 16-way support.
misc. past posts mentioning SMP (&/or compare&swap instruction):
https://www.garlic.com/~lynn/subtopic.html#smp
misc. recent posts mentioning charlie inventing compare&swap instruction
(compare&swap was chosen because CAS are charlie's initials):
https://www.garlic.com/~lynn/2010b.html#67 How long for IBM System/360 architecture and its descendants?
https://www.garlic.com/~lynn/2010c.html#47 Extracting STDOUT data from USS
https://www.garlic.com/~lynn/2010d.html#20 search engine history, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010e.html#15 search engine history, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010g.html#80 What is the protocal for GMT offset in SMTP (e-mail) header time-stamp?
https://www.garlic.com/~lynn/2010h.html#86 Itanium had appeal
https://www.garlic.com/~lynn/2010i.html#31 IBM Unix prehistory, someone smarter than Dave Cutler
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Ten years of IBM mainframe Linux Newsgroups: bit.listserv.ibm-main Date: 8 Jun 2010 13:59:05 -0700Ten years of IBM mainframe Linux
from above:
While IBM's System z, aka mainframes, revenue fell 17%, a billion bucks
or so of business still isn't anything to sneeze at. So what happened
to give the mainframe a new lease on life? In a word: Linux.
... snip ...
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Ten years of IBM mainframe Linux Newsgroups: bit.listserv.ibm-main Date: 8 Jun 2010 17:22:11 -0700re:
misc other past items in the same vein
IBM Enterprise Linux Server on z as mainframe savior
http://dancingdinosaur.wordpress.com/2009/12/13/ibm-enterprise-linux-server-on-z-as-mainframe-savior/
from above:
Clearly, IBM has identified Linux on System z as the most immediate
growth opportunity for the mainframe. To that end, it is willing to
slash prices to capture market share against the likes of HP and Sun
in the enterprise Linux segment.
... snip ...
10th Anniversary of Linux for the Mainframe: Beginning to Today
http://www.eweek.com/c/a/Linux-and-Open-Source/10th-Anniversary-of-Linux-for-the-Mainframe-Beginning-to-Today/
from above:
Then, in late 2008, IBM created new pricing for the new System z10
Business Class mainframe to make it more economical to use. This
amounted to about a 40 percent discount on Linux for System z
subscriptions for the System z10 Business Class servers. IBM also
dropped its prices for IFLs by more than 50 percent.
... snip ...
Mainframe Sales on the Rise, Why?
http://opseast.wordpress.com/2007/08/31/mainframe-sales-on-the-rise-why/
from above:
Linux, specifically SUSE Linux Enterprise, that's why.
... snip ...
Linux on IBM System z
http://www-03.ibm.com/systems/z/os/linux/
10 years of Enterprise Linux on System z; A simple idea that
changed the world
http://www-03.ibm.com/systems/z/news/announcement/20100517_annc.html
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Wed, 09 Jun 2010 10:02:34 -0400jmfbahciv <See.above@aol.com> writes:
or maybe not so huge (aka c5a) ... just look that way flying low ...
possible more like c-130 ... saw one a couple weeks ago at blue angels
performance ... "fat albert"
https://en.wikipedia.org/wiki/File:C-130T_Hercules_Blue_Angels.jpg
fat albert airlines
http://www.blueangels.navy.mil/fat_albert.htm
wiki page
https://en.wikipedia.org/wiki/C-130_Hercules
c-17 & c-130 reference
http://www.baseops.net/basictraining/airborne.html
Ft. Devens history
http://www.fortdevensmuseum.org/history.htm
from above:
Despite a valiant political effort to keep it open, Fort Devens closed
in 1996, under the national Base Realignment And Closing Act.
... snip ...
wiki ... includes reference to army air field:
https://en.wikipedia.org/wiki/Fort_Devens
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Wed, 09 Jun 2010 12:38:52 -0400Huge <Huge@nowhere.much.invalid> writes:
first business trip to germany, the locals were apprehensive about my being able to manage a rental car ... since I was american and none of the rental cars had automatic transmission (also issue about staying in local hotel where nobody but the children spoke any english).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: A "portable" hard disk Newsgroups: alt.folklore.computers Date: Wed, 09 Jun 2010 12:44:21 -0400Patrick Scheible <kkt@zipcon.net> writes:
BPA wiki
https://en.wikipedia.org/wiki/Bonneville_Power_Administration
BPA
http://www.bpa.gov/corporate/
combination of the hydro and hydro-electric power supposedly now accounts for several mega-datacenters in the region
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@garlic.com (Lynn Wheeler) Date: 09 June, 2010 Subject: The Durbin Amendment Ignites a Lobbying Frenzy on Capitol Hill Blog: Payment Systems NetworkThe Durbin Amendment Ignites a Lobbying Frenzy on Capitol Hill
from above:
The lobbying efforts come in the wake of another research report that
warns of unintended consequences from debit card price regulation and
estimates that government-benefit prepaid card programs could lose
more than $140 million in revenue should the amendment become law.
... snip ...
There was something similar when community banks were rallied to oppose walmart obtaining an ILC to become its own acquirer ... effectively eliminating the walmart acquiring interchange fees for a few very large institutions ... and wouldn't have impacted the community banks primarily issuing interchange fees. The scenario supposedly was that being its own acquirer was just a small step on the way to becoming an issuer.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: A "portable" hard disk Newsgroups: alt.folklore.computers Date: Wed, 09 Jun 2010 17:13:04 -0400re:
this talks about history of boeing, power, and aircraft production
in ww2:
http://www.taphilo.com/history/WWII/USAAF/Boeing/B17/index.shtml
more about columbia river hydropower and aluminum
http://www.nwcouncil.org/history/Aluminum.asp
this is earlier history ... including 1916 seaplane assembly in Lake
Union boathouse (merged with mcdonnel/douglas history):
http://www.boeing.com/history/chronology/chron01.html
other boeing history
http://www.centennialofflight.gov/essay/Aerospace/boeing-early/Aero17.htm
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Tube Talk 1952 computer design Newsgroups: alt.folklore.computers Date: Fri, 11 Jun 2010 09:33:16 -0400Peter Flass <Peter_Flass@Yahoo.com> writes:
recent thread/post about 709 cooling
https://www.garlic.com/~lynn/2010c.html#78 SLIGHTLY OT - Home Computer of the Future (not IBM)
other past posts mentioning cooling for the univ.'s 709
https://www.garlic.com/~lynn/94.html#53 How Do the Old Mainframes
https://www.garlic.com/~lynn/2000c.html#39 WHAT IS A MAINFRAME???
https://www.garlic.com/~lynn/2000f.html#74 Metric System (was: case sensitivity in file names)
https://www.garlic.com/~lynn/2003j.html#55 Origin of "Function keys" question
https://www.garlic.com/~lynn/2004f.html#49 can a program be run withour main memory?
https://www.garlic.com/~lynn/2009d.html#55 1401's in high schools?
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Tube Talk 1952 computer design Newsgroups: alt.folklore.computers Date: Fri, 11 Jun 2010 09:36:10 -0400Joe Pfeiffer <pfeiffer@cs.nmsu.edu> writes:
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main Date: 11 Jun 2010 07:39:30 -0700shmuel+ibm-main@PATRIOT.NET (Shmuel Metz , Seymour J.) writes:
they would needle me that the fixes were such that if I was left alone in the datacenter with the machine ... that even I couldn't access the FS documentation (part of this may have been in response to various unflattering comments that I had been making about the FS effort).
one friday afternoon when I was visiting to setup for some offshift dedicated time ... they got somewhat irritating about the subject ... which prompted me to reply it would take less than five minutes and involve changing one byte.
I first had to disable all access to the machine from other than the operators console. I then used the hardware console to patch a branch instruction in the kernel password checking routine ... so that everything entered was treated as valid password. I pointed out that countermeasure would require something like service console passwords for access to hardware functions (like display/alter storage).
misc. past posts mentioning Future System effort
https://www.garlic.com/~lynn/submain.html#futuresys
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Personal use z/OS machines was Re: Multiprise 3k for personal Use? Newsgroups: bit.listserv.ibm-main Date: 11 Jun 2010 08:22:15 -0700re:
it was also about the time that the corporation hired a new CSO, long distinguished career in gov. ... things like having been head of presidential detail; knew a lot about physical security. I got asked to run around with him some; supposedly the corporate computer/information security expert (a few details about physical security would rub off).
other posts in this thread:
https://www.garlic.com/~lynn/2010j.html#14 Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#17 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#18 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#19 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#20 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#22 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Sat, 12 Jun 2010 08:31:39 -0400jmfbahciv <See.above@aol.com> writes:
commerical online timesharing would recover their (billed) operational expenses by charge for use ... normally with off-shift usage billed at lower rate (to help encourage off-shift use) ... but early on, off-shift use could be extremely low & sporadic ... but system must be left available 7x24 for possible incoming offshift dialin calls.
system meter would run whether processor and/or i/o was "active". normally just leaving an active i/o on dial-in port ... to answer possible call ... would be sufficient to keep the system meter running. hack was to work on an i/o sequence that would allow system meter to stop ... but wakeup if there was incoming call (system available to be active when it was being used ... but system meter wasn't running whenever everything was quiet).
misc. past posts mentioning online commercial timesharing operation
https://www.garlic.com/~lynn/submain.html#timeshare
the system meter had characteristic that it would continue running (coast) for 400millisecond after everything had gone quiet. in the 70s, the pok favorite son operating system included an operation that wokeup and ran at least every 400milliseconds ... even tho by that time, most customers had moved to purchased machines (the wakeup feature was possible design left over from earlier times).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: IBM Rational Developer for System z Newsgroups: bit.listserv.ibm-main Date: 12 Jun 2010 05:57:28 -0700john_w_gilmore@MSN.COM (john gilmore) writes:
in the 60s, most systems were leases and monthly charges were based on number of shifts recorded by the system meter. the system meter would run whenever the processor and/or i/o was active ... and would continue to coast for 400milliseconds after everything had gone quiet (i.e. everything had to be quiet for at least 400milliseconds before the system meter actually stopped).
one of the challenges for cp67 use in 7x24 operation for online commercial timesharing service bureaus was figuring out an i/o programming hack for terminal connections and dialup so that system was available for incoming characters/connections but the system meter wouldn't being running if nothing was actually happening (commercial timesharing service bureaus recovered their operational expenses from use charges ... encouraging offshift use had to leave system available 7x24 ... but also minimize operational expenses ... like "system meter" ... during those periods when usage tended to be low; another area of reducing off-shift expenses was drastically reducing/eliminating requirement for human operator).
however, the pok favorite son operating system had something that would wakeup every 400milliseconds, even when nothing else was going on (this continued long after customers had been converted to purchased systems); aka the only way to stop the system meter would be to manually push the STOP buttom (or shutdown).
misc. past posts mentioning virtual machine based online commercial
timesharing service bureaus
https://www.garlic.com/~lynn/submain.html#timeshare
this has some overlap with recent thread mentioning that those commercial online operations also needed fairly high level of security ... since there was a lot of open use ... even from fierce business competitors, using the same machine concurrently.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Favourite computer history books? Newsgroups: alt.folklore.computers Date: Sat, 12 Jun 2010 16:33:54 -0400spring 96 MDC at moscone sort of was turning point. it had lots of discussion about supporting internet ... but the repeated theme & banners were "preserve your investment". this was all the people that had been doing development in various flavors of basic ... adding all kinds of scripting capability to various applications ... and that capability would be preserved as things moved to the internet (which is the root of enormous amount of current threats, vulnerabilites, and exploits).
the private discussions were that up until then ... new releases always came out with lots of new features that people had been waiting for ... but at the turning point, apps had 95+% of the features that 95+% of the people normally used. the issue was moving the market to new reasons to continue to always buy the newest release (sort of the '60s motivating consumers to buy a new car every year ... whether it was needed or not).
some past posts mentioning '96 mdc at moscone
https://www.garlic.com/~lynn/2003e.html#45 Computer programming was all about:
https://www.garlic.com/~lynn/2003h.html#22 Why did TCP become popular ?
https://www.garlic.com/~lynn/2004b.html#34 Next generation processor architecture?
https://www.garlic.com/~lynn/2009q.html#43 The 50th Anniversary of the Legendary IBM 1401
https://www.garlic.com/~lynn/2010c.html#63 who pioneered the WEB
https://www.garlic.com/~lynn/2010h.html#37 (slightly OT - Linux) Did IBM bet on the wrong OS?
m'soft had hired some old internet hands to do their tcp/ip and several were at moscone ... and they assured me that their browser had multiple A-record support.
we had been brought in to consult with small client/server startup that wanted to do payment transactions on their server ... they had also invented something they called SSL they wanted to use ... and the result is now frequently called "electronic commerce".
Part of the deployment was something called "payment gateway" ... that
sat on the internet and acted as gateway between webservers and the
payment transaction network. The payment gateway had multiple links into
the internet backbone locations for availability ... and initially was
planning on advertising routes. However, about that time, backbone
infrastructure made decision to start migrating to (ip-address)
hierarchical routing ... which pretty much left multiple A-record
support as the primary availability mechanism. I could mandate that all
webservers implement multiple A-record support for connections to the
payment gateway. misc. past posts mentioning "payment gateway"
https://www.garlic.com/~lynn/subnetwork.html#gateway
However, didn't have equivalent authority over browser implementation (for supporting high-availability webserver configurations). The browser group pushback was that multiple A-record support was too advanced ... even when I provided them samples of client implementations from BSD 4.3 tahoe distribution. I was still having periodic skirmishes with their browser group when MDC at moscone rolled around.
misc. past posts mentioning multiple A-record support:
https://www.garlic.com/~lynn/96.html#34 Mainframes & Unix
https://www.garlic.com/~lynn/99.html#16 Old Computers
https://www.garlic.com/~lynn/99.html#158 Uptime (was Re: Q: S/390 on PowerPC?)
https://www.garlic.com/~lynn/99.html#159 Uptime (was Re: Q: S/390 on PowerPC?)
https://www.garlic.com/~lynn/2002.html#23 Buffer overflow
https://www.garlic.com/~lynn/2002.html#32 Buffer overflow
https://www.garlic.com/~lynn/2002.html#34 Buffer overflow
https://www.garlic.com/~lynn/2003.html#30 Round robin IS NOT load balancing (?)
https://www.garlic.com/~lynn/2003.html#33 Round robin IS NOT load balancing (?)
https://www.garlic.com/~lynn/2003c.html#8 Network separation using host w/multiple network interfaces
https://www.garlic.com/~lynn/2003c.html#12 Network separation using host w/multiple network interfaces
https://www.garlic.com/~lynn/2003c.html#24 Network separation using host w/multiple network interfaces
https://www.garlic.com/~lynn/2003c.html#25 Network separation using host w/multiple network interfaces
https://www.garlic.com/~lynn/2003c.html#57 Easiest possible PASV experiment
https://www.garlic.com/~lynn/2004k.html#32 Frontiernet insists on being my firewall
https://www.garlic.com/~lynn/2004o.html#53 360 longevity, was RISCs too close to hardware?
https://www.garlic.com/~lynn/2005f.html#55 What is the "name" of a system?
https://www.garlic.com/~lynn/2005g.html#21 Protocol stack - disadvantages (revision)
https://www.garlic.com/~lynn/2005n.html#5 Wildcard SSL Certificates
https://www.garlic.com/~lynn/2005n.html#34 Data communications over telegraph circuits
https://www.garlic.com/~lynn/2005o.html#24 is a computer like an airport?
https://www.garlic.com/~lynn/2005r.html#32 How does the internet really look like ?
https://www.garlic.com/~lynn/2005r.html#39 What ever happened to Tandem and NonStop OS ?
https://www.garlic.com/~lynn/2006j.html#15 30 hop limit
https://www.garlic.com/~lynn/2007d.html#36 MAC and SSL
https://www.garlic.com/~lynn/2007h.html#67 SSL vs. SSL over tcp/ip
https://www.garlic.com/~lynn/2007i.html#44 latest Principles of Operation
https://www.garlic.com/~lynn/2007p.html#34 what does xp do when system is copying
https://www.garlic.com/~lynn/2007p.html#67 what does xp do when system is copying
https://www.garlic.com/~lynn/2007r.html#13 What do ATMS and card readers use?
https://www.garlic.com/~lynn/2008n.html#34 Builders v. Breakers
https://www.garlic.com/~lynn/2009d.html#60 Lack of bit field instructions in x86 instruction set because of patents ?
https://www.garlic.com/~lynn/2009m.html#32 comp.arch has made itself a sitting duck for spam
https://www.garlic.com/~lynn/2009n.html#41 Follow up
https://www.garlic.com/~lynn/2009n.html#43 Status of Arpanet/Internet in 1976?
https://www.garlic.com/~lynn/2009o.html#40 The Web browser turns 15: A look back;
https://www.garlic.com/~lynn/2009q.html#29 Check out Computer glitch to cause flight delays across U.S. - MarketWatch
https://www.garlic.com/~lynn/2009s.html#8 Union Pacific Railroad ditches its mainframe for SOA
https://www.garlic.com/~lynn/2010g.html#66 What is the protocal for GMT offset in SMTP (e-mail) header
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Sun, 13 Jun 2010 09:47:54 -0400jmfbahciv <See.above@aol.com> writes:
system meter was built into hardware ... not accessible by any software. was used by IBM for billing usage for machines on lease ... early 70s, company converted most customer machines from lease to sale. The issue was that the customer didn't own the machine ... and the customer was being charge based on (hardware) usage. there was all sorts of software, operating systems, and user code that ran on 360s ... some of it built from scratch by customers (MTS operating system done at Michigan univ. for example) and company couldn't necessary rely on any software correctly reflecting hardware use for basing machine charges.
for little topic drift
https://www.garlic.com/~lynn/2010j.html#35 IBM Rational Developer for System z
cp67 did cpu usage accounting using the location 80 (x'50') timer
... that was high-speed interval feature on 360/67 ... resolution
approx. 13microseconds. when switching tasks/charging, it would place
the new tasks value in x'54' and do an overlap move:
MVC X'4C'(8),X'50'
eight bytes starting at x'50' was (overlapping) moved to x'4C' ...
basically old (4byte timer) value in x'50' was x'4C' and the new value
at x'54' was moved to x'50' in one instruction.
Mechanism was used for accounting for individual virtual machine processor time, supervisor/kernel time spent on behalf of each virtual machine, time spent in wait state, and (originally) "overhead" time.
when initial cp67 was delivered to univ. in jan68, "overhead" time was
time that the dispatcher spent "running" master task list looking for
work to do ... and periodic scheduling functions adjusting dispatching
priorities. "overhead" increased (non-linear) proportional to the number
of logged on users. One of the first things I did when rewriting
high-use critical pathlengths ... was add ordered list of dispatchable
tasks ... effectively then the dispatcher only had to pop the first item
off the dispatch list to select next runnable ... rather than doing scan
of all possible elements on master list. "overhead" had been pushing 10%
of processor with more than 30-35 users. the change nearly eliminated
all "overhead". part of old presentation that I made at Fall68 (ibm user
group) Atlantic City SHARE meeting
https://www.garlic.com/~lynn/94.html#18 CP/67 & OS MFT14
above references both a lot of pathlength rewrite I had done for cp67 kernel ... measuring OS/360 MFT14 running in virtual machine ... as well as a lot of performance work I had done on OS/360 (applicable whether running native or in a virtual machine).
I was also starting to do other cp67 rewrite for the scheduling algorithms, page replacement algorithms, I/O subsystem on how I/O queueing was done for disk and paging drum ... bunch of other stuff.
misc. past posts related to dynamic adaptive resource management
algorithms (& fair share scheduling)
https://www.garlic.com/~lynn/subtopic.html#fairshare
misc. past posts related to page replacement algorithms
https://www.garlic.com/~lynn/subtopic.html#wsclock
Summer of '69, I added code to cp67 to make parts of the kernel code pageable. As part of doing that, I created stub, dummy virtual address space tables for the kernel image that identified which parts were pageable and which parts weren't pageable. Then to make the stub, dummy virtual address space tables work ... I created stub virtual machine control block for "SYSTEM" virtual machine. Then some of the residual stuff that was still related to "OVERHEAD" ... was switched to being accounted for against the "SYSTEM" virtual machine.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Information on obscure text editors wanted Newsgroups: alt.folklore.computers Date: Sun, 13 Jun 2010 10:23:48 -0400Peter Flass <Peter_Flass@Yahoo.com> writes:
the CMS file system kept index of physical block records for a file. If a file was "fixed" ... if application specified reading a specific record ... it could calculate the displacement into the file (using the fixed record length) and then select the physical block and read it directly from the physical block index.
variable length records were implemented with filesystem half-byte length field in front of each record. to read/write specific record, required running thru the file to find the corresponding record.
later os/dos simulation "variable block" support was layered on top of underlying CMS filesystem ... increasing efficiency of any "random" access for "variable block"
this references original cp67/cms editor (somewhat over from ctss)
which did stream bits between two disk files:
https://www.garlic.com/~lynn/2010j.html#11 Information on obscure text editors wanted
fairly soon a new "memory" cms editor became the standard ... i.e. (if it fit) load the whole file into (virtual) memory and operate on it there. the original editor then became "cedit" ... and were used for files that were too large to load into memory.
for other topic drift ... fairly early, I provided a cp67 layer that
overlayed cms filesystem on top of page mapped api (using 4k page size
block) ... which provided all sorts of pathlength, thruput and
functional improvments (compared to the channel program disk i/o
paradigm) ... misc past posts
https://www.garlic.com/~lynn/submain.html#mmap
old email mentioning moving lots of changes from cp67 base to
vm370 base ... including page mapped filesystem support:
https://www.garlic.com/~lynn/2006v.html#email731212
https://www.garlic.com/~lynn/2006w.html#email750102
https://www.garlic.com/~lynn/2006w.html#email750430
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Friden Flexowriter equipment series Newsgroups: alt.folklore.computers Date: Sun, 13 Jun 2010 10:52:31 -0400"Dave Wade" <g8mqw@yahoo.com> writes:
the displaywriter follow-on was being done in cp.r implemented in pl.8 ... however, retargeting for unix workstation market required unix and c language. they got the company that had done the at&t unix port to the pc for pc/ix ... to do "aix" for reworked displaywriter (pc/rt).
i have some vague memory of presentation on displaywriter follow-on about 200 pl.8 programmers working on the effort ... so when the displaywriter got canceled ... they needed something for all those pl.8 programmers to do. they came up with the unix implementation was going to be ontop of an abstract virtual machine implementation called "VRM" that would be done by the pl.8 programmers. there was some claim that would be faster and more efficient than if the outside company implemented directly to the bare ROMP processor.
the later counter-example was when the ACIS group was redirected from doing BSD port to 370 to doing the BSD port to the bare ROMP hardware (for what ACIS sold as "AOS").
misc. past posts mentioning 801, romp, rios, power, fort knox, somerset,
power/pc, etc
https://www.garlic.com/~lynn/subtopic.html#801
during HSDT effort
https://www.garlic.com/~lynn/subnetwork.html#hsdt
I tried to order a FE toolkit ... it was packaged to look like a large (suitcase-like) briefcase. For some reason, it got rejected when it became apparent that I wasn't in a service division ... I had to escalate to a executive level to get the order to go thru. It came with lots of tools that appear to be used for springs and other things that might be found in the innards of typewriters and misc. other things. They let me keep it when I "retired".
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: A "portable" hard disk Newsgroups: alt.folklore.computers Date: Mon, 14 Jun 2010 11:23:45 -0400"Charlie Gibbs" <cgibbs@kltpzyxm.invalid> writes:
370/155 had 2mic core memory ... with processor cache.
370/145 had 200-300ns electronic memory and no cache (electronic memory was much less expensive to manufacture).
http://www-03.ibm.com/ibm/history/exhibits/mainframe/mainframe_PP3145.html
from above:
In a major departure from conventional computer technology, International
Business Machines Corporation today introduced its first computer using
a main memory made entirely of monolithic circuits.
To store its data and instructions, the new IBM System/370 Model 145
uses silicon memory chips, rather than the magnetic core technology that
has been the mainstay of computer memories for the past 15 years. More
than 1,400 microscopic circuit elements are etched onto each
one-eighth-inch-square chip.
... snip ...
one of the changes from 155 to 158 was memory subsystem change that was similar to 145 (mitigating cache miss penalty) and less expensive.
360->370 transition period also started to see more switch-over from leased to purchase ... so lots of the 360 prices had been given in terms of monthly lease (not purchase).
from IBM 145 archive article:
Monthly rental for typical configurations of IBM System/370 Model 145
will range from about $14,950 (with 112,000 characters of main memory)
to $37,330 (512,000 characters), with purchase prices ranging from about
$705,775 to $1,783,000. Initial customer shipments will be scheduled for
late next summer.
... snip ...
assuming price range is mainly amount of memory ...
lease (37330-14950)/(512000-112000) ... is about $56k/mbyte
purchase (1783000-705775)/(512000-11200) ... is about $2.7m/mbyte
some product announcements:
http://ed-thelen.org/comp-hist/IBM-ProdAnn/index.html
picture of core memory:
http://www.thegalleryofoldiron.com/MISC.HTM
http://www.pbase.com/jimhwy/core_memory
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Monet, was Re: A "portable" hard disk Newsgroups: alt.folklore.computers Date: Mon, 14 Jun 2010 11:49:25 -0400Huge <Huge@nowhere.much.invalid> writes:
at the same time, the commercial/cash-management dial-up online banking operations were making presentations that they would never moved to the internet ... because of a large variety of security issues.
there have been a number of recent news items about exploits involving commercial internet banking ... including the fact that company/commercial banking doesn't have the same kind of "consumer" protections that apply to individual banking. most of the kinds of current exploits were described in some detail by the mid-90 presentations.
one of the current suggestions for online (internet) commercial/cash-management operations is companies have a separate PC that is *ONLY* used for online banking and is never used for any other purpose.
a few recent posts on the subject:
https://www.garlic.com/~lynn/2010b.html#21 security and online banking
https://www.garlic.com/~lynn/2010c.html#72 Users still make hacking easy with weak passwords
https://www.garlic.com/~lynn/2010c.html#79 Customers risk online banking fraud by reusing bank credentials
https://www.garlic.com/~lynn/2010c.html#86 NY Town's Bank Account Hacked; Poughkeepsie Loses $378K in Fraudulent Transfers
https://www.garlic.com/~lynn/2010f.html#6 Online Banking & Password Theft
One of the related scenarios is that in the very early part of this decade, there was fairly wide program to deploy smartcards for internet financial transactions and (internet) online banking ... that included serial-port smartcard readers that were either free or at a very nominal cost. The effort fairly quickly floundered due to enormous consumer support issues related to support serial-port devices. It somewhat can be characterised that in a period of something like give years ... the ephemeral institutional knowledge regarding difficulty with support serial-port (especially after-market) devices (attempting to work out all the serial-port issues) evaporated. Customers were calling in with BSOD and having to re-install their operating system from scratch in order to recover from the serial-port drivers.
The aftermath of the support difficulties resulted in the effort being abandoned and widely spreading opinion in the industry that smartcards weren't practical in the customer marketplace ... even tho the issue wasn't directly with the smartcards but with the serial-port smartcard readers. It should also be noted that all the serial-port difficulties was one of the major motivations for USB development. Another apparent side-effect was it was in the aftermath that microsoft canceled various smartcard efforts (including work on smartcard operating system).
a few past posts mentioning aborted internet smartcard deployment:
https://www.garlic.com/~lynn/2010d.html#25 Cambridge researchers show Chip and PIN system vulnerable to fraud
https://www.garlic.com/~lynn/2010f.html#26 Should the USA Implement EMV?
https://www.garlic.com/~lynn/2010f.html#27 Should the USA Implement EMV?
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Intelligence and foreign policy Newsgroups: alt.folklore.computers Date: Mon, 14 Jun 2010 13:07:58 -0400Patrick Scheible <kkt@zipcon.net> writes:
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Mon, 14 Jun 2010 15:17:57 -0400hancock4 writes:
part of it is "specialty" engines (basically mainframe processors licensed only for new kinds of workloads/applications) ... at drastically reduced/competitve prices.
somewhat related recent mainframe competitive posts:
https://www.garlic.com/~lynn/2010i.html#0 Processors stall on OLTP workloads about half the time--almost no matter what you do
https://www.garlic.com/~lynn/2010i.html#1 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010i.html#5 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010i.html#81 BMC reveals 'free money' mainframe and DB2 tools
https://www.garlic.com/~lynn/2010j.html#9 Windows, Linux propel Q1 server sales, Unix boxes, mainframes stalled
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Knuth Got It Wrong Newsgroups: alt.folklore.computers Date: Tue, 15 Jun 2010 08:18:22 -0400for the fun of it ... recent item ...
Developers Story Knuth Got It Wrong
http://developers.slashdot.org/story/10/06/14/2225251/Knuth-Got-It-Wrong
You're Doing It Wrong - ACM Queue
http://queue.acm.org/detail.cfm?id=1814327
from above:
Would you believe me if I claimed that an algorithm that has been on the
books as "optimal" for 46 years, which has been analyzed in excruciating
detail by geniuses like Knuth and taught in all computer science courses
in the world, can be optimized to run 10 times faster?
... snip ...
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Tue, 15 Jun 2010 09:00:14 -0400Walter Bushell <proto@panix.com> writes:
there was presentation that google does a lot of selection of commodity parts ... looking at variety of costs issues ... and builds servers for 1/3rd (or less) the cost of what one would pay for brand name.
while my previous post mentions that "specialty engines" have (comparitive) significant price reductions ... this article mentions that the standard processors (used for running traditional/legacy workloads) still carry quite a premium (compared to mainframe prices from earlier eras).
Financial Matters: Mainframe Processor Pricing History
http://www.zjournal.com/index.cfm?section=article&aid=346
from above (2006) article:
is that the price per MIPS today is approximately six times higher than
the $165 per MIPS that the traditional technology/price decline link
would have produced
... snip ...
recent posts mentioning above article:
https://www.garlic.com/~lynn/2010h.html#81 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010h.html#83 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010i.html#0 Processors stall on OLTP workloads about half the time--almost no matter what you do
https://www.garlic.com/~lynn/2010i.html#2 Processors stall on OLTP workloads about half the time--almost no matter what you do
https://www.garlic.com/~lynn/2010i.html#5 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010i.html#81 BMC reveals 'free money' mainframe and DB2 tools
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Tue, 15 Jun 2010 13:42:58 -0400re:
somewhat related
IBM goes elephant with Nehalem-EX iron; Massive memory for racks and
blades
http://www.theregister.co.uk/2010/04/01/ibm_xeon_7500_servers/
from above:
With so much of its money and profits coming from big Power and
mainframe servers, you can bet that IBM is not exactly enthusiastic
about the advent of the eight-core "Nehalem-EX" Xeon 7500 processors
from Intel and their ability to link up to eight sockets together in a
single system image. But IBM can't let other server makers own this
space either, so it had to make some tough choices.
... snip ...
for other topic drift, recent ibm-main posts about needing lots of
independent channels for mainframe half-duplex synchronous I/O paradigm
... but moving to full-duplex, asynchronous I/O operation ... becomes
primarily an issue of aggregate data transfer:
https://www.garlic.com/~lynn/2010e.html#36 What was old is new again (water chilled)
https://www.garlic.com/~lynn/2010f.html#13 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010f.html#16 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010f.html#18 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010h.html#62 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010h.html#63 25 reasons why hardware is still hot at IBM
one of the differences is that the servers are profit for the vendors ... but cost for customers. google has incentive to minimize its server costs (as opposed to vendors looking maximize server profit, which sometimes indirectly also means maximizing market share ... in competitive environment).
in the past, users didn't have scale to justify doing their own servers ... especially when it involved designing and manufacturing processors from scratch. with the microprocessor revolution ... more and more server vendors have reduced their costs by moving to industry commodity microprocessor chips. the issue here is that such a server paradigm change reduces the break-even level for customers to (also) use the same commodity sources and build their own servers.
operations like google somewhat led the way. however, other large businesses that are heavy data processing users have been looking at it for internal "cloud" operations ... recent item:
Global Banks Form Consortium To Counter HP, IBM, & Oracle
http://www.informationweek.com/news/global-cio/interviews/showArticle.jhtml?articleID=224700158
from above:
Move by three global banks to reduce their dependence on HP, IBM, and
Oracle by working together to build their own highly secure
cloud-based global infrastructure and network.
... snip ...
followup article:
Can HP, IBM, & Oracle Make Peace With Breakaway Banks? Three huge
global banks plan to revolutionize the CIO-vendor relationship -- will
their audacious plan trigger major changes from major IT vendors?
http://www.informationweek.com/news/global-cio/interviews/showArticle.jhtml?articleID=224700158
misc. recent posts referencing the cosortium reference
https://www.garlic.com/~lynn/2010h.html#71 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010h.html#83 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010i.html#12 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010i.html#22 Idiotic programming style edicts
https://www.garlic.com/~lynn/2010i.html#35 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010i.html#40 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010i.html#66 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010j.html#17 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
reference to Nehalem-EX targeted for mainframe server market:
https://www.garlic.com/~lynn/2010g.html#25 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#27 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#28 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#32 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#35 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010g.html#37 16:32 far pointers in OpenWatcom C/C++
https://www.garlic.com/~lynn/2010g.html#77 IBM responds to Oracle's Exadata with new systems
https://www.garlic.com/~lynn/2010h.html#10 Far and near pointers on the 80286 and later
https://www.garlic.com/~lynn/2010h.html#12 OS/400 and z/OS
https://www.garlic.com/~lynn/2010h.html#19 How many mainframes are there?
https://www.garlic.com/~lynn/2010h.html#24 How many mainframes are there?
https://www.garlic.com/~lynn/2010h.html#63 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010i.html#0 Processors stall on OLTP workloads about half the time--almost no matter what you do
https://www.garlic.com/~lynn/2010i.html#1 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010j.html#9 Windows, Linux propel Q1 server sales, Unix boxes, mainframes stalled
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Knuth Got It Wrong Newsgroups: alt.folklore.computers Date: Tue, 15 Jun 2010 20:39:43 -0400Patrick Scheible <kkt@zipcon.net> writes:
... it would have detracted from the "for the fun of it" phrase ... lots of people can complain about the slashdot characterization
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Knuth Got It Wrong Newsgroups: alt.folklore.computers Date: Wed, 16 Jun 2010 08:46:10 -0400Quadibloc <jsavard@ecn.ab.ca> writes:
one of the early experiences of this was on cp67/cms porting apl\360 to cms (with workstations the size of virtual memory) ... for cms\apl and having to redo apl's memory allocation & garbage collection.
standard apl\360 environment had been real storage with workspaces that were typically 16k-32k bytes ... where the complete workspace was always swapped as single unit. apl would always allocate a new storage location on every assignment (even if the variable had previously had allocation). this would continue until all memory had been exhausted and then it would perform garbage collection ... compacted allocations to contiguous area and then start all over.
in large virtual memory environment, this pretty much guaranteed that any apl application would quickly touch every virtual memory page before compacting back to contiguous area and repeating. an early form of vs/repack was used to trace typical apl instruction use and storage references. some number of this were printed/plotted on 1403 ... white backside of greenbar paper ... with x-axis memory storage address scalled to approx. 6ft vertical ... and time/instruction running for couple dozen feet vertical. these were then taped on the walls of science center corridors with storage references showing a very distinctive sawtooth pattern ... a slanted line of storage references as allocation quickly rose to top of virtual memory and then a solid line as garbage collection was performed and allocation compacted to bottom of the workspace.
in mid-70s, the vs/repack code was packaged, and shipped as product from the science center ... including doing semi-automated program reorganization for virtual memory environment. it was also used internally for analysing various large software application products ... not only for characterizing virtual memory operation but also "hot-spot" identification ... things like IMS database making transition from real-storage environment to virtual memory environment. One of the characteristics of large DBMS and evolving stuff like B-Tree ... was that their DBMS cache management tended to be tightly integrated with other parts of the application (where DBMS cache management has lots of similarities with virtual memory management) .... with the tight integration within the application of both cache management and the code that used the cache ... it was more natural that cache friendly implementations would evolve.
misc past psots mentioning vs/repack
https://www.garlic.com/~lynn/94.html#7 IBM 7090 (360s, 370s, apl, etc)
https://www.garlic.com/~lynn/99.html#68 The Melissa Virus or War on Microsoft?
https://www.garlic.com/~lynn/2000g.html#30 Could CDR-coding be on the way back?
https://www.garlic.com/~lynn/2001b.html#83 Z/90, S/390, 370/ESA (slightly off topic)
https://www.garlic.com/~lynn/2001c.html#31 database (or b-tree) page sizes
https://www.garlic.com/~lynn/2001c.html#33 database (or b-tree) page sizes
https://www.garlic.com/~lynn/2001i.html#20 Very CISC Instuctions (Was: why the machine word size ...)
https://www.garlic.com/~lynn/2002c.html#28 OS Workloads : Interactive etc
https://www.garlic.com/~lynn/2002c.html#45 cp/67 addenda (cross-post warning)
https://www.garlic.com/~lynn/2002c.html#46 cp/67 addenda (cross-post warning)
https://www.garlic.com/~lynn/2002c.html#49 Swapper was Re: History of Login Names
https://www.garlic.com/~lynn/2002e.html#50 IBM going after Strobe?
https://www.garlic.com/~lynn/2002f.html#50 Blade architectures
https://www.garlic.com/~lynn/2003f.html#15 Alpha performance, why?
https://www.garlic.com/~lynn/2003f.html#21 "Super-Cheap" Supercomputing
https://www.garlic.com/~lynn/2003f.html#53 Alpha performance, why?
https://www.garlic.com/~lynn/2003g.html#15 Disk capacity and backup solutions
https://www.garlic.com/~lynn/2003h.html#8 IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM
https://www.garlic.com/~lynn/2003j.html#32 Language semantics wrt exploits
https://www.garlic.com/~lynn/2004.html#14 Holee shit! 30 years ago!
https://www.garlic.com/~lynn/2004c.html#21 PSW Sampling
https://www.garlic.com/~lynn/2004m.html#22 Lock-free algorithms
https://www.garlic.com/~lynn/2004n.html#55 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2004o.html#7 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2004q.html#73 Athlon cache question
https://www.garlic.com/~lynn/2004q.html#76 Athlon cache question
https://www.garlic.com/~lynn/2005.html#4 Athlon cache question
https://www.garlic.com/~lynn/2005d.html#41 Thou shalt have no other gods before the ANSI C standard
https://www.garlic.com/~lynn/2005d.html#48 Secure design
https://www.garlic.com/~lynn/2005h.html#15 Exceptions at basic block boundaries
https://www.garlic.com/~lynn/2005j.html#62 More on garbage collection
https://www.garlic.com/~lynn/2005k.html#17 More on garbage collection
https://www.garlic.com/~lynn/2005m.html#28 IBM's mini computers--lack thereof
https://www.garlic.com/~lynn/2005n.html#18 Code density and performance?
https://www.garlic.com/~lynn/2005o.html#5 Code density and performance?
https://www.garlic.com/~lynn/2006b.html#15 {SPAM?} Re: Expanded Storage
https://www.garlic.com/~lynn/2006b.html#23 Seeking Info on XDS Sigma 7 APL
https://www.garlic.com/~lynn/2006e.html#20 About TLB in lower-level caches
https://www.garlic.com/~lynn/2006e.html#46 using 3390 mod-9s
https://www.garlic.com/~lynn/2006i.html#37 virtual memory
https://www.garlic.com/~lynn/2006j.html#18 virtual memory
https://www.garlic.com/~lynn/2006j.html#22 virtual memory
https://www.garlic.com/~lynn/2006j.html#24 virtual memory
https://www.garlic.com/~lynn/2006l.html#11 virtual memory
https://www.garlic.com/~lynn/2006o.html#23 Strobe equivalents
https://www.garlic.com/~lynn/2006o.html#26 Cache-Size vs Performance
https://www.garlic.com/~lynn/2006r.html#12 Trying to design low level hard disk manipulation program
https://www.garlic.com/~lynn/2006x.html#1 IBM sues maker of Intel-based Mainframe clones
https://www.garlic.com/~lynn/2006x.html#16 The Future of CPUs: What's After Multi-Core?
https://www.garlic.com/~lynn/2007g.html#31 Wylbur and Paging
https://www.garlic.com/~lynn/2007m.html#55 Capacity and Relational Database
https://www.garlic.com/~lynn/2007o.html#53 Virtual Storage implementation
https://www.garlic.com/~lynn/2007o.html#57 ACP/TPF
https://www.garlic.com/~lynn/2007s.html#41 Age of IBM VM
https://www.garlic.com/~lynn/2008c.html#24 Job ad for z/OS systems programmer trainee
https://www.garlic.com/~lynn/2008c.html#78 CPU time differences for the same job
https://www.garlic.com/~lynn/2008d.html#35 Interesting Mainframe Article: 5 Myths Exposed
https://www.garlic.com/~lynn/2008e.html#16 Kernels
https://www.garlic.com/~lynn/2008f.html#36 Object-relational impedence
https://www.garlic.com/~lynn/2008l.html#81 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008m.html#69 Speculation ONLY
https://www.garlic.com/~lynn/2008q.html#65 APL
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers Date: Wed, 16 Jun 2010 09:02:28 -0400Quadibloc <jsavard@ecn.ab.ca> writes:
some of the discussions about Nehalem in other forums about moving Itanium RAS features to x86 ... were about needing also software RAS features ... and about whether the vendors have given up on Itanium being the processor of choice for high-end servers.
one of the big problems with all the C-language based implementations is
the vast numbers of buffer length related bugs associated with storage
allocation paradigm for things like strings. ... misc. past posts on
the subject
https://www.garlic.com/~lynn/subintegrity.html#overflow
in the 80s & 90s ... one of the primary justifications given for hosting mainframe unix ports under vm/370 ... was requirement for RAS (in some cases, field people wouldn't sign-off on servicing the machine w/o appropriate mainframe software RAS) ... and that adding mainframe software RAS to unix base was possibly order-of-magnitude larger effort than the straight-forward port of unix to 370 architecture. running unix under vm/370 would depend on vm/370 to provide appropriate level of error retry and well as EREP recording and reporting structure.
on the other hand we did start high-availability HA/CMP cluster product
in the late 80s ... misc. past posts
https://www.garlic.com/~lynn/subtopic.html#hacmp
it is probably better known these days for the cluster scale-up part of
the effort (than the straight-forward high availability part); some past
email
https://www.garlic.com/~lynn/lhwemail.html#medusa
old post with jan92 meeting reference:
https://www.garlic.com/~lynn/95.html#13
even tho there are starting to be signs that there is renewed focus on
re-integrating some cluster scale-up back into the high-availability part
... some recent references:
https://www.garlic.com/~lynn/2009p.html#43 From The Annals of Release No Software Before Its Time
https://www.garlic.com/~lynn/2009p.html#46 From The Annals of Release No Software Before Its Time
https://www.garlic.com/~lynn/2009p.html#49 big iron mainframe vs. x86 servers
https://www.garlic.com/~lynn/2009p.html#54 big iron mainframe vs. x86 servers
https://www.garlic.com/~lynn/2009p.html#55 MasPar compiler and simulator
https://www.garlic.com/~lynn/2009p.html#57 MasPar compiler and simulator
https://www.garlic.com/~lynn/2009p.html#83 What would be a truly relational operating system ?
https://www.garlic.com/~lynn/2009p.html#85 Anyone going to Supercomputers '09 in Portland?
https://www.garlic.com/~lynn/2009q.html#19 Mainframe running 1,500 Linux servers?
https://www.garlic.com/~lynn/2009q.html#21 Is Cloud Computing Old Hat?
https://www.garlic.com/~lynn/2009q.html#42 The 50th Anniversary of the Legendary IBM 1401
https://www.garlic.com/~lynn/2009q.html#68 Now is time for banks to replace core system according to Accenture
https://www.garlic.com/~lynn/2009r.html#4 70 Years of ATM Innovation
https://www.garlic.com/~lynn/2009r.html#9 The 50th Anniversary of the Legendary IBM 1401
https://www.garlic.com/~lynn/2009r.html#33 While watching Biography about Bill Gates on CNBC last Night
https://www.garlic.com/~lynn/2010g.html#0 16:32 far pointers in OpenWatcom C/C++
https://www.garlic.com/~lynn/2010g.html#4 Handling multicore CPUs; what the competition is thinking
https://www.garlic.com/~lynn/2010g.html#33 SQL Server replacement
https://www.garlic.com/~lynn/2010g.html#44 16:32 far pointers in OpenWatcom C/C++
https://www.garlic.com/~lynn/2010g.html#48 Handling multicore CPUs; what the competition is thinking
https://www.garlic.com/~lynn/2010g.html#77 IBM responds to Oracle's Exadata with new systems
https://www.garlic.com/~lynn/2010g.html#81 What is the protocal for GMT offset in SMTP (e-mail) header time-stamp?
https://www.garlic.com/~lynn/2010h.html#63 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010i.html#29 someone smarter than Dave Cutler
https://www.garlic.com/~lynn/2010i.html#61 IBM to announce new MF's this year
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Knuth Got It Wrong Newsgroups: alt.folklore.computers Date: Wed, 16 Jun 2010 09:20:54 -0400re:
slightly more drift ... most of original relational/sql dbms
implementation was done in on 370/145 under vm370 (in bldg. 28) ... so
started out from a virtual memory prespective. misc. past posts
mentioning system/r
https://www.garlic.com/~lynn/submain.html#systemr
there are also some interesting intersections involved in operating an LRU managed cache paradigm in a (LRU managed) paged, virtual memory environment (unless you resort to just pinning the dbms cache pages in real memory).
now the first RDBMS product to ship was on multics (virtual memory base) ... beating sql/ds (productized system/r ... aka tech transfer from bldg. 28 in san jose to endicott for release as product).
for historical drift, some of the ctss people went to 5th flr multics
and some went to the science center on the 4th flr ... where (virtual
memory & virtual machine) cp67 was done ... misc. past posts mentioning
545 tech sq (i was there for most of the 70s before transferring to
bldg. 28 in san jose)
https://www.garlic.com/~lynn/subtopic.html#545tech
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Information on obscure text editors wanted Newsgroups: alt.folklore.computers Date: Wed, 16 Jun 2010 09:48:21 -0400jmfbahciv <See.above@aol.com> writes:
this gives site (and other multics) timelines:
https://www.multicians.org/site-timeline.html
most site showup late 70s (or early 80s).
being just one flr separated in 545 tech sq ... there was a little
competitive flavor ... as well as science center had originally
started out wanting the multics effort ... some amount of this
covered in melinda's history
https://www.leeandmelindavarian.com/Melinda#VMHist
it wouldn't be considered fair to base comparison on number of
customers. i've commented in the past, it wasn't even fair to compare
total Multics installations that ever existed against just the number
of internal corporate vm370 installations. I have noted that at one
period in the mid-70s, I was supporting approx. the same number of
internal csc/vm customers ... as total number of multics installations
that ever existed. ... some old email mentioning csc/vm:
https://www.garlic.com/~lynn/2006v.html#email731212
https://www.garlic.com/~lynn/2006w.html#email750102
https://www.garlic.com/~lynn/2006w.html#email750430
for other ... this mentions multics "new storage system"
https://www.multicians.org/nss.html
that some of the motivation for the new storage system was that MIT had a multics system that could take an hour to recover from a crash ... and in the same room, a cp67 system that could recover in a couple minutes.
this has a reference to modification to the TTY device support resulted
in CP67 crashing and restarting 27 times in the same day (multics taking
and hr to recover could hardly do that 8times/day)
https://www.multicians.org/thvv/360-67.html
i would be partially responsible for the crashes ... I had done the CP67 TTY support as undergraduate in the 60s ... and at one point fiddled something with using one byte to calculate line-length. The MIT modification was to increase the maximum TTY line-length to something like 1200 bytes (I think to support a dial-up tty plotter device over at harvard) ... and the one byte hack resulted in incorrect length calculation and clobbering storage. In any case, supposedly, the side-by-side crash recovery comparison was motivation for multics new storage system
other posts mentioning 545 tech sq
https://www.garlic.com/~lynn/subtopic.html#545tech
csc/vm (&/or sjr/vm) posts
https://www.garlic.com/~lynn/submisc.html#cscvm
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Wed, 16 Jun 2010 10:10:19 -0400art clemons <artclemons@aolSPAM.com> writes:
recent post mentioning doing some work on airline res
system in the mid-90s
https://www.garlic.com/~lynn/2010b.html#73 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#74 Happy DEC-10 Day
and part of thread that xscale processor used in treo
https://www.garlic.com/~lynn/2010b.html#78 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#79 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#80 Happy DEC-10 Day
has more raw computing power than what we had sized as needed to handle every airline passenger reservation for every passenger in the world for every airline in the world (in the mid-90s).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Wed, 16 Jun 2010 12:21:18 -0400re:
airline passenger res. system design was pretty much left over from the 60s ... done on 7090 & 360. started with "routes" which represented about 25% of the load ... aka lists direct & connecting flts to get from origin city/airport to destination city/airport.
res. system operation would take oag flat file, one entry for every flt segment, take-off/landing ... in the mid-90s nearly 500k and (periodically) rebuild database entries for the possibly origin/destination pairs with direct and connecting flts that would get between the two points (typically two connects at the most, origin/destinations that required more than two connects would have to be stitched together by agent/manually). database lookup/key was the combination of the origin & destination pair.
i redid it by building a compressed memory resident image of the original oag flat file ... and then dynamically doing search of that data for each query ... which turned out to be significantly faster than doing dbms query. one of the influences was that between the 60s & the 90s, real storage configurations got larger than (compressed version of) oag flat file. another possibly influence is that I had just recently done a contract that involved chip/board physical layout.
the search results then required some rule post-processing. as part of standard approach building dbms ... various airports have all sorts of special rules about getting minimum (connecting) time to get from gate on one concourse to gate on another concourse. then there are special airport codes for cities with multiple airports and special connecting flt rules involving gate/concourse at one airport and gate/councourse at different airport (instead of the rules being used as part of the dbms build ... the rules had to be applied during the real-time search).
one of the side-effects ... besides core lookup being nearly two-orders of magnitude faster ... was that it could handle any number of connections between two origin/destinations anywhere in the world.
the paradigm switch is analogous to various recent news items about speedups from designing dbms from scratch to be "in-memory" (claims of ten times performance improvement compared to traditional dbms design ... even when comparison involves traditional dbms with all disk data already loaded in cache).
a little x-over thread about selecting row/column operation
https://www.garlic.com/~lynn/2010j.html#44 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#47 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#48 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#50 Knuth Got It Wrong
it turned out that there was a factor of five difference in the implementation thruput depending on how the memory image of the OAG data was traversed (basically difference in processor hardware cache efficiency).
another side-effect was that it helped eliminate the traditional ACP/TPF dbms rebuild outage that was typically part of 3rd shift sunday outage that happened a couple times per month. the route process could be offloaded to high-availability cluster ... the OAG memory rebuild could be done asynchronously and then take a few seconds to switch one-at-a-time, each application from the previous OAG memory image to the new OAG memory image ... w/o requiring any downtime/outage (I remember being in the fareast on monday morning trying to change airline resverations and the system being down for its regular scheduled maint).
It wasn't long after doing the implementation that you could get PCs large enough to run the full route finding application ... and then as somewhat referrred to ... in theory current treos are capabile of running the whole thing, handling the load for the whole world.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers Date: Wed, 16 Jun 2010 13:05:17 -0400some RAS cross-over
Anne & Lynn Wheeler <lynn@garlic.com> writes:
it wouldn't be considered fair to base comparison on number of
customers. i've commented in the past, it wasn't even fair to compare
total Multics installations that ever existed against just the number of
internal corporate vm370 installations. I have noted that at one period
in the mid-70s, I was supporting approx. the same number of internal csc/vm
customers ... as total number of multics installations that ever
existed. ... some old email mentioning csc/vm:
https://www.garlic.com/~lynn/2006v.html#email731212
https://www.garlic.com/~lynn/2006w.html#email750102
https://www.garlic.com/~lynn/2006w.html#email750430
from this recent post:
https://www.garlic.com/~lynn/2010j.html#51 Information on obscure text editors wanted
charlie, in addition to working on fine-grain smp locking for the cp/67
kernel (at the science center) ... and inventing the compare&swap
instruction (CAS was chosen because they are charlie's initials)
... misc. past smp/cas posts
https://www.garlic.com/~lynn/subtopic.html#smp
was having some cp67 kernel integrity problems involving general area of
serialization ... so he undertook to rewrite the virtual machine reset
(including serialization that was used for i/o reset, virtual machine
reset/re-ipl, and user logoff ... whether running single processor or
multiprocessor). part of this involved re-assigning all pending i/o
to the "system" psuedo virtual machine. recent post mentioning having
create "system" psuedo virtual machine for paging part of kernel:
https://www.garlic.com/~lynn/2010j.html#37 Idiotic programming style edicts
this had the side-effect of eliminating crashes ... dangling operations/pointers that completed after associated control blocks were dissolved ... and hung/zombie users.
now the official cp67 dev. group split off from the science center and during morph for vm370 product ... took-over/absorbed the boston programming center on 3rd flr. of 545 tech sq. then as it continued to grow ... it moved out to burlington mall ... taking over the old/vacant sbc bldg. there. as part of the morph into vm370 ... there was quite a bit of simplification and also dropped a lot of stuff I had done in cp67 (an exception was the pageable cp67 kernel stuff i had done summer of 1969 ... and hadn't shipped in cp67 product was part of of the original standard vm370 release).
in any case, as mentioned in the old email ... at some point the science
center start planing for replacing the 360/67 (cp67) with 370/155-II
(and moving to vm370) ... which required moving a lot of stuff that was
in cp67/cms to vm370/cms base. I've mentioned in the past that the "H"
updates for cp67 including being able to simulate/virtualize a 370
machine (and run software that was designed to run on real 370
... including vm370). so a lot of the early work ... i could test out on
the cambridge cp67 operation. However, at some point, I needed do some
real 370 hardware test ... and had to get some weekend time on one of
the vm370 devel group machines. a couple recent refs. to getting
weekend test time on one of their machines (which also involves
security/integrity issue):
https://www.garlic.com/~lynn/2010e.html#6 Need tool to zap core
https://www.garlic.com/~lynn/2010j.html#32 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
in any case, as part of the cp67 & then vm370 algorithm, performance, and thruput work ... a automated benchmark process had evolved. This included being able to specify synthentic workloads with numerous operation characteristcs, rebuild and change kernels, change configurations, change system parameters ... turning it loose and having it run for days at a time. one of the side-effects was specifying a workload that was 10-20 times ever actually seen in the real-world. one of those side-effects was that the artificial heavy load was guaranteed to crash a standard vm370 system. so in addition to everything else referrenced in the cp67->vm370 transition ... I also redid charlie's rewrite of the kernel serialization function for vm370 (eliminating all known cases of crashes from dangling operations as well as zombie users).
as been previously mentioned ... during the future system effort
(planning on completely replacing 370), the 370 hardware and software
product pipelines were allowed to go empty. then with the death of FS
there was mad-rush to get products back into the 370 product pipeline
misc. past posts mentioning future system
https://www.garlic.com/~lynn/submain.html#futuresys
part of that included selecting a bunch of stuff I had been doing all thru the future system period (some of it having been in cp67 product releases and then dropped in the morph to vm370 ... other stuff was new). A small subset of the shared segment stuff for cms was shipped in standard release 3. A lot of the other stuff was packaged as a separate kernel product called the "resource manager".
Because of gov. and other litigation, there was 23jun69 unbundling
announcement to start charging for application software ... but the
case was made that kernel software continued to be free. misc. past
references to unbundling
https://www.garlic.com/~lynn/submain.html#unbundle
the claim is then the distraction of FS ... allowed the clone 370
processors to gain market foothold. with the death of FS and the mad
rush to get out 370 products ... it was also decided to start charging
for kernel software ... initially done in stages. The first guinea pig
for kernel software charging was my "resource manager" ... which met
that I had to spend a lot of time with business planning, lawyers, and
pricing people. Also the final stage of release the "resource manager"
involved doing over 2000 (automated) benchmarks that took over three
months elapsed time to run (representing a wide variety of workloads and
configurations). misc. past posts mentioning automated benchmarking
https://www.garlic.com/~lynn/submain.html#bench
and some other recent posts mentioning csc/vm
https://www.garlic.com/~lynn/2010d.html#70 LPARs: More or Less?
https://www.garlic.com/~lynn/2010d.html#79 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#1 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#2 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#17 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
https://www.garlic.com/~lynn/2010e.html#25 HONE Compute Intensive
https://www.garlic.com/~lynn/2010f.html#24 Would you fight?
https://www.garlic.com/~lynn/2010g.html#72 Interesting presentation
https://www.garlic.com/~lynn/2010j.html#20 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#22 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Wed, 16 Jun 2010 17:30:53 -0400Ben Pfaff <blp@cs.stanford.edu> writes:
3330 disk drive was 200mbyte capacity and 800kbyte/sec transfer.
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3330.html
i think DIALOG datacenter in late 70s (not far from stanford) had 300 3330-clone disk drives connected to a pair of 370/158 processors (in loosely-coupled/cluster configuration).
3380 disk drives attempted to address a number of things in the early
80s. the per "disk" capacity was increased to 630mbyte and the data
transfer rate to 3mbyte/sec.
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3380.html
the channel protocol was tweaked to "data streaming" (change to channel protocol from doing hardware handshake on ever byte transferred to being able to transfer multiple bytes per handshake). this made it easier to achieve 3mbyte/sec transfer (mitigating number of end-to-end half-duplex handshakes). It also allowed increasing maximum channel distance from 200ft to 400ft. In the earlier regime, lots of installations were running into physical configuration problems with placing processor in the middle of cicle and then getting all the devices packed into a 200ft radius. Some customers were resorting to multiple floor configuration ... extending number of devices that could be connected within the 200ft limitation by going from two-dimensional radius to starting to approach 3dimensional sphere. doubling the maximum channel from 200ft to 400ft allowed for a datacenter with placing devices in a cicle with 400ft radius.
complicating the situation was datacenters with numerous processors in cluster configuration ... placing all the processors at sort of the center of sphere ... and then needing to have all devices within the 200ft radius of all processors.
this is recent posts/threads talks about doing special effort for santa
teresa lab (since renamed silicon valley lab) in 1980, using
HYPERchannel for channel extension and allowing placing local channel
devices 10-15 miles away. Part of the HYPERchannel support was going to
downloading asynchronous channel programs ... as mentioned,
significantly increased the I/O thruput. Standard channel protocol
included a lot of busy time doing control operations associated with
channel programs ... being able to download channel programs (with
HYPERchannel) to be run remotely asynchronously ... effectively freed up
the mainframe channel for just doing data transfer.
https://www.garlic.com/~lynn/2010f.html#16 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010f.html#18 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010i.html#2 Processors stall on OLTP workloads about half the time--almost no matter what you do
some misc. other posts mentioning doing things with HYPERchannel
https://www.garlic.com/~lynn/subnetwork.html#hsdt
earlier posts in this thread:
https://www.garlic.com/~lynn/2010j.html#43 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#45 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#46 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#49 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#52 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#53 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#54 Article says mainframe most cost-efficient platform
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers Date: Wed, 16 Jun 2010 18:58:59 -0400"Charlie Gibbs" <cgibbs@kltpzyxm.invalid> writes:
pl.8 (theoretically a pl/i subset) was quite efficient. some old
comparisons between pl.8 and vs/pascal. one part of pl.8 technology was
doing a pascal language frontend for pl.8. this is old email comparing
the pascal/pl.8 on a number of platforms against vs/pascal on 3033
(high-end 370 for the period):
https://www.garlic.com/~lynn/2006t.html#email810808
in this post
https://www.garlic.com/~lynn/2006t.html#9 32 or even 64 registers for x86-64?
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Thu, 17 Jun 2010 09:28:16 -0400DMcCunney <plugh@xyzzy.com> writes:
mainframe channels could consume a significant amount of resources and the control & handshaking overhead between channels and devices/control units could be significant.
the legacy design from the 60s was trade-offs regarding extremely expensive and relatively scarce electronic storage ... as a result mainframe memory was used for channel programs ... and the channel would execute the channel program out of mainframe memory. this consumed lots of memory bus capacity (independent of just memory bus for doing data transfer). it was less of an issue for "integrated channels" ... where the same engine that executed the processor microcode also was being shared with executing the channel micrcode. however, it represented more & more of bottleneck as things scaled up ... and deployed things like independent processors for executing channel (including latency issue between the channel processor and frequent accesses to mainframe memory for every channel command in channel program).
I've periodically mentioned that adding tty/ascii terminal support to
cp67 as undergraduate in the 60s ... I tried to make the terminal
controller do something that it couldn't quite do. this was motivation
for the univ. to start a clone controller effort ... using an interdata3
programmed to emulate the mainframe terminal controller.
https://www.garlic.com/~lynn/submain.html#360pcm
part of this required reverse engineering the channel interface and building a interface board for the interdata3. the 360/67 had high-speed timer option which "tic'ed" the location 80 timer approx. ever 13microseconds (and if memory bus was reserved by channel, locking out timer from updating location 80 ... and a 2nd tic come around while previous timer tic update was pending ... it would force machine fault). one of the first bugs attaching the interdata3 to the channel was locking up the memory bus for longer than timer update interval.
four of us then got written up for responsible for clone controller
business. interdata began selling the boxes as clone controllers ...
and this continued after P/E bought interdata under the perkin/elmer
label. the clone controller business has been cited as major motivation
for the future system effort ... some past posts
https://www.garlic.com/~lynn/submain.html#futuresys
and then the distraction of future system effort is attributed with allowing clone processors to get market foothold.
another such trade-off was in the design of disk (DASD) I/O commands.
standard disk channel program includes a search command ... comparing
some part of each record against specified critera. The disk device
search argument is refetched from mainframe memory for every record that
passes under the read/write head (checking if the record meets the
search criteria). This results in enormous overhead on every part of the
I/O infrastructure as well as introducing latency limits on distance
from device to mainframe memory. some past posts mentioning the
mainframe disk design & CKD search characteristics
https://www.garlic.com/~lynn/submain.html#dasd
the trade-offs in the designs was beneficial in the 60s because of scarce electronic storage ... by at least the mid-70s ... the designs were becoming a bottleneck with significant increase in availability of electronic storage.
other posts in this thread:
https://www.garlic.com/~lynn/2010j.html#43 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#45 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#49 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#52 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#53 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#54 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#56 Article says mainframe most cost-efficient platform
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Thu, 17 Jun 2010 09:38:09 -0400Ahem A Rivet's Shot <steveo@eircom.net> writes:
of course, we had started ha/cmp for rs/6000 in the same time-frame
https://www.garlic.com/~lynn/subtopic.html#hacmp
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Thu, 17 Jun 2010 10:20:29 -0400Anne & Lynn Wheeler <lynn@garlic.com> writes:
the trade-off design regarding amount of available electronic storage has repeatedly shown up. there is the whole channel and disk i/o design trade-offs from the 60s.
i've commented this also shows up in a little discord between the
60s (physical) databases and rdbms effort ... misc. past posts
mentioning original relational/sql implementation
https://www.garlic.com/~lynn/submain.html#systemr
end of the 70s, the IMS group in STL would somewhat bash relational effort pointing out that the implicit indexes typically doubled the physical disk space (compared to IMS) and significantly increased the disk accesses (reading various index levels) to get to actual record. the relational retort was that the direct record pointers (exposed in IMS data) significantly increased administrative effort (things like re-organization).
going into the 80s ... the amount of disk storage significantly increased at the same time the cost significantly decreased ... mitigated the disk issues related to implicit indexes in relational. at the same time, futher increases in availability of electronic storage allowed significant portion of indexes to be cached in memory (mitigating number of disk reads to transverse the index).
MULTICS shipped the first RDBMS product
https://en.wikipedia.org/wiki/Multics_Relational_Data_Store
https://www.multicians.org/mgm.html
and several other vendors were in the market before system/r
tech. transfer from bldg. 28 to Endicott for sql/ds. then one
of the people mentioned in this jan92 meeting
https://www.garlic.com/~lynn/95.html#13
claimed to have handled most of the tech. transfer from endicott back to STL for DB2. Possibly one of the delays for productizing system/r was the other DBMS product offerings the company had.
design of indexes ... requiring significant access overhead
... also shows up in this recent thread/article:
https://www.garlic.com/~lynn/2010j.html#44 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#47 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#48 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#50 Knuth Got It Wrong
IMS reference:
https://en.wikipedia.org/wiki/Information_Management_System
for other topic drift ... old email reference to Jim handing off a
number of things to me when he departed for tandem ... including
consulting with the IMS group
https://www.garlic.com/~lynn/2007.html#email801006
https://www.garlic.com/~lynn/2007.html#email801016
above references starting to get calls from places like Bank Of America on RDBMS stuff.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Information on obscure text editors wanted Newsgroups: alt.folklore.computers Date: Thu, 17 Jun 2010 17:35:23 -0400Rich Alderson <news@alderson.users.panix.com> writes:
recent reference to MIT running cp67 and MULTICS side-by-side in the same
machine room
https://www.garlic.com/~lynn/2010j.html#51 Information on obscure text editors wanted
comparison somewhat motivation for doing "new storage system" for multics
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Information on obscure text editors wanted Newsgroups: alt.folklore.computers Date: Thu, 17 Jun 2010 17:56:37 -0400Quadibloc <jsavard@ecn.ab.ca> writes:
somewhat related reference ... but with respect to 4th flr 545 tech. sq
... rather than 5th flr:
https://web.archive.org/web/20090117083033/http://www.nsa.gov/research/selinux/list-archive/0409/8362.shtml
for other MULTICS
https://www.garlic.com/~lynn/2002l.html#42 Thirty Years Later: Lessons from the Multics Security Evaluation
https://www.garlic.com/~lynn/2002l.html#43 another 30 year thing
https://www.garlic.com/~lynn/2002l.html#44 Thirty Years Later: Lessons from the Multics Security Evaluation
aka multics implementation in pli supposedly never had a buffer overflow problem.
references article
http://www.acsac.org/2002/papers/classic-multics.pdf
and the original
http://csrc.nist.gov/publications/history/karg74.pdf
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Thu, 17 Jun 2010 18:04:42 -0400reference to asynchronous, full-duplex i/o implying big part is for latency compensation (as datarates increase)
references using HYPERchannel as channel extender ... and being able
to download channel program to remote location (with it executing
remotely) ... significant mitigating latency
https://www.garlic.com/~lynn/2010j.html#55 Article says mainframe most cost-efficient platform
this was in 1980 as part of remoting 300 people from the IMS group to
offsite bldg ... and being able to preserve the human factors of local
3270 terminal operation (the alternative was remote 3270s which were
abysmal human factors in comparison). this is unrelated to item about
doing DBMS consulting to the IMS group ... mentioned here:
https://www.garlic.com/~lynn/2010j.html#59 Article says mainframe most cost-efficient platform
this reference to old jan92 meeting on scale-up
https://www.garlic.com/~lynn/95.html#13
it also references 9333/harrier and would have liked to see it it become interoperable with FCS ... instead of becoming SSA. harrier had taken SCSI devices and SCSI commands and packetized the commands and flowed them over serial copper links (dedicated copper for flow in each direction) ... running them asynchronously. this significantly reduced SCSI controller busy and overhead ... since the commands were packetized sent out asynchronously for execution at the devices. For even half-dozen disks ... it resulted in significantly higher aggregate thruput.
this ibm-main talks about various efforts in the late 80s and early
90s moving to asynchronous serial operation as approach to (at least)
being latency compensation.
https://www.garlic.com/~lynn/2010f.html#7 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010f.html#8 What was the historical price of a P/390?
this is part of thread mentioning mainframes highlighting the large
number of I/O channels (possibly as compensation for channel trade-off
design resulting in numerous bottlenecks & latencies as things moved
to higher datarates):
https://www.garlic.com/~lynn/2010f.html#13 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010f.html#18 What was the historical price of a P/390?
old email regarding doing cluster scale-up in conjunction with high
availability work
https://www.garlic.com/~lynn/lhwemail.html#medusa
with this email possibly just hrs being informed that the scale-up
work was transferred and we weren't supposed to work on anything
with more than four processors.
https://www.garlic.com/~lynn/2006x.html#email920129
and then 920217 press item with cluster scale-up announced as product:
https://www.garlic.com/~lynn/2001n.html#6000clusters1
and then rubbing salt into the wound ... press item about how
it caught them totally by surprise:
https://www.garlic.com/~lynn/2001n.html#6000clusters2
past post in comp.arch thread regarding events fall of '91 leading up
to transfer of cluster scale-up (and being told not to work on
configurations with more than four processors):
https://www.garlic.com/~lynn/2009p.html#57 MasPar compiler and simulator
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Information on obscure text editors wanted Newsgroups: alt.folklore.computers Date: Thu, 17 Jun 2010 18:29:51 -0400Anne & Lynn Wheeler <lynn@garlic.com> writes:
the original mainframe tcp/ip product implementation was done in vs/pascal ... there was some issues with the implementation and/or the controller box sold for the product ... resulting in extremely high cpu utilization for relatively low thruput.
I did the modifications to add rfc 1044 support for the product and
got better than two orders of magnitude improvement in instructions
per byte moved (drastic reduction in pathlength and transferring at
channel media rate). misc past posts mentioning doing the rfc 1044
support
https://www.garlic.com/~lynn/subnetwork.html#1044
as far as I know, the pascal-based implementation ... also never had a
buffer overflow problem ... one of the reasons in the past, I've made
lots of comments about buffer overflows being strongly correlated with
use of C. misc. past posts:
https://www.garlic.com/~lynn/subintegrity.html#overflow
misc. other posts in this thread:
https://www.garlic.com/~lynn/2010j.html#11 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#38 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#51 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#60 Information on obscure text editors wanted
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Fri, 18 Jun 2010 10:28:05 -0400Anne & Lynn Wheeler <lynn@garlic.com> writes:
other rubbing salt into the wound
https://www.garlic.com/~lynn/2000c.html#21 Cache coherence [was Re: TF-1]
as mentioned in this post
https://www.garlic.com/~lynn/2009p.html#57 MasPar compiler and simulator
the retirement of senior executive in fall of '91 kicked off chain of events.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers, phl.media Date: Fri, 18 Jun 2010 11:08:35 -0400jmfbahciv <See.above@aol.com> writes:
somebody's comp.arch post from summer 1994:
I benchmarked a Tandem Integrity S2 machine about a year ago,
that we had for evaluation. The processor performance was 10
Vax MIPS, measured using Dhrystone 2.0. This is about the same
as a 25 Mhz 486 PC.
>
Admittedly the Tandem S2 is a fairly old machine. It has three
MIPS R2000 processors, for redundancy reasons. Performance is
equivalent to one R2000. The price tag was pretty high.
>
There must be better ways of getting redundancy. Loosely coupled
systems, for instance. Of course, then you pay the big money
for the distributed software to make it work.
... snip ...
or not ... misc. past posts mentioning ha/cmp
https://www.garlic.com/~lynn/subtopic.html#hacmp
tandem wiki page
https://en.wikipedia.org/wiki/Tandem_Computers
part of uniform 90 trip report, 22-25jan1990 in washington dc (tandem
claiming AT&T ordering 40k S2s for telecommuncation):
Tandem
Tandem proudly displayed its newest fault tolerant system, the Integrity
S2. A Tandem employee involved in the S2's development said that
AT&T will order 40,000 of these systems. In addition, he said
that Tandem expects to capture fifteen percent of the telecommunications
market in Europe with the S2. As these countries become more open, they
will need to standardize their telecommunications systems. I was told
that this market is estimated to be $100 Billion, and of course,
fifteen percent of $100 Billion is $15 Billion!
The Integrity S2 runs NonStop-UX, Tandem's implementation of UNIX
System V. It does not run GuardianOS, Tandem's proprietary operating
system, and is thus incompatible with Tandem's other products.
NonStop-UX supports NFS, TCP/IP, and X-Window. FORTRAN, C, Pascal,
and MicroFocus COBOL compilers are available. In regards to
database products, the S2 can run either Informix, Oracle, or Ingres
software.
The S2 also has a very different architecture from Tandem's other
products. The S2 is based on the 16.67 MHz MIPS R2000 CPU using
a triple modular redundant architecture. This means that three of
these CPUs are present, each executing the very same instruction stream
and each residing on a separate board.
Each CPU has its own local memory which can be up to 8 MB,
and each CPU attaches to two voters. The voters compare the results
generated by the three CPUs before writing to global memory. Each
voter is associated with a global memory as large as 32 MB.
In the event that a CPU produces a different result from the other
two CPUs, the voters designate that CPU to be faulty, and it is
automatically taken down while the other components continue
operating.
From the standpoint of the end-user, a failure like this is
completely transparent. In fact, a CPU board, I/O processor, or
memory board can actually be replaced while the system is running,
and this capability was demonstrated at UniForum. Each board is
inside of a plastic enclosure and clearly labeled. Therefore, no
special training nor skills are required to replace parts. When
a part is inserted back into the system, this process is referred
to as "reintegration" and normally takes about 90 seconds for the
component to be recognized once again by the other parts.
The Integrity S2 also contains dual data paths, dual ported controllers,
battery backup, and mirrored disks. The maximum DASD supported is
greater than 17 GB. The system enclosure measures 29.5 inches
high by 36 inches wide by 16.4 inches deep, while the mass storage
cabinet is 29.5 inches high by 16.5 inches wide by 16.4 inches deep.
The Integrity S2 comes with one system enclosure and from one to
four mass storage cabinets. Additional CPUs cannot be added, at least
at the present time. Finally, the S2 is priced from $172,000 to
$248,000 depending on the amount of DASD purchased.
Besides Tandem, other vendors also displayed fault tolerant UNIX
systems. These included Stratus, Sequoia, and IMP. Stratus
demonstrated FTX, its UNIX implementation that is now available for
the existing Stratus XA2000 line. Sequoia Systems showed its Series
300 system designed for on-line transaction processing. It runs
TOPIX, Sequoia's version of UNIX. IMP displayed the XR650 which is
based on a dual redundant architecture. Two 25 MHz Motorola
MC68030 CPUs execute the same instructions, and the results are
compared. If they differ, then diagnostics are automatically run
on each CPU in order to determine which one is faulty.
... snip ...
part of uniform 91 trip report, 22-24jan1991 in dallas
Commercial Processing
At last year's conference a paper was presented by an AT&T employee on
a transaction monitor designed for UNIX called Tuxedo. This year
several vendors announced products based on Tuxedo, and NCR introduced
Top End, its own transaction monitor. UNIX has not been popular for
on-line transaction processing, but that is primarily due to a lack
of applications. With the introduction of these transaction monitors,
many more commercial applications for UNIX should become available
during the next few years.
NCR's Top End runs on the NCR System 3000 Family of loosely coupled
multiprocessors. It is based on a client/server architecture and
allows processes to be distributed over multiple parallel processors
at the same time by making several copies of the client. Top End
creates an audit trail and also provides automated recovery of the
database in the event of an error. It works with the C, COBOL, and
C++ languages. Prices range from $3,200 to $150,000 depending on the
hardware configuration, and it will be available Second Quarter 1991.
Sequent and AT&T announced a joint agreement to enhance Tuxedo's
performance in a symmetrical multiprocessing environment. In addition,
Sequent announced new systems based on the Intel 486 CPU. The Symmetry
2000 Series supports from two to thirty tightly coupled CPUs. Each
processor board contains two CPUs that operate independently. These
systems are meant for database processing, and Sequent claimed the
performance to be up to 354 TPS based on an Oracle TP1 benchmark.
Sequent said that this is "the highest performance in the industry"
and claimed a price/performance ratio "three to five times better
than traditional superminicomputers and seven to ten times better
than mainframes." A uniprocessor model starts at $23,500, while a
thirty CPU system sells for $2.5 Million. Sequent's existing install
base was said to be greater than 3,000 systems.
The leading fault tolerant vendors were also present at UniForum.
Tandem demonstrated its one year old Integrity S2. Boards were pulled
out of the system as it continued to run. Stratus went even further
in its demonstration. Besides pulling boards out of the running system,
Stratus shut the power off and a few seconds later turned the power
back on. The system began running again by itself right where it had
left off as if nothing had happened. FTX, Stratus's version of UNIX
that was announced at last year's UniForum, has been available for the
past month.
Sequoia, the third largest fault tolerant vendor, also had a large
exhibit. TOPIX, the Sequoia Operating System, has always been based
on UNIX. The Sequoia Series 300 uses the Motorola 68030 CPU and is
claimed to support from two to 64 processor boards. However, the
largest system shipped to date consists of 16 processor boards. The
Sequoia system is a shared memory computer containing duplexed memory
boards, and each processor board contains dual self-checking CPUs.
There are four processor boards per cabinet, and every board in the
system has its own power supply associated with it, allowing any
type of hardware repair to be performed on-line.
Pyramid, known for its multiprocessor UNIX servers based on a
proprietary RISC architecture, is now positioning itself against the
fault tolerant vendors. "TUFF-ENUFF" was the marketing phrase used
to describe Pyramid's "new" Reliant Series. The Reliant Series is
actually two separate systems connected to a Gandalf StarMaster, which
appeared to be some sort of terminal server. The disks are mirrored,
and there is a UPS. The two systems are existing Pyramid MIServer
Systems, each supporting from one to twelve CPUs. There is new
software though, and it provides automatic switchover in the event
that one system fails. Service is restored within three minutes.
Although Pyramid claims that the Reliant Series is a low cost
alternative to fault tolerance, a typical configuration sells for
$800,000. Pyramid supplies both AT&T and Olivetti with systems.
... snip ...
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: A "portable" hard disk Newsgroups: alt.folklore.computers Date: Fri, 18 Jun 2010 20:28:32 -0400re:
Water flowing over top of Grand Coulee Dam
http://www.krem.com/news/local/Water-flowing-over-top-of-Grand-Coulee-Dam-96617344.html
from above:
"The last time Grand Coulee Dam looked like this was 1997. Most people
can't remember that far back. But this is even bigger. That spill is
three feet deep and pouring 500 thousand gallons of water every second,"
... snip ...
original dam just had the two powerhouses and quite a bit of water went over the spillway ... but after 3rd powerhouse was built, there was little water left for the spillway. w/o the tremendous amount of water going over the spillway ... there has been lot less wear&tear at the bottom of the spillway.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers Date: Sat, 19 Jun 2010 10:59:23 -0400Louis Krupp <lkrupp_nospam@indra.com.invalid> writes:
the unofficial system for 360/67, cp67/cms was with dozen or fewer
people (I've commented that the difference was close to two orders of
magnitude for much of the period). cp67/cms was done at the science
center on 4th flr of 545 tech sq ... with some number of the people
having worked on ctss (other ctss people went to work on multics on 5th
flr of 545 tech sq). misc. past posts mentioning 545 tech sq.
https://www.garlic.com/~lynn/subtopic.html#545tech
ctss at mit:
https://www.multicians.org/thvv/7094.html
https://en.wikipedia.org/wiki/Compatible_Time-Sharing_System
Michigan did MTS for 360/67 for much of the same reason that
the science center did MTS. some reference here:
http://www.eecis.udel.edu/~mills/gallery/gallery8.html
other MTS pages have gone 404 ... but still are at wayback machine:
https://web.archive.org/web/20050212073808/www.itd.umich.edu/~doc/Digest/0596/feat01.html
https://web.archive.org/web/20050212073808/www.itd.umich.edu/~doc/Digest/0596/feat02.html
https://web.archive.org/web/20050212183905/www.itd.umich.edu/~doc/Digest/0596/feat03.html
Note that Stanford did Orvyl for 360/67.
http://www.stanford.edu/~guertin/manuals/SPIDES.HTML#I.1.1.3.4
Orvyl (and MTS) were later ported to 370:
http://www.slac.stanford.edu/spires/explain/manuals/ORVMAN.HTML
the period at Stanford is possibly better remembered for the Wylbur
editor (part of Orvyl):
http://www.stanford.edu/dept/its/support/wylorv/
http://texteditors.org/cgi-bin/wiki.pl?Wylbur
other reference
http://datacenter.cit.nih.gov/interface/interface206/if206-01.htm
http://www.depts.ttu.edu/itts/documentation/wylbur/wylbur.html
melinda's history of cp67 & vm370 talks about early days of 360/67 and
many univ. bought 360/67 in anticipation of tss/360. When tss/360 didn't
materialize like expected, that provided motivation for developing other
uses for 360/67.
https://www.leeandmelindavarian.com/Melinda#VMHist
available in a number of formats:
https://www.leeandmelindavarian.com/Melinda/25paper.listing
https://www.leeandmelindavarian.com/Melinda/25paper.ps
https://www.leeandmelindavarian.com/Melinda/25paper.pdf
recent posts mentioning ctss:
https://www.garlic.com/~lynn/2010.html#11 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#22 Why is JCL so bad was Re: Basic question on passing JCL set symbol to proc
https://www.garlic.com/~lynn/2010.html#74 1964 CTSS film
https://www.garlic.com/~lynn/2010b.html#31 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010b.html#97 "The Naked Mainframe" (Forbes Security Article)
https://www.garlic.com/~lynn/2010c.html#14 Processes' memory
https://www.garlic.com/~lynn/2010e.html#12 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
https://www.garlic.com/~lynn/2010e.html#14 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
https://www.garlic.com/~lynn/2010e.html#16 Senior Java Developer vs. MVS Systems Programmer (warning: Conley rant)
https://www.garlic.com/~lynn/2010f.html#73 Handling multicore CPUs; what the competition is thinking
https://www.garlic.com/~lynn/2010g.html#9 Far and near pointers on the 80286 and later
https://www.garlic.com/~lynn/2010g.html#53 Far and near pointers on the 80286 and later
https://www.garlic.com/~lynn/2010h.html#53 IBM 029 service manual
https://www.garlic.com/~lynn/2010i.html#10 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010j.html#11 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#38 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#50 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#60 Information on obscure text editors wanted
earlier posts in this thread:
https://www.garlic.com/~lynn/2010j.html#43 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#45 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#46 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#49 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#52 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#53 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#54 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#55 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#56 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#57 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#58 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#59 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#62 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#64 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#65 Article says mainframe most cost-efficient platform
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers Date: Sat, 19 Jun 2010 11:58:03 -0400Louis Krupp <lkrupp_nospam@indra.com.invalid> writes:
past post in thread on resurrecting some old diskettes (with borland
software, pascal in additions to C)
https://www.garlic.com/~lynn/2006s.html#57 Turbo C 1.5 (1987)
in 1984, there was somebody (in rochester) that did several enhancements ... from somewhat long ago and far away:
I have sent you the following file MC PAS -- pascal source for MicroCalc MC HLP -- help file for MicroCalc MCDEMO MSC -- demonstration file MC COM -- compiled verion that matches source Borland stated the source was public domain for Turbo V2.0. See prologue for some of the changes I have made. All I ask is that if you have any neat ideas for changes, that you send them to me. If you find any problems, also please send them. I am still trying to find out why I can not put this on IBMPC. I am adding right justified and centered strings like 1-2-3. "string ::= right justified in cell(s)" 'string := Centered in cell(s) string := left justified -- this is the way it works today I am also looking at sorting the formulas so that forward references will work. I wrote this at home for my Z-80 system first and converted it to the IBM PC.... snip ...
IBMPC was for computing conferencing similar to IBMVM (vmtools originally developed for internal vm370 computer conferencing ... supported usenet like operation, mailing list like operation, as well as a kind of remote file request). There was restrictions on code in PC computer conferencing. Later "pctools" was setup for purpose of code and executable.
The Rochester person then did tinycalc enhancement and placed it on the internal "pctools" distribution
SUBJECT: TinyCalc I have sent you the following file TinyCalc HLP -- help file for MicroCalc TinyDemo TSC -- demonstration file TinyCalc COM -- compiled verion This is a tiny spread sheet with external characteristics like Lotus 123 (TM). I wrote it at home to help me do budgeting and taxes. It supports most of the common spread sheet operation including formula move and copy. This version should run on either a color or mono monitor with 80x25 lines.It is designed to try an limit the amount of main store used. ............. Version 1.01 ................................. Changes: 1.Cleaned up Bounce Bar Menu 2.Added creating of file.BAK for SAVE option so as to not replace the original. 3.Changed Help to allow ESC to quit exit. 4.Changed cursor positioning for pageup and pagedown. .............. Version 2.00 ................................. 1. Support Move, Copy, ClrRange for cells 2. Support merging spread sheets -- be careful what you do here you could really clobber what you have. 3. Support color screen. 4. Improved and shortened hlp 5. Flag cells with formulas that could not convert on Move/Copy with @ERR around term and flashing error indication. 6. Redid screen layout some to use all 80 columns.... snip ...
listserv later evolved on external EARN (bitnet in europe) ... sort of
just the mailing list portion of vmtools
http://www.lsoft.com/products/listserv-history.asp
http://www.lsoft.com/corporate/history_listserv.asp
another version of listserv history:
http://www.livinginternet.com/l/li.htm
http://www.livinginternet.com/l/lli.htm
misc. past posts mentioning bitnet (&/or earn)
https://www.garlic.com/~lynn/subnetwork.html#bitnet
old email from person setting up EARN ... looking for some assistance:
https://www.garlic.com/~lynn/2001h.html#email840320
in this post
https://www.garlic.com/~lynn/2001h.html#65 UUCP email
bitnet history reference:
http://www.livinginternet.com/u/ui_bitnet.htm
misc. past posts mentioning internal network
https://www.garlic.com/~lynn/subnetwork.html#internalnet
in the late 70s and early 80s, I had gotten blamed for computer
conferencing on the internal network (which was larger than the
arpanet/internet from just about the beginning until sometime late 85
or early 86). misc. past posts mentioning computer mediated
conversation:
https://www.garlic.com/~lynn/subnetwork.html#cmc
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Article says mainframe most cost-efficient platform Newsgroups: alt.folklore.computers Date: Sat, 19 Jun 2010 12:34:59 -0400"Charlie Gibbs" <cgibbs@kltpzyxm.invalid> writes:
recent reference:
https://www.garlic.com/~lynn/2010i.html#82 Favourite computer history books?
for tiny drift ... the above mentions reworking the declassification tag
that they gave me for the ww2 army reports (had to show up in image of
each page photographed)
https://www.garlic.com/~lynn/dectag.jpg
there is a little similarity to the way the tag appears (on pages) and
the "serial number" that IBM retrofitted to all the corporate copying
machines in the early 70s. there was an incident where some copied
documentation regarding 370 virtual memory (before virtual memory had
been announced) found its way to a journalist. after a lot of
investigation (where they couldn't prove ... just had suspicions),
they went about retrofitting tiny serial numbers under the glass of
every corporate copier. The serial number then would appear on every
page made on that copier (providing ability to at least trace back to
what copier was used). recent mention in another thread:
https://www.garlic.com/~lynn/2010j.html#32 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: RISC load-store verses x86 Add from memory. Newsgroups: comp.arch Date: Sat, 19 Jun 2010 16:52:00 -0400Brett Davis <ggtgp@yahoo.com> writes:
in that time-frame there was effort to replace large variety of different internal microprocessors with 801. 801 iliad chips had specialized features for aid in emulation.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: taking down the machine - z9 series Newsgroups: bit.listserv.ibm-main, alt.folklore.computers Date: Mon, 21 Jun 2010 10:12:00 -0400eamacneil@YAHOO.CA (Ted MacNEIL) writes:
description (some) of 3081 technology
http://www.jfsowa.com/computer/memo125.htm
misc. past posts mentioning frying 3081 TCMs
https://www.garlic.com/~lynn/2000b.html#36 How to learn assembler language for OS/390 ?
https://www.garlic.com/~lynn/2000b.html#38 How to learn assembler language for OS/390 ?
https://www.garlic.com/~lynn/2001k.html#4 hot chips and nuclear reactors
https://www.garlic.com/~lynn/2002d.html#13 IBM Mainframe at home
https://www.garlic.com/~lynn/2004p.html#35 IBM 3614 and 3624 ATM's
https://www.garlic.com/~lynn/2004p.html#36 IBM 3614 and 3624 ATM's
https://www.garlic.com/~lynn/2004p.html#41 IBM 3614 and 3624 ATM's
https://www.garlic.com/~lynn/2008i.html#96 A Blast from the Past
https://www.garlic.com/~lynn/2009b.html#77 Z11 - Water cooling?
https://www.garlic.com/~lynn/2010d.html#43 What was old is new again (water chilled)
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 1952 use of transistors in computers Newsgroups: alt.folklore.computers, misc.transport.rail.americas Date: Mon, 21 Jun 2010 13:15:45 -0400Philip Nasadowski <nasadowsk@usermale.com> writes:
from above:
IBM designed IMS with Rockwell and Caterpillar starting in 1966 for the
Apollo program. IMS's challenge was to inventory the very large bill of
materials (BOM) for the Saturn V moon rocket and Apollo space
vehicle. However, by some accounts it was accepted too late in the
process to make significant contributions to the Apollo program.
... snip ...
recent posts mentioning IMS
https://www.garlic.com/~lynn/2010d.html#0 PDS vs. PDSE
https://www.garlic.com/~lynn/2010d.html#12 The origins of CICS
https://www.garlic.com/~lynn/2010d.html#62 LPARs: More or Less?
https://www.garlic.com/~lynn/2010d.html#71 LPARs: More or Less?
https://www.garlic.com/~lynn/2010d.html#72 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#50 LPARs: More or Less?
https://www.garlic.com/~lynn/2010e.html#61 z9 / z10 instruction speed(s)
https://www.garlic.com/~lynn/2010e.html#83 Entry point for a Mainframe?
https://www.garlic.com/~lynn/2010f.html#2 Entry point for a Mainframe?
https://www.garlic.com/~lynn/2010f.html#18 What was the historical price of a P/390?
https://www.garlic.com/~lynn/2010f.html#47 Nonlinear systems and nonlocal supercomputing
https://www.garlic.com/~lynn/2010f.html#48 Nonlinear systems and nonlocal supercomputing
https://www.garlic.com/~lynn/2010g.html#28 Intel Nehalem-EX Aims for the Mainframe
https://www.garlic.com/~lynn/2010h.html#66 25 reasons why hardware is still hot at IBM
https://www.garlic.com/~lynn/2010i.html#35 Global CIO: Global Banks Form Consortium To Counter HP, IBM, & Oracle
https://www.garlic.com/~lynn/2010i.html#41 Idiotic programming style edicts
https://www.garlic.com/~lynn/2010i.html#50 Ten examples of why the humble ATM = innovation in 2010
https://www.garlic.com/~lynn/2010i.html#85 Idiotic programming style edicts
https://www.garlic.com/~lynn/2010j.html#48 Knuth Got It Wrong
https://www.garlic.com/~lynn/2010j.html#59 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#62 Article says mainframe most cost-efficient platform
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: IBM 3670 Brokerage Communications System Newsgroups: alt.folklore.computers Date: Thu, 24 Jun 2010 08:28:09 -0400hancock4 writes:
in the mid-90s ... did some work with NIH library and a couple of the
oriignal implementers were still around ... started in the late 60s
using mainframe bdam ... but they had invented their own online
environment (originally CICS had started out as effort at customer shop
... before being selected to be turned into product). recent post
mentioning NLM
https://www.garlic.com/~lynn/2010e.html#9 Adventure - Or Colossal Cave Adventure
early part of the century, I was in datacenter that mentioned well over 100 CICS regions (at the time CICS didn't have multiprocessor support ... so method for taking advantage of additional processors, was to have multiple copies of CICS). the datacenter handled dataprocessing outsourcing for most of tv cable companies in the US ... customer support terminals, billing, sending commands down to customer TV settop boxes, etc
IMS wiki page ... mentions that most of ATM transactions in the
world goes thru some IMS system:
https://en.wikipedia.org/wiki/IBM_Information_Management_System
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Disk replacing Tape? Newsgroups: bit.listserv.ibm-main Date: 24 Jun 2010 08:55:02 -0700SanDisk's SD card can store data for 100 years
from above:
The WORM (write once, read many) card is "tamper proof" and data cannot
be altered or deleted, SanDisk said in a statement. The card is designed
for long-time preservation of crucial data like legal documents, medical
files and forensic evidence, SanDisk said.
... snip ...
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: What is the protocal for GMT offset in SMTP (e-mail) header Newsgroups: comp.mail.misc, alt.folklore.computers, comp.os.linux.advocacy Date: Fri, 25 Jun 2010 00:25:08 -0400bonomi@host122.r-bonomi.com (Robert Bonomi) writes:
starting in the late '60s there were a number of commerical online time-sharing service bureaus formed started out using (virtual machine) cp67 as base (later moving to vm370)
aug76, tymshare started offering their (vm370/cms based) online computer
conferencing to (ibm user group) SHARE ... for "vmshare" (archives back
to aug76)
http://vm.marist.edu/~vmshare/
tymshare wiki
https://en.wikipedia.org/wiki/Tymshare
the above mentions moving tymnet to interdata in 70s. for other topic
drift ... as undergraduate in the 60s ... at the univ. i had to add
tty/ascii support to cp67. as part of doing the tty/ascii support, i
tried to make the terminal controller to do something that it couldn't
quite do. this was somewhat the motivation for the univ. to start a
clone controller project, reverse engineering the mainframe channel
interface, building channel interface board and programming interdata/3
to emulate the standard terminal controller. this was later upgraded to
interdata/4 (handling channel interface) ... with multiple interdata/3
handling line-scanner functions. four of us got written up as being
responsible for clone controller business. misc. past posts
https://www.garlic.com/~lynn/submain.html#360pcm
tymshare had also developed their own 370 timesharing system, GNOSIS. As
part of m/d (now boeing) buying tymshare in the 80s ... I was brought
in for audit of gnosis as part of its spin-off to key logic:
http://cap-lore.com/CapTheory/upenn/Gnosis/Gnosis.html
http://cap-lore.com/CapTheory/upenn/
and wiki refs:
https://en.wikipedia.org/wiki/GNOSIS
https://en.wikipedia.org/wiki/KeyKOS
for other drift, minor note about internal network getting gateway in
fall of '82 (before the arpanet cut-off to internetworking tcp/ip
protocol)
https://www.garlic.com/~lynn/internet.htm#email821022
in this post
https://www.garlic.com/~lynn/internet.htm#0
the internal network was larger than the arpanet/internet from
just about the beginning until sometime late '85 or early '86 ...
some past posts:
https://www.garlic.com/~lynn/subnetwork.html#internalnet
I had gotten blamed for computer conferencing on the internal network in the late 70s and early 80s.
bitnet (& later earn) was corporate sponsored network for educational
institutions using technology similar to that used for the internal
network ... some past posts
https://www.garlic.com/~lynn/subnetwork.html#bitnet
recent post mentioning bitnet/earn
https://www.garlic.com/~lynn/2010j.html#68 Article says mainframe most cost-efficient platform
... slightly related to the web ... two of the people mentioned in
this jan92 meeting reference:
https://www.garlic.com/~lynn/95.html#13
later show up at a small client/server startup responsible for something called the "commerce server". We were brought in to consult because they wanted to do financial transactions on their servers. The startup had also invented this technology called SSL they wanted to use ... oh, btw, the result is no frequently called "electronic commerce".
Part of deploying support for servers doing finanical transfer
transactions was something called a "payment gateway" ... aka sat on the
internet and acted interface between the webservers doing financial
transactions and the payment networks. some past post mentioning
payment gateway
https://www.garlic.com/~lynn/subnetwork.html#gateway
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: What is the protocal for GMT offset in SMTP (e-mail) header Newsgroups: comp.mail.misc, alt.folklore.computers, comp.os.linux.advocacy Date: Fri, 25 Jun 2010 10:23:55 -0400Anne & Lynn Wheeler <lynn@garlic.com> writes:
HONE was one of the largest online virtual machine based systems that
started with cp67 ... after the 23jun69 unbundling announcement
https://www.garlic.com/~lynn/submain.html#unbundle
it evolved into providing internal worldwide corporate sales & marketing
support ... with HONE clones sprouting up all over the
world. misc. past posts mentioning commercial online timesharing service
bureaus.
https://www.garlic.com/~lynn/submain.html#timeshare
recent post mentioning HONE and joke about four shift workweek:
https://www.garlic.com/~lynn/2010i.html#85 Idiotic programming style edicts
local (vm) group train trip from oakland to reno. In the mid-70s, HONE
had consolidated various US datacenters on the pennisula; do
(sat. photo) address lookup for the (relatively new) facebook
bldg. ... it was in the bldg. next door. misc. past posts mentioning
HONE
https://www.garlic.com/~lynn/subtopic.html#hone
Date: 17 January 1986, 16:11:35 PST
From: xxxxx at HONE9 <HONE VM Systems Support>
To: distribution
Subject: VM Training
Hi -
I plan to call my travel agent tomorrow (Sat.) with the list of attendees
to get our reservations made. If there is anyone who has had a change of
plans and can NOT participate, please let me know tonight or tomorrow
morning.
I will be at Eric's for Lynn's weekly Friday symposium later today and
can be reached at home at (415) xxx-xxxx.
... snip ... top of post, old email index, HONE email
references from recent facebook discussion about san jose plant site:
When Eric's first opened across from bldg. 28 ... the backroom was
usually kept closed ... but for some reason they had my name on the
door and they would open it up fri. nights. they would also give us
half-price on pitchers of anchor steam
... snip ...
comment about san jose plant site earthquake retrofit
some people that complained were issued industrial hearing protection
gear; seemed to discount productivity issues in noisy workplace (this
is research, we don't need to worry about productivity?).
they started with bldg. 14 ... moved everybody to offsite bldg (86)
while doing construction for the earthquake retrofit. Then doing
bldg. 28 ... which was external addition wrapped around the old
bldg. ... story was that old bldg. was attached/hung from the external
skeleton. They didn't evacuate "old" bldg. 28 during construction and
the noise was really deafening ... there were some complaints filed
with gov. about noise in the workplace.
... snip ...
one of the bldg. 28 distinctions was that was where the work on the
original relational/sql System/R occurred on vm370 370/145 ... some past
posts
https://www.garlic.com/~lynn/submain.html#systemr
plant site sat. photo:
5600 Cottle Rd ... (at least) bldg 12, 25, 26, 28 (old research),
homestead, pond between bldg 28 & homestead ... various other ... all
gone.
http://maps.google.com/maps?f=q&source=s_q&hl=en&geocode&q=5600+Cottle+Road%2C+San+Jose%2C+CA&sll=37.0625%2C-95.677068&sspn=65.047864%2C48.779297&ie=UTF8&hq&hnear=5600+Cottle+Rd%2C+San+Jose%2C+Santa+Clara%2C+California+95123&ll=37.245293%2C-121.799154&spn=0.015869%2C0.011909&t=h&z=16
msnmaps is still showing 1998 satellite photos, before all the bldgs
had been plowed under.
http://msrmaps.com/image.aspx?T=1&S=11&Z=10&X=1516&Y=10306&W=1&qs=5600+cottle+rd|san+jose|ca|&Addr=5600+Cottle+Rd%2c+San+Jose%2c+CA+95123-3696&ALon=-121.8040000&ALat=37.2495300
... snip ...
for a little drift back to internet ... misc. past posts about being
in booth at interop '88
https://www.garlic.com/~lynn/subnetwork.html#interop88
there were monthly (baybunch) vm user group meetings at SLAC ... which regularly saw several of the local customers (like tymshare) as well as number of vendors (Amdahl, NAS, 2pi, etc)
slac also had the first webserver (outside cern) on their vm370 system:
https://ahro.slac.stanford.edu/wwwslac-exhibit
reference to original cern webserver being hacked up sgml
http://infomesh.net/html/history/early/
SGML was the ISO standard version of GML ... misc. past posts
https://www.garlic.com/~lynn/submain.html#sgml
which had been invented at the science center in 1969 ... science
center was 4th flr, 545 tech sq ... misc. past posts
https://www.garlic.com/~lynn/subtopic.html#545tech
which was also did virtual machine system cp67/cms as well as the technology for the internal network (later also saw use in bitnet & earn).
melinda has history reference to early period
https://www.leeandmelindavarian.com/Melinda#VMHist
available in a number of formats:
https://www.leeandmelindavarian.com/Melinda/25paper.listing
https://www.leeandmelindavarian.com/Melinda/25paper.ps
https://www.leeandmelindavarian.com/Melinda/25paper.pdf
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Fri, 25 Jun 2010 11:15:49 -0400Huge <Huge@nowhere.much.invalid> writes:
endicott mid-range 370 4341/4331 also saw similar big upswing selling in the same market segment ... selling about the same as vax in single or few machines at time ... but overall outsold because of large corporate orders of several hundreds at a time.
there was similar volume expectation for the 4341/4331 follow-ons (4381 & 4361) ... which didn't materialize ... seeing similar effect to the vax numbers (reference above) with the mid-range market starting to move to workstations and larger PCs in the mid-80s.
370 mainframe sales survived this ... but it was mostly the high-end machines ... which then started to see the "killer micro" scenario in the 90s.
some posts in recent mainframe thread:
https://www.garlic.com/~lynn/2010j.html#43 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#45 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#46 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#49 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#52 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#53 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#54 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#55 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#56 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#57 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#58 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#59 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#62 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#64 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#65 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#67 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#68 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#69 Article says mainframe most cost-efficient platform
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: What is the protocal for GMT offset in SMTP (e-mail) header Newsgroups: comp.mail.misc, alt.folklore.computers, comp.os.linux.advocacy, microsoft.public.mail.misc Date: Fri, 25 Jun 2010 19:17:11 -0400Morten Reistad <first@last.name> writes:
possibly part of the nsfnet & AUP was telcos. going into mid-80s, there was huge amount of excess bandwidth (dark fiber, other stuff) ... but big telcos were in chicken&egg scenario; they had huge fixed run-rate ... new bandwidth hungry applications wouldn't appear until tariffs came down significantly ... and tariffs wouldn't come down significantly w/o the new generation of bandwidth hungry apps (i.e. just lowering tarrifs significantly would have resulted in big telcos operating at loss for extended period until the new generation of bandwidth hungry apps appeared).
winning nsfnet backbone bid was $11.2M ... but estimate that commercial interests contributed well over four times that for the backbone ... with the possible hope that there wouldn't be any commercial traffic bleeding off onto the backbone ... but at the same time providing significant available bandwidth as technology incubator for the evolution of the bandwidth hungry applications.
misc. old email mentioning nsfnet backbone RFP
https://www.garlic.com/~lynn/lhwemail.html#nsfnet
we had T1 operational with internal backbone prototype ... and was anticipating getting the nsfnet backbone RFP (some conjecture that one of the reasons that the backbone RFP specified T1 ... was our example of operational T1 infrastructure). there were some internal politics and we weren't allowed to bid on the RFP. The director of NSF tried to help by writing a letter to the company 3Apr1986, NSF Director to IBM Chief Scientist and IBM Senior VP and director of Research, copying IBM CEO) ... ... but that just resulted in aggravating the internal politics (little statements like what we already had running was at least five years ahead of all RFP bid submissions ... to build something new).
misc. past posts mentioning nsfnet backbone
https://www.garlic.com/~lynn/subnetwork.html#nsfnet
btw, the winning nsfnet T1 backbone bid ... didn't actually install T1 links ... 440kbit links were installed ... and then somewhat to meet the letter of the RFP ... T1 trunks were installed with telco multiplexor running multiple 440kbit links over the T1 trunks.
then possibly trying to quiet the sniping ... I was asked to be the red-team on the nsfnet T3 backbone rfp ... while something like 30 people from 7-8 labs around the world were the blue team. at the final review ... I got to present first ... followed by the blue team. something like 5-10 minutes into the blue team presentation ... the person running the review, pounded on the table and said that they would lay down in front of a garbage truck before they let any but the blue team proposal go forward.
it turns out that some of the people involved in the nsfnet backbone
internal politics were also later involved in transferring the cluster
scale-up work and stating that we weren't allowed to work on anything
with more than four processors. ... reference to jan92 cluster scale-up
meeting before effort was transferred
https://www.garlic.com/~lynn/95.html#13
misc. old email related to cluster scale-up
https://www.garlic.com/~lynn/lhwemail.html#medusa
and past posts mentioning the work
https://www.garlic.com/~lynn/subtopic.html#hacmp
shortly after the transfer at the very end of jan ... there
started to be press things like:
https://www.garlic.com/~lynn/2001n.html#6000clusters1 17feb92
and then something later that summer
https://www.garlic.com/~lynn/2001n.html#6000clusters2 11May92
other posts in this thread:
https://www.garlic.com/~lynn/2010g.html#66 What is the protocal for GMT offset in SMTP (e-mail) header
https://www.garlic.com/~lynn/2010g.html#68 What is the protocal for GMT offset in SMTP (e-mail) header
https://www.garlic.com/~lynn/2010g.html#70 What is the protocal for GMT offset in SMTP (e-mail) header
https://www.garlic.com/~lynn/2010g.html#75 What is the protocal for GMT offset in SMTP (e-mail) header header time-stamp?
https://www.garlic.com/~lynn/2010g.html#76 What is the protocal for GMT offset in SMTP (e-mail) header header time-stamp?
https://www.garlic.com/~lynn/2010g.html#80 What is the protocal for GMT offset in SMTP (e-mail) header time-stamp?
https://www.garlic.com/~lynn/2010g.html#81 What is the protocal for GMT offset in SMTP (e-mail) header time-stamp?
https://www.garlic.com/~lynn/2010g.html#85 What is the protocal for GMT offset in SMTP (e-mail) header
https://www.garlic.com/~lynn/2010h.html#0 What is the protocal for GMT offset in SMTP (e-mail) header time-stamp?
https://www.garlic.com/~lynn/2010h.html#4 What is the protocal for GMT offset in SMTP (e-mail) header
https://www.garlic.com/~lynn/2010h.html#9 What is the protocal for GMT offset in SMTP (e-mail) header
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: zVM training Newsgroups: bit.listserv.ibm-main Date: 25 Jun 2010 18:11:00 -0700gahenke@GMAIL.COM (George Henke) writes:
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Idiotic programming style edicts Newsgroups: alt.folklore.computers Date: Sat, 26 Jun 2010 09:23:39 -0400Quadibloc <jsavard@ecn.ab.ca> writes:
maintenance was split off (& separately charged for), hardware interface/operation manuals published, SE services split off (& separately charged for), and starting to charge for application software (although case was made that kernel software continued to be free).
The starting to charge for SE services was part of the original
motivation for HONE ... online cp67 virtual machine services available
to branch office ... some past posts
https://www.garlic.com/~lynn/subtopic.html#hone
prior to unbundling, a lot of SE training was sort of apprentice program ... part of large team, onsite at customer shop. after unbundling ... all those people were being charged for ... and nobody was able to figure out the charge for SEs in-training. HONE (Hands-On Network Environment) started out with several cp67 datacenters ... providing online access for SEs in branch offices to sharpen their skills ... practicing with (guest) operating systems running in virtual machines.
the science center had also done port of apl\360 to cms for cms\apl. there started to be some number of sales & marketing support applications done in cms\apl ... which eventually started to dominate all HONE activity and the SE training activity eventually disappeared.
univ. had started work on clone controller before unbundling ... so
figuring out the channel interface had to be done by reverse
engineering. a few recent posts mentioning clone controller:
https://www.garlic.com/~lynn/2010j.html#11 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#19 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#57 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#75 What is the protocal for GMT offset in SMTP (e-mail) header
as mentioned ... clone controller business was major motivation for
future system effort ... that was going to completely 360/370 and
was significantly different ... misc. past posts
https://www.garlic.com/~lynn/submain.html#futuresys
during future system effort ... the 370 software and hardware product
pipeline was allowed to dry up. some reference here:
http://www.jfsowa.com/computer/memo125.htm
the lack of 370 products during the period is attributed with allowing
clone processors to gain market foothold. with the failure of the future
system effort (w/o even being announced), there was a mad rush to get
stuff back into the 370 product pipelines. this possibly contributed to
deciding to release a lot of 370 work I had been doing all during the
period (as well as sometimes making less than flattering comments about
the future system effort). The decision was also made to start charging
for kernel software ... and my resource manager was selected as the
guinea pig to start the transition for charging for kernel software.
some past posts related to resource management
https://www.garlic.com/~lynn/subtopic.html#fairshare
In Apple II period ... my brother was regional apple market rep (had largest physical region in CONUS). one of the things he had setup was using his Apple II to call into Apple corporate hdqtrs S/38, to monitor the hardware build and delivery schedules.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Percentage of code executed that is user written was Re: Delete all members of a PDS that is allocated Newsgroups: bit.listserv.ibm-main Date: 26 Jun 2010 09:05:21 -0700Peter.Farley@BROADRIDGE.COM (Farley, Peter x23353) writes:
they had activity counter snap shots ... starting with the science center cp67 ... eventually every 5-10 mins, 7x24 for nearly decade. the convention was followed by internal systems installing first cp67 and then vm370 ... so there were hundreds of systems with lots of different workloads & configurations with years of data.
another activity was several system simulators ... some written in pli others in apl ... that could simulate effects of different algorithms. one of the apl simulators ... using the enormous amount of snapshot information evolved into the performance predictor ... made available on the worldwide sales & marketing hone system; customer support people could enter profiles of their customers workload & configuration ... and then ask what-if questions regarding what happens selling an additional mbyte of memory (or other configurations changes) &/or what happens when there are workload changes. in the 90s, rights to a descendent of the performance predictor were sold ... the person ran it thru and apl-to-c convertor and then made a living doing consulting at number of high-end datacenters around the world.
the science center also did some number of other kinds of hot-spot
monitoring. not at cambridge ... but from palo alto science center (by
people that brought you the 370/145 apl microcode assist) ... they
gen'ed up a 145 microcode path for me that sampled instruction address.
I used this for help in identifying pieces of code to be dropped into
microcode for the ECPS performance assist. this is old post with
result of some of the other kinds of analysis done on the kernel
code as part of selecting what went into the microcode:
https://www.garlic.com/~lynn/94.html#21 370 ECPS VM microcode assist
The science center also did a project that started out with the POK "redcap" instruction simulator (recorded information about each instruction simulated ... was used in helping design processors &/or new instruction). Several changes to the information recording were made at the science center ... which was then used as input into a virtual memory simulator. Science center eventually released this as "vs/repack" product ... which included ability to do semi-automated program re-organization. It was used by a number of corporate products ... both as an instruction "hot-spot" identifier ... as well as helping in transition from real storage environment to virtual memory environment (i.e. products like IMS made a lot of use of it).
about the turn of the century had a chance to look at a large, 450+K statement cobol application that ran all night on something like 40+ CECs (billion plus in mainframe equipment). the consultant that was using the descendent of the performance predictor had been called in on the project ... to look for possible system and application bottlenecks. the operation also had a large performance group that had been tending the application for years (decades?) ... mostly with hotspot analysis. the consultant turned up some things that had gone unnoticed by the group doing hot-spot analysis ... but there was still strong feeling that there was significant thruput inefficiencies (in part based on cpu processing/operation from decades earlier and currently).
One of the things done at the science center decades earlier was multiple regression analysis of the activity counters. I decided to try something similar for this application; collected fairly large number of application counts from several nights run across large number of CECs. It turned up some "meta" level activity that was quite anomolous ... and when reworked, resulted in significant thruput improvement. Part of the issue was that the system analytical models and the hot-spot analysis had been highlighting "micro" level activity ... but didn't catch the application doing several anomolous unnecessary passes of whole operations.
misc. past posts mentioning performance predictor &/or "vs/repack":
https://www.garlic.com/~lynn/94.html#7 IBM 7090 (360s, 370s, apl, etc)
https://www.garlic.com/~lynn/99.html#68 The Melissa Virus or War on Microsoft?
https://www.garlic.com/~lynn/2000g.html#30 Could CDR-coding be on the way back?
https://www.garlic.com/~lynn/2001b.html#83 Z/90, S/390, 370/ESA (slightly off topic)
https://www.garlic.com/~lynn/2001c.html#31 database (or b-tree) page sizes
https://www.garlic.com/~lynn/2001c.html#33 database (or b-tree) page sizes
https://www.garlic.com/~lynn/2001i.html#20 Very CISC Instuctions (Was: why the machine word size ...)
https://www.garlic.com/~lynn/2001i.html#46 Withdrawal Announcement 901-218 - No More 'small machines'
https://www.garlic.com/~lynn/2002b.html#64 ... the need for a Museum of Computer Software
https://www.garlic.com/~lynn/2002c.html#28 OS Workloads : Interactive etc
https://www.garlic.com/~lynn/2002c.html#45 cp/67 addenda (cross-post warning)
https://www.garlic.com/~lynn/2002c.html#46 cp/67 addenda (cross-post warning)
https://www.garlic.com/~lynn/2002c.html#49 Swapper was Re: History of Login Names
https://www.garlic.com/~lynn/2002e.html#50 IBM going after Strobe?
https://www.garlic.com/~lynn/2002f.html#50 Blade architectures
https://www.garlic.com/~lynn/2002q.html#28 Origin of XAUTOLOG (x-post)
https://www.garlic.com/~lynn/2003f.html#15 Alpha performance, why?
https://www.garlic.com/~lynn/2003f.html#21 "Super-Cheap" Supercomputing
https://www.garlic.com/~lynn/2003f.html#53 Alpha performance, why?
https://www.garlic.com/~lynn/2003g.html#15 Disk capacity and backup solutions
https://www.garlic.com/~lynn/2003h.html#8 IBM says AMD dead in 5yrs ... -- Microsoft Monopoly vs. IBM
https://www.garlic.com/~lynn/2003j.html#32 Language semantics wrt exploits
https://www.garlic.com/~lynn/2003p.html#29 Sun researchers: Computers do bad math ;)
https://www.garlic.com/~lynn/2004.html#14 Holee shit! 30 years ago!
https://www.garlic.com/~lynn/2004c.html#21 PSW Sampling
https://www.garlic.com/~lynn/2004g.html#42 command line switches [Re: [REALLY OT!] Overuse of symbolic constants]
https://www.garlic.com/~lynn/2004k.html#31 capacity planning: art, science or magic?
https://www.garlic.com/~lynn/2004m.html#22 Lock-free algorithms
https://www.garlic.com/~lynn/2004n.html#55 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2004o.html#7 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2004o.html#10 Multi-processor timing issue
https://www.garlic.com/~lynn/2004q.html#73 Athlon cache question
https://www.garlic.com/~lynn/2004q.html#76 Athlon cache question
https://www.garlic.com/~lynn/2005.html#4 Athlon cache question
https://www.garlic.com/~lynn/2005d.html#1 Self restarting property of RTOS-How it works?
https://www.garlic.com/~lynn/2005d.html#6 [Lit.] Buffer overruns
https://www.garlic.com/~lynn/2005d.html#33 Thou shalt have no other gods before the ANSI C standard
https://www.garlic.com/~lynn/2005d.html#41 Thou shalt have no other gods before the ANSI C standard
https://www.garlic.com/~lynn/2005d.html#48 Secure design
https://www.garlic.com/~lynn/2005h.html#1 Single System Image questions
https://www.garlic.com/~lynn/2005h.html#15 Exceptions at basic block boundaries
https://www.garlic.com/~lynn/2005j.html#12 Performance and Capacity Planning
https://www.garlic.com/~lynn/2005j.html#62 More on garbage collection
https://www.garlic.com/~lynn/2005k.html#17 More on garbage collection
https://www.garlic.com/~lynn/2005m.html#28 IBM's mini computers--lack thereof
https://www.garlic.com/~lynn/2005n.html#18 Code density and performance?
https://www.garlic.com/~lynn/2005o.html#5 Code density and performance?
https://www.garlic.com/~lynn/2005o.html#30 auto reIPL
https://www.garlic.com/~lynn/2005o.html#34 Not enough parallelism in programming
https://www.garlic.com/~lynn/2006b.html#15 {SPAM?} Re: Expanded Storage
https://www.garlic.com/~lynn/2006b.html#17 {SPAM?} Re: Expanded Storage
https://www.garlic.com/~lynn/2006b.html#23 Seeking Info on XDS Sigma 7 APL
https://www.garlic.com/~lynn/2006e.html#20 About TLB in lower-level caches
https://www.garlic.com/~lynn/2006e.html#46 using 3390 mod-9s
https://www.garlic.com/~lynn/2006f.html#22 A very basic question
https://www.garlic.com/~lynn/2006f.html#30 A very basic question
https://www.garlic.com/~lynn/2006g.html#34 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#25 The Pankian Metaphor
https://www.garlic.com/~lynn/2006i.html#37 virtual memory
https://www.garlic.com/~lynn/2006j.html#18 virtual memory
https://www.garlic.com/~lynn/2006j.html#22 virtual memory
https://www.garlic.com/~lynn/2006j.html#24 virtual memory
https://www.garlic.com/~lynn/2006l.html#3 virtual memory
https://www.garlic.com/~lynn/2006l.html#11 virtual memory
https://www.garlic.com/~lynn/2006o.html#23 Strobe equivalents
https://www.garlic.com/~lynn/2006o.html#25 CPU usage for paging
https://www.garlic.com/~lynn/2006o.html#26 Cache-Size vs Performance
https://www.garlic.com/~lynn/2006r.html#12 Trying to design low level hard disk manipulation program
https://www.garlic.com/~lynn/2006s.html#24 Curiousity: CPU % for COBOL program
https://www.garlic.com/~lynn/2006t.html#28 Why these original FORTRAN quirks?
https://www.garlic.com/~lynn/2006x.html#1 IBM sues maker of Intel-based Mainframe clones
https://www.garlic.com/~lynn/2006x.html#16 The Future of CPUs: What's After Multi-Core?
https://www.garlic.com/~lynn/2007g.html#31 Wylbur and Paging
https://www.garlic.com/~lynn/2007k.html#65 Non-Standard Mainframe Language?
https://www.garlic.com/~lynn/2007m.html#55 Capacity and Relational Database
https://www.garlic.com/~lynn/2007o.html#53 Virtual Storage implementation
https://www.garlic.com/~lynn/2007o.html#57 ACP/TPF
https://www.garlic.com/~lynn/2007r.html#68 High order bit in 31/24 bit address
https://www.garlic.com/~lynn/2007s.html#41 Age of IBM VM
https://www.garlic.com/~lynn/2007u.html#21 Distributed Computing
https://www.garlic.com/~lynn/2008c.html#24 Job ad for z/OS systems programmer trainee
https://www.garlic.com/~lynn/2008c.html#78 CPU time differences for the same job
https://www.garlic.com/~lynn/2008d.html#35 Interesting Mainframe Article: 5 Myths Exposed
https://www.garlic.com/~lynn/2008e.html#16 Kernels
https://www.garlic.com/~lynn/2008f.html#36 Object-relational impedence
https://www.garlic.com/~lynn/2008l.html#81 Intel: an expensive many-core future is ahead of us
https://www.garlic.com/~lynn/2008m.html#42 APL
https://www.garlic.com/~lynn/2008m.html#69 Speculation ONLY
https://www.garlic.com/~lynn/2008p.html#41 Automation is still not accepted to streamline the business processes... why organizations are not accepting newer technologies?
https://www.garlic.com/~lynn/2008q.html#65 APL
https://www.garlic.com/~lynn/2009d.html#5 Why do IBMers think disks are 'Direct Access'?
https://www.garlic.com/~lynn/2009h.html#76 A Math Geek's Plan to Save Wall Street's Soul
https://www.garlic.com/~lynn/2009l.html#43 SNA: conflicting opinions
https://www.garlic.com/~lynn/2009r.html#17 How to reduce the overall monthly cost on a System z environment?
https://www.garlic.com/~lynn/2010d.html#62 LPARs: More or Less?
https://www.garlic.com/~lynn/2010j.html#48 Knuth Got It Wrong
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Percentage of code executed that is user written was Re: Delete all members of a PDS that is allocated Newsgroups: bit.listserv.ibm-main, alt.folklore.computers Date: Sat, 26 Jun 2010 12:24:13 -0400re:
for slightly different slant on the subject ... in the early 70s ... Amdahl gave a talk at MIT in large auditorium on subject of forming his own clone processor company. one of the questions from student in the audience was what was the pitch/spin he used to get venture capital & funding for his corporation. his reply was something about customers had already spent billions in 360 software development ... that even if IBM was to completely walk away from 360/370 ... that software base would be enough to keep him in business through the end of the century.
the reference to possibility of IBM completely walking away from 360/370
... might be considered a veiled reference to future system effort
... which was planning on doing just that. misc. past posts mentioning
future system
https://www.garlic.com/~lynn/submain.html#futuresys
this has some reference to the future system period letting 370 software
& hardware product pipeline go dry:
http://www.jfsowa.com/computer/memo125.htm
and this fergus & morris book reference
https://www.garlic.com/~lynn/2001f.html#33 IBM's "VM for the PC" c.1984??
makes mention that letting the 370 product pipeline go dry, in large part contributed to letting clone processor vendors gain market foothold.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Percentage of code executed that is user written was Re: Delete all members of a PDS that is allocated Newsgroups: bit.listserv.ibm-main, alt.folklore.computers Date: Sat, 26 Jun 2010 14:14:05 -0400Al Kossow <aek@bitsavers.org> writes:
come from reply to somebody's post in 2001 ... their post included the
reference ... since in was 2001 ... it most likely wasn't the 2003 book
https://www.garlic.com/~lynn/2001f.html#33 IBM's "VM for the PC" c.1984??
so presumably has to the Ferguson & Morris 1993 book
https://www.amazon.com/Computer-Wars-Future-Western-Technology/dp/0812921569
another in similar genre is:
https://www.ecole.org/en/session/49-the-rise-and-fall-of-ibm
https://www.ecole.org/en/session/49-the-rise-and-fall-of-ibm
from above:
After 40 years of unrivalled success, IBM is now in serious trouble.
What has happened? Jean-Jacques Duby explains how the company's values
and the cogs and wheels of its internal management system doomed IBM to
failure, in the light of long developments in the technical, economic
and commercial environment. But why there should have been such a sudden
shock remains a mystery. Perhaps IBM's mighty power had delayed its
downfall, making this all the more brutal as a result, like the
earthquake which follows the sudden encounter of two continental plates.
... snip ...
sort of at the height of the killer micros ... & after the company had
gone in the red ... recent reference
https://www.garlic.com/~lynn/2010j.html#77 Idiotic programming style edicts
also from same article ... referring to at least part of future system
motivation was clone controller business:
IBM tried to react by launching a major project called the 'Future
System' (FS) in the early 1970's. The idea was to get so far ahead
that the competition would never be able to keep up, and to have such
a high level of integration that it would be impossible for
competitors to follow a compatible niche strategy. However, the
project failed because the objectives were too ambitious for the
available technology. Many of the ideas that were developed were
nevertheless adapted for later generations. Once IBM had acknowledged
this failure, it launched its 'box strategy', which called for
competitiveness with all the different types of compatible
sub-systems. But this proved to be difficult because of IBM's cost
structure and its R&D spending, and the strategy only resulted in
a partial narrowing of the price gap between IBM and its rivals.
... snip ...
misc posts mentioning future system
https://www.garlic.com/~lynn/submain.html#futuresys
misc. posts mentioning clone controllers
https://www.garlic.com/~lynn/submain.html#360pcm
and a couple recent references to clone controllers:
https://www.garlic.com/~lynn/2010.html#4 360 programs on a z/10
https://www.garlic.com/~lynn/2010b.html#64 SYSENTER/SYSEXIT_vs._SYSCALL/SYSRET
https://www.garlic.com/~lynn/2010d.html#53 SWTL and 522
https://www.garlic.com/~lynn/2010e.html#62 z9 / z10 instruction speed(s)
https://www.garlic.com/~lynn/2010f.html#9 Entry point for a Mainframe?
https://www.garlic.com/~lynn/2010f.html#11 IBM And Microsoft Clash Over Unbundling Policy
https://www.garlic.com/~lynn/2010g.html#34 someone smarter than Dave Cutler
https://www.garlic.com/~lynn/2010h.html#18 How many mainframes are there?
https://www.garlic.com/~lynn/2010h.html#33 45 years of Mainframe
https://www.garlic.com/~lynn/2010j.html#11 Information on obscure text editors wanted
https://www.garlic.com/~lynn/2010j.html#19 Personal use z/OS machines was Re: Multiprise 3k for personal Use?
https://www.garlic.com/~lynn/2010j.html#57 Article says mainframe most cost-efficient platform
https://www.garlic.com/~lynn/2010j.html#75 What is the protocal for GMT offset in SMTP (e-mail) header
https://www.garlic.com/~lynn/2010j.html#80 Idiotic programming style edicts
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: USPTO Grants Bezos Patent on '60s-Era Chargebacks Newsgroups: alt.folklore.computers Date: Sat, 26 Jun 2010 14:29:47 -0400USPTO Grants Bezos Patent on '60s-Era Chargebacks
from above ...
Another example of why it's not wise to grant software patents when
people don't know much about computer history.
... snip ...
the charging scenarios came up frequently in the various online
commercial time-sharing service bureaus ... misc. reference:
https://www.garlic.com/~lynn/submain.html#timeshare
recent posts with one of the "recoverables" was that the equipment (in
the 60s) was leased ... and companies being billed based on the "system
meter":
https://www.garlic.com/~lynn/2010g.html#13 An Interview with Watts Humphrey, Part 6: The IBM 360
https://www.garlic.com/~lynn/2010g.html#57 An Interview with Watts Humphrey, Part 6: The IBM 360
https://www.garlic.com/~lynn/2010j.html#13 A "portable" hard disk
https://www.garlic.com/~lynn/2010j.html#34 Idiotic programming style edicts
https://www.garlic.com/~lynn/2010j.html#35 IBM Rational Developer for System z
https://www.garlic.com/~lynn/2010j.html#37 Idiotic programming style edicts
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: USPTO Grants Bezos Patent on '60s-Era Chargebacks Newsgroups: alt.folklore.computers Date: Sat, 26 Jun 2010 19:00:51 -0400John Levine <johnl@iecc.com> writes:
online timesharing service bureaus frequently tried to shift more use offshift & low-usage periods by offering various kinds of pricing incentives (with peak charges for periods of peak demands) ... somewhat analogous to telco offshift tariffs.
dynamically predicting low-usage gets more interesting ... somewhat akin to what i was doing as undergraduate in the 60s with dynamic adaptive resource management and (attempting) scheduling to the bottleneck.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: taking down the machine - z9 series Newsgroups: alt.folklore.computers Date: Sun, 27 Jun 2010 08:19:22 -0400hancock4 writes:
Solar's Great Leap Forward
http://www.technologyreview.com/energy/25565/page1/
Solar prices fall due to nanotech
http://www.reportage-enviro.com/2010/05/solar-prices-reduce-thanks-to-nanotechnology/
Highly Efficient Solar Cells Could Result from Quantum Dot Research
http://www.nanotech-now.com/news.cgi?story_id=38787
Nanotech could make solar energy as easy and cheap as growing grass
http://www.physorg.com/news109253445.html
.. some amount of trade-off between cost of manufacturing versus efficiency. satellites would see some of this since quite a bit of the cost is putting weight into orbit. higher efficiency (power per pound) would easily pay for increased cost of manufacturing.
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: taking down the machine - z9 series Newsgroups: alt.folklore.computers Date: Sun, 27 Jun 2010 08:29:34 -0400Huge <Huge@nowhere.much.invalid> writes:
... originally over million across were planned, but currently only
little under 700k acres
https://en.wikipedia.org/wiki/Columbia_Basin_Project
http://users.owt.com/chubbard/gcdam/html/irrigate.html
https://web.archive.org/web/20111007024826/http://www.usbr.gov/pn/project/columbia_index.html
http://www.nwcouncil.org/history/ColumbiaBasinProject.asp
--
42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: taking down the machine - z9 series Newsgroups: alt.folklore.computers Date: Sun, 27 Jun 2010 08:59:23 -0400Quadibloc <jsavard@ecn.ab.ca> writes:
there was article in the 60s about solar power generation ... that the power utilities were in conflict ... solar power farms where power passed thru current utility facilities and they still collected revenue vis-a-vis solar panels on homes where utilities wouldn't collecting fees.
past post about annual economic conference ... making reference to
congress is the most corrupt institution on earth ... and going to flat
tax would eliminate enormous amount of the corruption ... since it would
eliminate the enormous amounts of money spent on lobbying for special
tax provisions
https://www.garlic.com/~lynn/2008m.html#49 Taxes
https://www.garlic.com/~lynn/2008m.html#87 Fraud due to stupid failure to test for negative
https://www.garlic.com/~lynn/2009e.html#53 Are the "brightest minds in finance" finally onto something?
https://www.garlic.com/~lynn/2009h.html#20 China's yuan 'set to usurp US dollar' as world's reserve currency
https://www.garlic.com/~lynn/2009i.html#13 64 Cores -- IBM is showing a prototype already
https://www.garlic.com/~lynn/2009p.html#31 Opinions on the 'Unix Haters' Handbook
https://www.garlic.com/~lynn/2009p.html#39 Opinions on the 'Unix Haters' Handbook
https://www.garlic.com/~lynn/2009q.html#77 Now is time for banks to replace core system according to Accenture
https://www.garlic.com/~lynn/2010d.html#49 search engine history, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010f.html#40 F.B.I. Faces New Setback in Computer Overhaul
regardless of all the other pros & cons regarding "flat tax" ... significantly reducing the enormous corruption would be sufficient justification. there was also reference to going to flat-tax could increase GDP by 3-6% because of enormous resources devoted to dealing with current (complex) tax provisions (i.e. lost productivity could be applied to something more useful). they concluded with semi-humorous about ireland was (also) lobbying against US flat tax ... since companies had given complexity dealing with US tax code as part of motivation for setting up in ireland.
read an article recently ... where barney frank canceled all fund raisers and meetings with special interests during work on the financial regulation bill ... to avoid being exposed to demands. the article also expressed surprise that frank had opened the deliberations to cspan ... which have been doing quite a bit of reruns (a lot of frank's quips during the deliberations were humorous ... don't know if he was just playing to the cameras or not).
--
42yrs virtualization experience (since Jan68), online at home since Mar1970