From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Problem with XP scheduler? Newsgroups: microsoft.public.win32.programmer.kernel,alt.folklore.computers,comp.os.ms-windows.programmer.nt.kernel-mode Date: Fri, 01 Jan 2010 10:24:31 -0500Peter Flass <Peter_Flass@Yahoo.com> writes:
thnks, comments I had seen were DEC (former VMS) people going to m'soft to do NT ... and lots of DEC technology possibly being part of NT. The main point was that people didn't have long SMP history contributing to decision involving sequent in NT SMP scale-up. This isn't exactly the same as straight performance (i.e. NT having equivalent thruput to UNIXes of the period on the same hardware) ... but is likely work on finer-grain locks in the kernel ... so adding processors showed some incremental thruput (and weren't spending all their time waiting/spinning on locks). Some of the (other) performance issues was implied in the comments about weeks away from having to explain to their CEO that a high-end UNIX would be used for some deployment (rather than NT) ... until the responsible executive changed the definition of the deployment to be only that which NT could do.
when charlie was doing work on cp67 fine-grain locking and invented the
compare&swap instruction ... one of the scenarios was showing
compare&swap directly updating variables ... as opposed to (barrier)
locks around the updating of the variables (scenario for various kinds
of applications like database management systems ... for multithreaded
work ... whether running in a multiprocessor environment or not). misc.
past posts mentioning smp &/or compare&swap instruction
https://www.garlic.com/~lynn/subtopic.html#smp
The recent Larrabee thread in comp.arch has touched on wide variety
of topics ... but yesterday there were couple posts discussing
compare&swap variations. misc. past posts in the Larrabee thread:
https://www.garlic.com/~lynn/2009s.html#18 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#20 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#22 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#23 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#29 channels, was Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#30 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#31 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#32 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#33 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#34 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#35 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#36 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#42 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#43 Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#48 Larrabee delayed: anyone know what's happening?
... couple weeks and it has been 42yrs since I first got to play with cp67
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: DEC-10 SOS Editor Intra-Line Editing Newsgroups: alt.folklore.computers,alt.sys.pdp10 Date: Fri, 01 Jan 2010 12:16:58 -0500glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:
a little x-over from thread mentioning character sorting order
https://www.garlic.com/~lynn/2009s.html#63 CAPS Fantasia
before Jim left for tandem ... he would frequently join us
for friday after work ... recent post with old email regarding
friday after work (although this email was after Jim had left
for tandem):
https://www.garlic.com/~lynn/2009r.html#email831110
https://www.garlic.com/~lynn/2009r.html#email831202
in this post
https://www.garlic.com/~lynn/2009r.html#62 some '83 references to boyd
anyway, one friday we were sitting around and bemoaning the fact that very few of the corporate management had any familiarity actually using computers (they designed them, they built them, the sold them, they maintained them, but they didn't actually use computers). discussion turned to possible "silver bullet" computer application that would attract the large number of people that had no experience using computer. we settled on online phone books as an application ... and setup guidelines that Jim would write the lookup routine (that had to be faster than somebody reaching for plantsite "paper" copy), and I would write the ancillary stuff ... *AND* neither of us were allowed to spend more than 40hrs on the effort. Infrastructure had to be simple and relatively lightweight ... so no DBMS could be use.
design was sorted flat file; lastname, firstname. to get performance, lookup would be radix search. frequency distribution of first two letters of last name was calculate for employees ... and initially built into the lookup routine.
files were sorted based on modified "lastname, firstname" value; everything converted to lowercase, special characters and blanks were removed. lookup did something similar for lastname lookup when calculating the first (radix) probe (i.e. "D'A" ... became "da"). lookup would take the number of records in the file ... then the first two letters in the search value was used to calculate the initial record number "probe" (based on name first two letter frequency distribution) ... aka based on simple distribution, "da" would be (4*26/26*26) times the number of records in the file. For 16,000 record file, binary search would avg. around 14 reads ... radix might be as few as 4-5. With 80 byte records and 4096 byte physical disk records (50 file records per disk record) ... radix might be in the correct physical record within first probe.
This was later incorporated into the corporate PROFS deployment and the PROFS people frequently called it the PROFS phone books (somewhat akin to PROFS picking up VMSG for the PROFS email function ... and then trying to the VMSG author fired).
later porting to UNIX ... there wasn't information about number of records ... just the total length of the file. there was simplifying assumption was that the records tended to be approx. same size. I did the initial probe based on letter frequency distribution using the filesize (instead of number of records) ... that gave a byte location for the initial read. Then the code had to find a record separator to find start of record ... for the actual match comparison (aka probes to byte displacement could be anyplace in actual record ... so there was some jitter to find start of an actual record).
at the celebration for Jim .. the tandem people mentioned Jim doing the
online telephone book for tandem ... so I got up and mentioned that he
had previously worked on online telephone book when he was at SJR:
https://www.garlic.com/~lynn/2008i.html#32 A Tribute to Jim Gray: Sometimes Nice Guys Do Finish First
https://www.garlic.com/~lynn/2008i.html#50 Microsoft versus Digital Equipment Corporation
https://www.garlic.com/~lynn/2008i.html#51 Microsoft versus Digital Equipment Corporation
above mentions i was on the 23083 podcast ... at about 1:14 minutes
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Fri, 01 Jan 2010 12:34:59 -0500despen writes:
the issue was that compared to PC applications ... CMS applications was much more real storage hungry and disk i/o intensive. CMS did disk "diagnose i/o" to vm370 ... and the (modified) vm370 did all i/o by sending request to cp88 running on the PC side. cp88 then handled i/o requests between vm370 and keyboard, display, (dos) harddisk, etc.
initial xt/370 had 384k "370" storage (both vm370 kernel and cms paging). I did various page "thrashing" tests ... showing that for almost anything, system was spending much of its time paging (to the 100ms/access dos/xt harddisk). this resulted in revising xt/370 to 512k "370" storage and blaming me for six months slip in the schedule.
even with the reduction in paging ... the shipped product still appeared to be quite slow ... with CMS applications being more disk intensive than compared to similar PC applications; ... and compared to real mainframe (cms files on 3380 drives vis-a-vis on 100ms/access dos/xt drive).
later there was A74 ... which was a separate box ... with a cable that
connected into pc/at. 4mbyte 370 storage ... and faster dos/at disks
made it more palatable ... but also more expensive:
https://www.garlic.com/~lynn/2002d.html#4 IBM Mainframe at home
of course ... there is currently pure software simulation with Hercules.
other old posts mentioning xt/370 &/or A74
https://www.garlic.com/~lynn/94.html#42 bloat
https://www.garlic.com/~lynn/96.html#23 Old IBM's
https://www.garlic.com/~lynn/2000e.html#52 Why not an IBM zSeries workstation?
https://www.garlic.com/~lynn/2001f.html#28 IBM's "VM for the PC" c.1984??
https://www.garlic.com/~lynn/2001i.html#20 Very CISC Instuctions (Was: why the machine word size ...)
https://www.garlic.com/~lynn/2001k.html#24 HP Compaq merger, here we go again.
https://www.garlic.com/~lynn/2002b.html#45 IBM 5100 [Was: First DESKTOP Unix Box?]
https://www.garlic.com/~lynn/2002f.html#49 Blade architectures
https://www.garlic.com/~lynn/2002f.html#50 Blade architectures
https://www.garlic.com/~lynn/2002f.html#52 Mainframes and "mini-computers"
https://www.garlic.com/~lynn/2003f.html#8 Alpha performance, why?
https://www.garlic.com/~lynn/2003f.html#56 ECPS:VM DISPx instructions
https://www.garlic.com/~lynn/2003h.html#40 IBM system 370
https://www.garlic.com/~lynn/2004h.html#29 BLKSIZE question
https://www.garlic.com/~lynn/2004m.html#7 Whatever happened to IBM's VM PC software?
https://www.garlic.com/~lynn/2004m.html#10 Whatever happened to IBM's VM PC software?
https://www.garlic.com/~lynn/2004m.html#11 Whatever happened to IBM's VM PC software?
https://www.garlic.com/~lynn/2004m.html#13 Whatever happened to IBM's VM PC software?
https://www.garlic.com/~lynn/2004o.html#9 Integer types for 128-bit addressing
https://www.garlic.com/~lynn/2005f.html#6 Where should the type information be: in tags and descriptors
https://www.garlic.com/~lynn/2005f.html#10 Where should the type information be: in tags and descriptors
https://www.garlic.com/~lynn/2006.html#10 How to restore VMFPLC dumped files on z/VM V5.1
https://www.garlic.com/~lynn/2006f.html#2 using 3390 mod-9s
https://www.garlic.com/~lynn/2006j.html#36 The Pankian Metaphor
https://www.garlic.com/~lynn/2006m.html#56 DCSS
https://www.garlic.com/~lynn/2006n.html#5 Not Your Dad's Mainframe: Little Iron
https://www.garlic.com/~lynn/2006n.html#14 RCA Spectra 70/25: Another Mystery Computer?
https://www.garlic.com/~lynn/2006y.html#29 "The Elements of Programming Style"
https://www.garlic.com/~lynn/2006y.html#30 "The Elements of Programming Style"
https://www.garlic.com/~lynn/2007.html#1 "The Elements of Programming Style"
https://www.garlic.com/~lynn/2007c.html#14 How many 36-bit Unix ports in the old days?
https://www.garlic.com/~lynn/2007c.html#23 How many 36-bit Unix ports in the old days?
https://www.garlic.com/~lynn/2007d.html#7 Has anyone ever used self-modifying microcode? Would it even be useful?
https://www.garlic.com/~lynn/2007d.html#25 modern paging
https://www.garlic.com/~lynn/2007e.html#5 Is computer history taugh now?
https://www.garlic.com/~lynn/2007j.html#41 z/VM usability
https://www.garlic.com/~lynn/2007l.html#61 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007s.html#43 Intel Ships Power-Efficient Penryn CPUs
https://www.garlic.com/~lynn/2008g.html#22 Was CMS multi-tasking?
https://www.garlic.com/~lynn/2008h.html#73 Microsoft versus Digital Equipment Corporation
https://www.garlic.com/~lynn/2008r.html#33 What if the computers went back to the '70s too?
https://www.garlic.com/~lynn/2008r.html#38 "True" story of the birth of the IBM PC
https://www.garlic.com/~lynn/2008r.html#68 New machine code
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: dead zone Newsgroups: bit.listserv.ibm-main Date: Fri, 01 Jan 2010 11:13:58 -0500Steve Samson <ssamson@dc.rr.com> writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Fri, 01 Jan 2010 18:17:41 -0500hancock4 writes:
I didn't have any knowledge of that. As undergraduate ... and doing
tty/ascii support for cp67 ... tried to get the 2702 to do something, it
couldn't quite do ... helped led to motivation to build clone controller
using Interdata/3 (four of us at the univ, later getting blamed for that
clone controller business) ... mentioned in a couple recent threads:
https://www.garlic.com/~lynn/2009s.html#29 channels, was Larrabee delayed: anyone know what's happening?
https://www.garlic.com/~lynn/2009s.html#53 DEC-10 SOS Editor Intra-Line Editing
Later at SJR (before research moved up the hill to almaden, was in bldg. 28 on the main disk division plant site) ... i got involved doing various things in the disk division & got to play disk engineer.
Among other things ... I would get requested to participate in conference calls with channel engineers in POK. I once asked why I was getting sucked into ... and was told that had been previously been handled by senior engineers ... but most of them had been hired away. The atmosphere around silicon valley was startups offering lots of compensation for experienced people ... not necessarily just larger salaries ... but frequently also promises of large equity.
Disk division silicon valley during the go-go years ... was quite a bit different from many other corporate locations (for instance, there were some of us that would periodically go by Tandem on friday afternoons, to see Jim, or the monthly baybunch user group meetings ... where there would also be people from Amdahl and maybe NAS). People could change employers several times w/o ever having to move.
There was some of that in boston area ... but not nearly as much. There is the scenario where head of POK is considered a major contributer to VMS ... because of convincing corporate to kill off vm370, shutdown burlington mall development group and move everybody to POK (as necessary to making the mvs/xa ship schedule) ... some number of people left and went to work for DEC (on VMS) instead (Endicott eventually managed to save the vm370 product mission but had to reconstitute a group from scratch).
In any case, some of the other discussions about Future Systems claim a
major motivation for FS was the rise of the clone controllers (and
then the distraction of FS gave clone processors their opening)
https://www.garlic.com/~lynn/submain.html#futuresys
in the early 80s ... there was a mini-version ... after some number of
the 801/risc projects were being killed off, people were leaving (AMD,
HP, etc). some old email
https://www.garlic.com/~lynn/2003e.html#email811006
https://www.garlic.com/~lynn/2003e.html#email811006b
https://www.garlic.com/~lynn/2003e.html#email811113
https://www.garlic.com/~lynn/2003e.html#email811115
in this post
https://www.garlic.com/~lynn/2003e.html#65 801 (was Re: Reviving Multics
there were references wondering why I wasn't part of the exodus
... especially since info about this scenario had got some
dissemination
https://www.garlic.com/~lynn/2009r.html#50 "Portable" data centers
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: DEC-10 SOS Editor Intra-Line Editing Newsgroups: alt.folklore.computers,alt.sys.pdp10 Date: Fri, 01 Jan 2010 21:46:00 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
it turned out that online phone book app became quite popular ... but
the real silver bullet ... was when rumor started that senior executive
committee was starting to use email to communicate with each other; then
there was mad rush by much of the rest of the management ranks to at
least have a 3270 terminal on their desk ... recent post
https://www.garlic.com/~lynn/2009q.html#49 The 50th Anniversary of the Legendary IBM 1401
so in some ways ... it was the case of careful what you ask for.
actually ... some percentage of the branch office poeple had been
exposed to online computer use via the (virtual machine vm370-based)
HONE system ... as more and more of customer orders had to be processed
thru various HONE applications ... misc. past post mentioning HONE
https://www.garlic.com/~lynn/subtopic.html#hone
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Larrabee delayed: anyone know what's happening? Newsgroups: comp.arch Date: Sat, 02 Jan 2010 10:10:00 -0500rpw3@rpw3.org (Rob Warnock) writes:
as mentioned, 8100 used a very slow & underpowered processor that was developed for control function; my wife had been asked to do a technical audit of the 8100 ... and very shortly later, 8100 was killed.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: CAPS Fantasia Newsgroups: alt.folklore.computers Date: Sat, 02 Jan 2010 10:23:27 -0500Jean-Marc Bourguet <jm@bourguet.org> writes:
thread originated on ibm-main (mainframe) mailing list, i added a.f.c. x-post when I replied about some of ascii stuff.
ibm-main originated on bitnet in the 80s ... using listserv mailing list
processing. some number of the bitnet mailing list are gateway to usenet
in the "bit.listserv" hierarchy. listserv was something of subset of
function of TOOLSRUN running on the internal network
https://www.garlic.com/~lynn/subnetwork.html#internalnet
misc. past posts mentioning bitnet (&/or EARN)
https://www.garlic.com/~lynn/subnetwork.html#bitnet
listserv historical reference (paris, 1985):
http://www.lsoft.com/products/listserv-history.asp
the bitnet extension in europe was EARN. old email from somebody taking
assignment to setup EARN
https://www.garlic.com/~lynn/2001h.html#email840320
in this post
https://www.garlic.com/~lynn/2001h.html#65 UUCP email
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: The 50th Anniversary of the Legendary IBM 1401 Newsgroups: alt.folklore.computers Date: Sat, 02 Jan 2010 10:33:52 -0500jmfbahciv <jmfbahciv@aol> writes:
and post from today in comp.arch referring to '79 email about attempts
at getting 801 chips for lisp machine:
https://www.garlic.com/~lynn/2010.html#6
symbolics wiki page
https://en.wikipedia.org/wiki/Symbolics
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sat, 02 Jan 2010 11:54:16 -0500William Hamblen <william.hamblen@earthlink.net> writes:
this was motivation for business purchase cards ... basically a credit card ... that had some additional tailored business rules on the backend ... as well as driver for level3 data for transactions.
basic credit card transaction for straight-forward authorization is around 60 bytes; can include merchant code (merchant type) ... simplest business rule is purchase card limited to specific kind of merchant and/or max. value per transaction (in addition to standard credit limit). level3 data would have the full itemized purchase with SKU codes. business rules with level3 data could limit purchases to specific (quantities) of product (in addition to specific merchant type, specific merchant, and/or even specific merchant location).
to help cut the purchase order expense for GM ... as part of cost reduction driving thru business purchase card ... credit (purchase) card statements was converted to EDI and delivered to GM in (large batch) electronic form directly into their standard EDI processing system.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sat, 02 Jan 2010 15:36:59 -0500Pat Farrell <pfarrell@pfarrell.com> writes:
similar story for cp67/cms (& later vm370/cms) ... some number of
commercial online timesharing service bureaus starting with cp67/cms in
the 60s (i.e. cms was the "personal computing" from 60s well into the
80s ... running in its own virtual machine) ... for instance tymshare
offering both mainframe vm370 and pdp10.
https://en.wikipedia.org/wiki/Tymshare
tymshare developed vm370 online computer conferencing and offered it
"free" to the ibm share user group as vmshare (starting aug76) ...
vmshare archives:
http://vm.marist.edu/~vmshare/
midrange systems saw big uptick starting in the late 70s ... both
vax/vms as well a vm/43xx (big difference between volume of vm/43xx and
vax/vms ... was a lot of big corporate w/orders of multi-hundreds at a
time for vm/43xx). old post with vax/vms volumes sliced & diced by
model, yr, us/non-us, etc:
https://www.garlic.com/~lynn/2002f.html#0 Computers in Science Fiction
A major (corporate) spreadsheet application was travel expense form ... there was a CMS implementation done in REXX and (at least) a Lotus version; there was spreedsheet that came with turbo pascal ... and several people got the official corporate (lotus) travel expense spreedsheet running on turbo pascal spreedsheet.
mid-range (both vax/vms & vm/43xx) saw drop-off in mid-80s, as workstations and large PCs moved up the value chain.
some of the (other) high-end mainframe VM use dropped off starting in late 80s with incorporation of major subset of VM function into the "native" hardware as LPARs. Early part of this decade saw some resurgance of high-end mainframe VM use with growing installations of large number of (mainframe) linux virtual machines (some numbers starting in early part of this decade about major portion of new mainframe installations were these vm/linux operations ... as opposed to purely replacement/upgrade of existing mainframe installations).
I did some number of memos in the early 80s ... that one of the reasons
for the growth in the distributed vm/43xx installations was overcoming
the relatively rigidity and lack of agile adaptability of the mainframe
datacenter operations. misc. old vm/43xx email/memo
https://www.garlic.com/~lynn/lhwemail.html#43xx
I also sponsored Boyd's briefings at IBM in this period ... which had a
major agility theme in his OODA-loop paradigm. misc. past posts
mentioning Boyd &/or OODA-loop
https://www.garlic.com/~lynn/subboyd.html
lots of past posts mentioning online commercial (virtual machine based)
timesharing service bureau
https://www.garlic.com/~lynn/submain.html#timeshare
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sat, 02 Jan 2010 16:11:09 -0500Walter Bushell <proto@panix.com> writes:
one of the applications reimplemented on trs80 was the cms script
document formating. cms script started off with "dot" formating commands
(somewhat from ctss runoff) ... and after GML was invented at science
center in '69 ... support for GML tags were added (which evolved into
international standard as SGML ... later also morphed into HTML and XML)
... some old posts
https://www.garlic.com/~lynn/2004l.html#74 Specifying all biz rules in relational data
https://www.garlic.com/~lynn/2004m.html#20 Whatever happened to IBM's VM PC software?
https://www.garlic.com/~lynn/2008s.html#0 First successful PC OS
also some old email mentioning stopping by to see him:
https://www.garlic.com/~lynn/2007.html#email801016
in this post
https://www.garlic.com/~lynn/2007.html#1 "The Elements of Programming Style"
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Small-Job Efficiency Newsgroups: alt.folklore.computers Date: Sat, 02 Jan 2010 16:38:03 -0500Lars Poulsen <lars@beagle-ears.com> writes:
somewhere along the way DTU got some vm370 ... because I remember user group meeting and some number of other visits to Lyngby in the 70s.
slightly related
https://www.garlic.com/~lynn/96.html#4a John Hartmann's Birthday Party
and ...
http://vm.marist.edu/~piper/
and
http://vm.marist.edu/~piper/party/jph-12.html#wheeler
your page only has passing reference
http://www.beagle-ears.com/lars/engineer/comphist/dan_hist.htm
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sat, 02 Jan 2010 20:19:15 -0500Patrick Scheible <kkt@zipcon.net> writes:
a lot of 3270s were actually in use for 6-10 yrs
recent reference to jim's "mip envy"
https://www.garlic.com/~lynn/2009q.html#49 The 50th Anniversary of the Legendary IBM 1401
20sep80 version of jim's "mip envy"
https://www.garlic.com/~lynn/2007d.html#email800920
and a 24sep80 version
https://web.archive.org/web/20081115000000*/http://research.microsoft.com/~gray//papers/CritiqueOfIBM%27sCSResearch.doc
3270 capital planning was also part of annual budget planning ... which
was problem when management/executives decided that they needed (annual)
3270 terminals rerouted to their desk ... recent reference:
https://www.garlic.com/~lynn/2010.html#5 DEC-10 SOS Editor Intra-Line Editing
other posts in this thread:
https://www.garlic.com/~lynn/2009r.html#52 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#57 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#68 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#69 360 programs on a z/10
https://www.garlic.com/~lynn/2009s.html#60 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#4 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#9 360 programs on a z/10
misc. other posts mentioning mip envy:
https://www.garlic.com/~lynn/2001g.html#7 New IBM history book out
https://www.garlic.com/~lynn/2002k.html#39 Vnet : Unbelievable
https://www.garlic.com/~lynn/2002o.html#73 They Got Mail: Not-So-Fond Farewells
https://www.garlic.com/~lynn/2002o.html#74 They Got Mail: Not-So-Fond Farewells
https://www.garlic.com/~lynn/2002o.html#75 They Got Mail: Not-So-Fond Farewells
https://www.garlic.com/~lynn/2004c.html#15 If there had been no MS-DOS
https://www.garlic.com/~lynn/2004l.html#28 Shipwrecks
https://www.garlic.com/~lynn/2004l.html#31 Shipwrecks
https://www.garlic.com/~lynn/2005c.html#50 [Lit.] Buffer overruns
https://www.garlic.com/~lynn/2005u.html#41 Mainframe Applications and Records Keeping?
https://www.garlic.com/~lynn/2006n.html#26 sorting was: The System/360 Model 20 Wasn't As Bad As All That
https://www.garlic.com/~lynn/2006o.html#50 When Does Folklore Begin???
https://www.garlic.com/~lynn/2007.html#1 "The Elements of Programming Style"
https://www.garlic.com/~lynn/2007.html#13 "The Elements of Programming Style"
https://www.garlic.com/~lynn/2007d.html#17 Jim Gray Is Missing
https://www.garlic.com/~lynn/2007d.html#45 Is computer history taugh now?
https://www.garlic.com/~lynn/2007d.html#63 Cycles per ASM instruction
https://www.garlic.com/~lynn/2007e.html#50 Is computer history taught now?
https://www.garlic.com/~lynn/2007f.html#13 Why is switch to DSL so traumatic?
https://www.garlic.com/~lynn/2007f.html#70 Is computer history taught now?
https://www.garlic.com/~lynn/2009l.html#41 another item related to ASCII vs. EBCDIC
https://www.garlic.com/~lynn/2009p.html#8 WSJ.com - IBM Puts Executive on Leave
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Oldest Instruction Set still in daily use? Newsgroups: alt.folklore.computers Date: Sun, 03 Jan 2010 10:09:03 -0500"invalid" <invalid@invalid.invalid> writes:
360 was announced 1964 (models started shipping 1965) ... a22-6821-0 360
principles of operation (with 24bit addresses)
http://www.bitsavers.org/pdf/ibm/360/princOps/A22-6821-0_360PrincOps.pdf
this is not only all the 360 application instructions continue to be supported today ... but things like compatible program libraries and kernel calls continue to work on modern systems.
more recent principles of operation
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/dz9zr003/CCONTENTS
with 31-bit and 64-bit addressing (in addition to 24-bit addressing). not only do the 24-bit address instructions continue to work ... the programs written using 24-bit address instructions ... including use of library, system services, kernel calls, i/o requests (aka 360 assembler, machine language programs continue to run ... with or w/o re-assemly).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sun, 03 Jan 2010 11:09:36 -0500Morten Reistad <first@last.name> writes:
misc. past post mentioning datahub:
https://www.garlic.com/~lynn/96.html#4a John Hartmann's Birthday Party
https://www.garlic.com/~lynn/2000g.html#40 No more innovation? Get serious
https://www.garlic.com/~lynn/2002f.html#19 When will IBM buy Sun?
https://www.garlic.com/~lynn/2002g.html#79 Coulda, Woulda, Shoudda moments?
https://www.garlic.com/~lynn/2002o.html#33 Over-the-shoulder effect
https://www.garlic.com/~lynn/2003e.html#26 MP cost effectiveness
https://www.garlic.com/~lynn/2003f.html#13 Alpha performance, why?
https://www.garlic.com/~lynn/2004f.html#16 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2005p.html#23 What ever happened to Tandem and NonStop OS ?
https://www.garlic.com/~lynn/2005q.html#9 What ever happened to Tandem and NonStop OS ?
https://www.garlic.com/~lynn/2005q.html#36 Intel strikes back with a parallel x86 design
https://www.garlic.com/~lynn/2006l.html#39 Token-ring vs Ethernet - 10 years later
https://www.garlic.com/~lynn/2006y.html#31 "The Elements of Programming Style"
https://www.garlic.com/~lynn/2007f.html#17 Is computer history taught now?
https://www.garlic.com/~lynn/2007j.html#49 How difficult would it be for a SYSPROG ?
https://www.garlic.com/~lynn/2007n.html#21 The Development of the Vital IBM PC in Spite of the Corporate Culture of IBM
https://www.garlic.com/~lynn/2007n.html#86 The Unexpected Fact about the First Computer Programmer
https://www.garlic.com/~lynn/2007p.html#35 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007v.html#53 folklore indeed
https://www.garlic.com/~lynn/2008e.html#8 MAINFRAME Training with IBM Certification and JOB GUARANTEE
https://www.garlic.com/~lynn/2008p.html#36 Making tea
https://www.garlic.com/~lynn/2008r.html#68 New machine code
https://www.garlic.com/~lynn/2009e.html#58 When did "client server" become part of the language?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sun, 03 Jan 2010 12:19:16 -0500Walter Bushell <proto@panix.com> writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sun, 03 Jan 2010 12:58:23 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
during future system years ... I would sometimes draw parallels between FS effort and a cult film that had been playing down in central sq. for several years.
later when I sponsored Boyd's briefings at IBM ... Boyd would tell this
story about reviewing the specs on the airforce air-to-air missile that
would be used in vietnam ... which seemed to carry some fantasy
aspect ... similar to some of others comments about FS
http://www.jfsowa.com/computer/memo125.htm
partial retelling of the (vietnam airforce air-to-air missile) story:
https://www.garlic.com/~lynn/2008c.html#52 Current Officers
which also had more than a little "fantasy" aspect to it.
other recent reference to Boyd:
https://www.garlic.com/~lynn/2009r.html#50 "Portable" data centers
past posts mentioning Boyd
https://www.garlic.com/~lynn/subboyd.html#boyd1
misc. URLs from around the web mentioning Boyd (&/or his OODA-loops):
https://www.garlic.com/~lynn/subboyd.html#boyd2
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: DEC-10 SOS Editor Intra-Line Editing Newsgroups: alt.folklore.computers,alt.sys.pdp10 Date: Sun, 03 Jan 2010 15:03:44 -0500re:
from long ago (nearly 30yrs) and far away:
Date: 04/07/80 21:27:09
From: wheeler
To: gray
for whatever reason, pls is generating code in the teleprob text deck
which resets the entry point to teleprob that is why i have to
load telex (reset telex
to overide the pls text deck entry point change and then do
genmod telex
the r1 you are seeing is the cms r1 pointing to the cms argument
string in low core from the command. since there were no arguments
passed 8c0 shows x'ffffffff'.
... snip ... top of post, old email index.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sun, 03 Jan 2010 15:24:47 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
from long ago (nearly 30 yrs) and far away:
Date: 03/11/80 20:36:01
From: gray
To: wheeler
Lynn: Bank of America needs your help. They are going to get 60 4341 and
spread them around the world. All will run CP and guest operating sytems on top
They are getting xxxxxx to help them sith a multi-domain architecture.
They need help in the areas of:
Maintenance of a large net of machines from a single central site
Multidomain support.
Perhaps this will force you to write the code.
... snip ... top of post, old email index.
Jim also was sucking me into various (relational, system/r) customers
... including Bank of America, when he left for Tandem. old email
from later in 1980
https://www.garlic.com/~lynn/2007.html#email801016
in this post
https://www.garlic.com/~lynn/2007.html#1 "The Elements of Programming Style"
the above email/post also mentions LA SE that did ATM cash machine
transactions in vm370 virtual machines for large (not BofA)
cal. financial institutions.
https://www.garlic.com/~lynn/2010.html#11 Happy DEC-10 Day
misc. poast posts mentioning original relational/sql
https://www.garlic.com/~lynn/submain.html#systemr
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sun, 03 Jan 2010 15:54:40 -0500Walter Bushell <proto@panix.com> writes:
early 70s, it was relatively new 3-4 story bldg. in residential section of sindelfingen (outside stuttgart, near boeblingen plant site), seemed to be "business" person hotel ... almost no english spoken (look out window in the morning and see the children on the way to school). in the 80s there were a number of american businesses that had been established in the boeblingen/sindelfingen area ... and even "american" brand hotels ... but not in the early 70s.
such exorbitant surcharges was common back then ... could easily be larger than daily room rate. I vaguely remember something ... possibly in the late 80s ... when there was big push back against such huge hotel long distance surcharges.
beoblingen was paying my way over to talk to them about vm370 systems. while I was there I got a call asking if I could stop by paris on my way back ... and look at the (vm370-based) HONE system going in (it was part of transfer of EMEA hdqtrs from westchester to europe).
EMEA hdqtrs HONE was going into brand new three office bldg. complex in La Defense (two of the bldgs. hadn't been finished and landscaping hadn't even been started). Since there was no other facilities in the area yet ... they put me up in Paris hotel within a block or two of RER station (& the opera, and american express office) and I would take train over to the La Defense & walk.
misc. past posts mentioning (virtual machine based) HONE system
https://www.garlic.com/~lynn/subtopic.html#hone
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sun, 03 Jan 2010 20:27:49 -0500Charles Richmond <frizzle@tx.rr.com> writes:
cics "transaction" processing was very lightweight multithreading in single address space ... small assembler, cobol, pli applications doing various kinds of lightweight transaction processing.
bdam (disk) access method had direct disk record numbers as part of the data (application being able to go directly to desired disk record).
that made for lots of efficiency in small constrained environment ... but represented lots of problems as things scaled up.
when i was at univ. in the 60, the univ. library got ONR grant for
digital/online catalog. part of the money was used to get a 2321
"data-cell" ... reference:
http://members.optushome.com.au/intaretro/2321DCD.htm
http://www.columbia.edu/cu/computinghistory/datacell.html
project was also selected to be one of the beta-test sites for the
original CICS (it had originally been developed at customer site ... and
was then being turned into product). I got tasked to support/debug the
deployment at the library. misc. past posts mentioning cics &/or bdam
https://www.garlic.com/~lynn/submain.html#cics
in part because the extremely lightweight environment ... there was little protection/separation between transactions ... application program error could take down whole cics operation.
NIH did something similar to cics (roll-your-own) for national library
of medicine (heavily taking advantage of BDAM optimization).
https://www.garlic.com/~lynn/2009.html#44 Lawyers & programming (x-over from a.f.c discussion)
https://www.garlic.com/~lynn/2009m.html#88 Continous Systems Modelling Package
https://www.garlic.com/~lynn/2009n.html#27 Continous Systems Modelling Package
https://www.garlic.com/~lynn/2009o.html#38 U.S. house decommissions its last mainframe, saves $730,000
https://www.garlic.com/~lynn/2009q.html#25 Old datasearches
doing its own multithreading ... didn't mean that cics had multiprocessor support ... because of various data structures and other things ... it was limited to single processor operation (until just a few years ago). up until then (and possibly even still) a large cics operation might have over hundred different CICS "regions" (in different address spaces); separate CICS instances, all running on same mvs system.
ACP (airline control program) was whole (360) system ... along the lines
of cics ... but w/o the rest of the os/360 overhead and infrastructure.
As some large financial networks started using ACP for financial
transactions ... the product was renamed TPF (transaction processing
facility). It was highly streamlined ... and TPF didn't get
multiprocessor support until late. This caused quite a bit of problem
for 3081 ... which was originally designed to never have a
non-multiprocessor option. Eventually a 3083 (single processor) was done
... in large part motivated by TPF ... as well as other special 3083
features tailored for TPF environment. misc. recent posts mentioning
3083:
https://www.garlic.com/~lynn/2009g.html#66 Mainframe articles
https://www.garlic.com/~lynn/2009g.html#68 IT Infrastructure Slideshow: The IBM Mainframe: 50 Years of Big Iron Innovation
https://www.garlic.com/~lynn/2009g.html#70 Mainframe articles
https://www.garlic.com/~lynn/2009h.html#77 Operating Systems for Virtual Machines
https://www.garlic.com/~lynn/2009l.html#55 IBM halves mainframe Linux engine prices
https://www.garlic.com/~lynn/2009l.html#65 ACP, One of the Oldest Open Source Apps
https://www.garlic.com/~lynn/2009l.html#67 ACP, One of the Oldest Open Source Apps
https://www.garlic.com/~lynn/2009m.html#39 ACP, One of the Oldest Open Source Apps
https://www.garlic.com/~lynn/2009r.html#70 While watching Biography about Bill Gates on CNBC last Night
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Why is JCL so bad was Re: Basic question on passing JCL set symbol to proc Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sun, 03 Jan 2010 21:13:25 -0500PaulGBoulder@AIM.COM (Paul Gilmartin) writes:
more information ... cp67/cms version 3 (oct70) users guide:
http://www.bitsavers.org/pdf/ibm/360/cp67/GH20-0859-0_CP67_Version_3_Users_Guide_Oct70.pdf
os macros .... pg 294
Later after the moph to vm370/cms (and cambridge monitor system renamed conversational monitor system) ... there was some amount of DOS simulation also added. There was also some joke about CMS 32kbyte os/360 simulation code ... did almost as good as the 8mbyte mvs os/360 simulation code.
The biggest change from cp67/cms to vm370/cms ... was reorging the kernel with application program loading at x'12000' to take advantage of (original) 370 (64k) r/o shared segment support. Application programs then were moved to starting at x'20000' ... and cms kernel "shared code" moved to start at segment boundary at x'10000' (segment 0 was then data and non-shared code ... since virtual page zero couldn't be shared, and therefor virtual segment 0 couldn't be shared).
however, all the effort came to *NOT* ... before the full 370 virtual memory architecture could be announced and shipped ... retrofitting the full 370 virtual memory architecture to 370/165 ran into all sorts of problems. 165 petitioned for dropping several features to gain back a lot of the schedule; and one of the things dropped was 370 r/o shared segment support. that forced m370 to quickly come up with a quick&dirty hack using storage protection keys ... to simulate the original 370 virtual memory r/o segment protect.
with the addition of dos simulation ... for running dos programs in cms
... there then had to be option to specify whether non-cms service
requests went thru os simulation or dos simulation ... aka "set dos on"
http://publib.boulder.ibm.com/infocenter/zvm/v5r4/topic/com.ibm.zvm.v54.dmsa5/hcsd2b00514.htm
in the wake of demise of FS and the head of POK convincing corporate to kill vm370, shutdown the burlington development group and moving all the people to POK (justification was that otherwise wouldn't be able to make mvs/xa ship schedule) .... there was joke that head of POK was major contributor to vax/vms (i.e. the people that didn't move to pok ... but went to DEC to work on vax/vms); endicott eventually was able to obtain the vm370 product mission ... but had to reconstitute the organization from scratch.
the issue here was that one of the people in burlington (that didn't move to pok) had done a pretty complete os disk i/o routines for os disk r/w, vtocs, handle most kinds of files, understood and could process PDS, etc (aka wasn't os/360 file simulation in cms filesystem, but supported full os/360 operation on os/360 disks). All of that evaporated/disappeared with the shutdown of burlington development group.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sun, 03 Jan 2010 21:25:20 -0500Pat Farrell <pfarrell@pfarrell.com> writes:
renamed from ACP to TPF when financial networks started using it
wiki tpf page:
https://en.wikipedia.org/wiki/Transaction_Processing_Facility
above includes mention of sabre, amadeus, visa, amex, eds, holiday inn, etc
a few past posts mentioning my wife did short stint as amadeus chief
architect:
https://www.garlic.com/~lynn/2008c.html#53 Migration from Mainframe to othre platforms - the othe bell?
https://www.garlic.com/~lynn/2008i.html#19 American Airlines
https://www.garlic.com/~lynn/2008i.html#34 American Airlines
https://www.garlic.com/~lynn/2008p.html#41 Automation is still not accepted to streamline the business processes... why organizations are not accepting newer technologies?
https://www.garlic.com/~lynn/2009j.html#33 IBM touts encryption innovation
https://www.garlic.com/~lynn/2009l.html#55 IBM halves mainframe Linux engine prices
https://www.garlic.com/~lynn/2009r.html#59 "Portable" data centers
ibm tpf page:
http://www-01.ibm.com/software/htp/tpf/index.html
ibm tpf article
http://www.ibmsystemsmag.com/mainframe/enewsletterexclusive/10019p1.aspx
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sun, 03 Jan 2010 21:44:15 -0500hancock4 writes:
i was told that it had been the base cost of doing purchase order processing ... all the other stuff would have been on top of the base processing.
it was major motivation for commercial purchase cards
some discussions of purchase cards and level 2/3 data
http://www.gotmerchant.com/level3_credit_card_processing.php
Purchase cards, GSA SmartPay, and Level 2 & 3 credit
card processing ("feds save $1B in 2006 using p-cards")
http://www.braintreepaymentsolutions.com/blog/purchase-cards-gsa-smartpay-and-level-2-3-credit-card-processing/
as mentioned ... one of the "special" things done for GM was converting
(commercial/payment) credit card statements into (GMs) EDI 8x0 records
... vendor web pages talking about gm edi solutions:
http://www.dicentral.com/general_motor_edi/
http://www.123edi.com/partners/edi-general-motors.asp
http://www.edipipeline.com/General_Motors.asp
http://www.infoconn.com/EDI/Partners/General_Motors.htm
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sun, 03 Jan 2010 22:03:39 -0500glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:
datasets are contiguous with start/base and limit (with limited number of "Extents") ... so (even) relative ttr was (mostly) simple add to start/base (didn't require complex lookup). TTR would have issue if move to different device type with different number of records per track (unless simulation/translation layer is thrown in). sometimes copy to different device (of same type) ... to eliminate EXTENTS ... i.e. single larger area.
one of the issues brought up for move to FBA ... where just have
sequential record numbers and knowledge about things like records/track,
tracks/cylinder, etc ... isn't exposed. misc. past posts
discussing ckd, multi-track search, fba, etc
https://www.garlic.com/~lynn/submain.html#dasd
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Sun, 03 Jan 2010 22:22:18 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
lots of business pushed to have things all done thru EDI 8x0 records ... significantly reducing cost of processing.
however, there was still a lot of transactions with suppliers that weren't setup to do things via EDI ... like local retail outlets.
the commercial/purchase credit cards ... could be setup with real-time authorization business rules (based on level 2&3 dat) ... and have the credit card issuer do the conversion of credit card statements into EDI 8x0 records.
the credit card processors got new transactions (and fees) ... and the businesses got real-time authorization business rules ... and conversion of lots of stuff to EDI ... that had been resistant/difficult otherwise.
merchant POS terminals had to be upgraded to supply level 2 (& possibly level 3) data ... but it wasn't as if it was data that they weren't already processing (for other reasons).
EDI wiki page
https://en.wikipedia.org/wiki/Electronic_Data_Interchange
X12 EDIFACT Mapping
https://en.wikipedia.org/wiki/X12_EDIFACT_Mapping
EDI standards body
http://www.x12.org/x12org/index.cfm
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Oldest Instruction Set still in daily use? Newsgroups: alt.folklore.computers Date: Mon, 04 Jan 2010 05:48:28 -0500Charles Richmond <frizzle@tx.rr.com> writes:
funsoft had made major product platform choice of sequent ... and after ibm bought sequent ... that seemed to better cement such relationship ... but then ibm discontinued all the (numa-q) sequent stuff ... and more recently there has been a lot of contention with the 360 software emulators running on intel platforms (including various legal action with one of the other players).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Mon, 04 Jan 2010 06:02:41 -0500Charles Richmond <frizzle@tx.rr.com> writes:
i had uncle that spent couple yrs during ww2 in europe as tank mechanic ... he was big ... which is possibly reason that ruled him out as being co'opted for tank crew (stories about everybody being co'opted for tank crew, including cooks ... they lost lots of crews and tanks ... but apparently the new tanks kept arriving faster than they could find replacement crews).
misc. past posts mentioning boyd (&/or OODA-loops)
https://www.garlic.com/~lynn/subboyd.html
misc. past posts mentioning boyd's sherman story:
https://www.garlic.com/~lynn/2000c.html#85 V-Man's Patton Quote (LONG) (Pronafity)
https://www.garlic.com/~lynn/2001.html#30 Review of Steve McConnell's AFTER THE GOLD RUSH
https://www.garlic.com/~lynn/2001m.html#3 mainframe question
https://www.garlic.com/~lynn/2001m.html#10 mainframe question
https://www.garlic.com/~lynn/2001m.html#11 mainframe question
https://www.garlic.com/~lynn/2001m.html#16 mainframe question
https://www.garlic.com/~lynn/2003n.html#27 Controversial paper - Good response article on ZDNet
https://www.garlic.com/~lynn/2004b.html#24 The SOB that helped IT jobs move to India is dead!
https://www.garlic.com/~lynn/2004g.html#53 Chained I/O's
https://www.garlic.com/~lynn/2004l.html#19 FW: Looking for Disk Calc program/Exec (long)
https://www.garlic.com/~lynn/2005j.html#11 The 8008
https://www.garlic.com/~lynn/2005j.html#14 The 8008
https://www.garlic.com/~lynn/2006f.html#14 The Pankian Metaphor
https://www.garlic.com/~lynn/2006q.html#28 was change headers: The Fate of VM - was: Re: Baby MVS???
https://www.garlic.com/~lynn/2008g.html#21 WWII supplies
https://www.garlic.com/~lynn/2008j.html#11 To: Graymouse -- Ireland and the EU, What in the H... is all this about?
https://www.garlic.com/~lynn/2008j.html#21 To: Graymouse -- Ireland and the EU, What in the H... is all this about?
https://www.garlic.com/~lynn/2008q.html#17 realtors (and GM, too!)
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Mon, 04 Jan 2010 06:20:44 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
there are various stories that for "desert storm" (previous middle east
conflict) boyd's battle plan won out against up-the-middle tank battle &
slug it out until last man standing (as well as some quotes about
problem with current conflict, was that boyd had died in '97) ...
http://findarticles.com/p/articles/mi_m1571/is_32_18/ai_91210683/
http://www.freerepublic.com/focus/news/732364/posts
http://www.statemaster.com/encyclopedia/John-Boyd-%28military-strategist%29
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Mon, 04 Jan 2010 06:27:33 -0500re:
Boyd's tactics and Operation Iraqi Freedom (illuminating background on
Iraq strategy)
http://www.freerepublic.com/focus/news/899525/posts
from above:
Lamb: What part of the Gulf War in 1991 plan did John Boyd have some responsibility for?
Coram: All of it. The multiple thrust, the feints, the ambiguity, the Marine feint, the ...
Lamb: You mean the landing in Kuwait, the early landing?
Coram: Yes. Yes.
Lamb: That was his idea?
Coram: It was his idea. He was behind every bit of it.
... snip ...
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Larrabee delayed: anyone know what's happening? Newsgroups: comp.arch,alt.folklore.computers Date: Mon, 04 Jan 2010 07:17:01 -0500medusa was all about footprint, cooling (handling heat as more & more was compressed into smaller area), and interconnect ... and we were doing cluster scale-up as part of ha/cmp ... which also was heavily into availability
in fact, when I was out marketing ha/cmp, I had coined the terms
disaster survivability and geographic survivability to differentiate
from disaster/recovery
https://www.garlic.com/~lynn/submain.html#available
primary person behind medusa was engineer that I previously mentioned
involved with 6000 serial-link-adapter (SLA) and then worked in FCS
committee ... recent reference:
https://www.garlic.com/~lynn/2009s.html#32 Larrabee delayed: anyone know what's happening?
misc. old medusa email &/or cluster scale-up
https://www.garlic.com/~lynn/lhwemail.html#medusa
this is old email (lots of stuff about working with LLNL &/or parties
doing stuff for LLNL.) ... possibly just hrs before being told it was
being transferred and we weren't to work on anything with more than
4-processors
https://www.garlic.com/~lynn/2006x.html#email920129
and then little over 2weeks later announcement (only for scientific
and technical)
https://www.garlic.com/~lynn/2001n.html#6000clusters1
and then later that summer (about the time we were leaving)
... clusters caught us by surprise
https://www.garlic.com/~lynn/2001n.html#6000clusters2
other recent threads mentioning above:
https://www.garlic.com/~lynn/2009o.html#81 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#85 Anyone going to Supercomputers '09 in Portland?
https://www.garlic.com/~lynn/2009q.html#27 Supercomputers Are Still Fast, but Less Super
https://www.garlic.com/~lynn/2009q.html#33 Check out Computer glitch to cause flight delays across U.S.
as previously mentioned ... some connection between cluster scale-up
and electronic commerce ... reference to jan92 meeting about
cluster scale-up for commercial dataprocessing
https://www.garlic.com/~lynn/95.html#13
two of the people in that meeting later leave and show up at small client/server startup responsible for something called "commerce server" and we were brought in because they wanted to do payment transactions on the server. The "commerce server" started out as collection of servers providing a multi-store "mall" paradigm implemented with large oracle dbms backend. the startup had also invented this technology called "SSL" they wanted to use.
Now before we had left ... besides ha/cmp, FCS, HIPPI and misc. other stuff ... we also played some in SCI. Part of ha/cmp issue involved purity of "801" and simplified hardware ... and major 801 simplification theme was no cache coherency (ruling out shared memory multiprocessing). Since there was no cache coherency & multiprocessing ... that forced scale-up to purely interconnect solution.
While doing electronic commerce ... we also talked some to convex about their exemplar sci implementation (using HP risc chips). Then HP acquired both Verifone (a major point-of-sale terminal vendor that was looking at moving into the electronic commerce value chain) and Convex ... and we spent some amount of time at HP, talking to the respective responsible executives.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Why is JCL so bad was Re: Basic question on passing JCL set symbol to proc Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Mon, 04 Jan 2010 11:04:11 -0500hancock4 writes:
the thread originated in ibm-main ... and I added in a.f.c.. earlier in
the (ibm-main) part of the thread there; can be viewed in google groups:
http://groups.google.com/group/bit.listserv.ibm-main/browse_thread/thread/8fb4c75d62fc6353/b31fc2570a47f078?q=why+is+jcl+so+bad+group:bit.listserv.ibm-main#b31fc2570a47f078
.... one of the earlier posts (by somebody else) starts out with:
I transcribed this from a talk given by Fred Brooks, Jr. on the 40th year anniversary of the System 360. This is a wonderful talk given by the people who were involved in the original design, recorded at the Computer History Museum. This is what Fred Brooks said about JCL:
... snip ...
and then goes on for a couple paragraphs
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Oldest Instruction Set still in daily use? Newsgroups: alt.folklore.computers Date: Mon, 04 Jan 2010 11:19:18 -0500jmfbahciv <jmfbahciv@aol> writes:
instruction nearly all hardware these days (change from the days when most low/mid range 370s were vertical microcode, and high-end was horizontal microcode) ... except for a few of the more complex instructions done in "millicode" (nearly all of the original 360 instructions fall into the "less complex" category).
Millicode in an IBM zSeries processor
http://findarticles.com/p/articles/mi_qa3751/is_200405/ai_n9388162/
from above:
While the hardware can execute many of the logically less complex and high-performance instructions, millicode is required to implement the more complex instructions, as well as to provide additional support functions related primarily to the central processor. This paper is a review of millicode on previous zSeries CMOS systems and also describes enhancements made to the z990 system for processing of the millicode. It specifically discusses the flexibility millicode provides to the z990 system.
... snip ...
there are slightly different issues of what is the 1) oldest machine
still running, 2) oldest software still running on new/existing
hardware, and 3) oldest software still running on emulated
hardware. reference to old software running on new "intel" hardware
under 360 simulator
https://www.garlic.com/~lynn/2010.html#27 Oldest Instruction Set still in daily use?
and posts in thread about 360 programs running on current hardware:
https://www.garlic.com/~lynn/2009r.html#52 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#57 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#68 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#69 360 programs on a z/10
https://www.garlic.com/~lynn/2009s.html#60 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#4 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#9 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#13 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#16 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#17 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#20 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#24 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#26 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#28 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#29 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#30 360 programs on a z/10
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Mon, 04 Jan 2010 11:47:52 -0500Howard Brazee <howard.brazee@cusys.edu> writes:
there was story about marines being forced into some number of
Abrams. the issue was that Abrams contract called for per tank discount
above some minimum number ... which the army didn't quite justify ... in
order to get the total up to the minimum, the marines had to take
a few. ... this mentions 4796 for the army and 221 for the marines
(would 5000 have been contract minimum for discount?)
http://www.army-technology.com/projects/abrams/
the problem for the marines is that they traditionally have a very different mission profile than the army ... well over 90% of their mission profiles are in areas with 30-35ton load limit ... and the Abrams runs 65-70 tons. they had to spend big part of their budget on something they couldn't use (other than as supporting some army operation), as well as the people needed to populate the tank corp, as well as not buying what they really needed (trivial example, I remember a relative commenting that their army unit had larger computer allocation than the whole marine corp).
along the lines of earmarks and directed appropriations ... not so bad if there is actual additional money ... but can get really bad when department isn't allocated any additional money ... but is directed on how they spend their existing money. in the mid-90s we visited department that had built large, new, expensive marble bldg ... which was mostly empty. People that were there complained that they didn't need the bldg ... they needed their regular budget to do things that the department was suppose to do. Instead, congress had directed that major portion of departments budget go for the new building (subsidy by some congressman to builder in his state that got the contract).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: dead zone Newsgroups: bit.listserv.ibm-main Date: 4 Jan 2010 09:07:19 -0800Scott.Rowe@JOANN.COM (Scott Rowe) writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 12:35:18 -0500Mark Crispin <mrcrispin@panda.com> writes:
basically other programs are so starting to swamp the total budget and defense spending as percentage of total budget has been on decline.
saving our future requires tough choices today:
http://www.gao.gov/cghome/d08241cg.pdf
extract from above (also in this post):
https://www.garlic.com/~lynn/2007t.html#14
first page in the above gives percent of federal spending for 1966, 1986 and 2006
1966 1986 2006 defense 43 28 20 net interest 7 14 9 all other spending 34 29 32 social security 15 20 21 medicare 1 10 19... snip ...
lots of other posts referencing comptroller general rants:
https://www.garlic.com/~lynn/2006f.html#41 The Pankian Metaphor
https://www.garlic.com/~lynn/2006f.html#44 The Pankian Metaphor
https://www.garlic.com/~lynn/2006g.html#9 The Pankian Metaphor
https://www.garlic.com/~lynn/2006g.html#14 The Pankian Metaphor
https://www.garlic.com/~lynn/2006g.html#27 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#2 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#3 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#4 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#17 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#19 The Pankian Metaphor
https://www.garlic.com/~lynn/2006h.html#33 The Pankian Metaphor
https://www.garlic.com/~lynn/2006o.html#61 Health Care
https://www.garlic.com/~lynn/2006p.html#17 Health Care
https://www.garlic.com/~lynn/2006r.html#0 Cray-1 Anniversary Event - September 21st
https://www.garlic.com/~lynn/2006t.html#26 Universal constants
https://www.garlic.com/~lynn/2007j.html#20 IBM Unionization
https://www.garlic.com/~lynn/2007j.html#91 IBM Unionization
https://www.garlic.com/~lynn/2007k.html#19 Another "migration" from the mainframe
https://www.garlic.com/~lynn/2007o.html#74 Horrid thought about Politics, President Bush, and Democrats
https://www.garlic.com/~lynn/2007p.html#22 U.S. Cedes Top Spot in Global IT Competitiveness
https://www.garlic.com/~lynn/2007q.html#7 what does xp do when system is copying
https://www.garlic.com/~lynn/2007s.html#1 Translation of IBM Basic Assembler to C?
https://www.garlic.com/~lynn/2007t.html#13 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007t.html#14 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007t.html#15 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007t.html#24 Translation of IBM Basic Assembler to C?
https://www.garlic.com/~lynn/2007t.html#25 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007t.html#33 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007t.html#35 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2007v.html#26 2007 Year in Review on Mainframes - Interesting
https://www.garlic.com/~lynn/2008.html#57 Computer Science Education: Where Are the Software Engineers of Tomorrow?
https://www.garlic.com/~lynn/2008d.html#40 Computer Science Education: Where Are the Software Engineers of Tomorrow?
https://www.garlic.com/~lynn/2008e.html#50 fraying infrastructure
https://www.garlic.com/~lynn/2008f.html#86 Banks failing to manage IT risk - study
https://www.garlic.com/~lynn/2008g.html#1 The Workplace War for Age and Talent
https://www.garlic.com/~lynn/2008h.html#3 America's Prophet of Fiscal Doom
https://www.garlic.com/~lynn/2008h.html#26 The Return of Ada
https://www.garlic.com/~lynn/2008i.html#98 dollar coins
https://www.garlic.com/~lynn/2008n.html#8 Taxcuts
https://www.garlic.com/~lynn/2008n.html#9 Taxcuts
https://www.garlic.com/~lynn/2008n.html#17 Michigan industry
https://www.garlic.com/~lynn/2009f.html#20 What is the real basis for business mess we are facing today?
https://www.garlic.com/~lynn/2009n.html#55 Hexadecimal Kid - articles from Computerworld wanted
https://www.garlic.com/~lynn/2009p.html#86 Opinions on the 'Unix Haters' Handbook
https://www.garlic.com/~lynn/2009p.html#87 IBM driving mainframe systems programmers into the ground
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 13:45:46 -0500"Charlie Gibbs" <cgibbs@kltpzyxm.invalid> writes:
the comptroller general's rants are that the unfunded mandates done over the last 20-30 yrs ... totally swamp everything else ... by tens of trillions.
service on the debt is increasing ... but that also is in the noise range compared to the unfunded mandates that the former comptroller general has been ranting about.
big part of recent debt are things like TARP and stiumuls funds started under the previous administration. TARP was originally targeted for purchase of toxic/troubled asset purchases from regulated financial institution.
the basic process was that unregulated loan originators were able to buy triple-A ratings for the packaged (CDO) loans from the rating agencies. This 1) provided unregulated loan originators almost unlimited source of funds (which hadn't existed previously) and 2) eliminated any motivation for them to do any loan qualification (they could immediately unload all loans they could write; with the triple-A ratings, regardless of actual quality). In fact, a fertile untapped mortgage market was speculators that wouldn't be able to ordinarily obtain such loans (non-owner occupied and planning on flipping before rate adjusted; one percent, interest only, no-documentation, no-down had trivial carrying cost compared to real-estate inflation in many parts of the country)
A lot of these toxic CDOs were then bought up by unregulated investment banking arms (courtesy of repeal of Glass-Steagall) of regulated financial institutions and carried off-balance. When the previous administration looked at starting to spend the TARP funds ... they found that it was trivial drop in the (toxic assets/CDO) bucket compared to toxic assets/CDOs that were being carried (off-balance) by just the top four regulated financial institutions ... and invented a totally different mechanism for using the money.
one of the possible motivations for this whole new mechanism ... was large percentage of the parties involved were taking percentages of the total deal (as opposed to bonuses tied to whether or not it lost money). this is slightly analogous to the old-time stock portfolio "churn" as mechanism for parties to increase their compensation.
another part of it was general social programs and retiree programs ... as the baby boomers were going thru their most productive working years (could really be soaked to maintain everybody else). there was recent numbers that baby boomer (bubble) is four times larger than the previous generation (taxes on baby boomers used to provide benefits to those retirees), but the generation following the baby boomer (bubble) is only half as large. that means as the baby boomers become the retirees, the ratio of workers to retirees changes by factor of eight times (1/8th as many workers per retiree as when the baby boomers were in their productive work years).
misc. past posts mentioning baby boomer generation moving
into retirement:
https://www.garlic.com/~lynn/2007v.html#26 2007 Year in Review on Mainframes - Interesting
https://www.garlic.com/~lynn/2008i.html#98 dollar coins
https://www.garlic.com/~lynn/2008l.html#37 dollar coins
https://www.garlic.com/~lynn/2008m.html#3 Medical care
https://www.garlic.com/~lynn/2008n.html#13 Michigan industry
https://www.garlic.com/~lynn/2008n.html#20 Michigan industry
https://www.garlic.com/~lynn/2008o.html#8 The end of the baby boomers, US bonds maturing, and then what?
https://www.garlic.com/~lynn/2008o.html#58 Everyone is getting same deal out of life: babyboomers can't retire but they get SS benefits intact
https://www.garlic.com/~lynn/2009m.html#61 August 7, 1944: today is the 65th Anniversary of the Birth of the Computer
https://www.garlic.com/~lynn/2009m.html#64 August 7, 1944: today is the 65th Anniversary of the Birth of the Computer
https://www.garlic.com/~lynn/2009o.html#37 Young Developers Get Old Mainframers' Jobs
https://www.garlic.com/~lynn/2009o.html#72 I would like to understand the professional job market in US. Is it shrinking?
misc. posts mentioning toxic CDOs:
https://www.garlic.com/~lynn/2009c.html#39 'WHO IS RESPONSIBLE FOR THE GLOBAL MELTDOWN'
https://www.garlic.com/~lynn/2009c.html#51 How to defeat new telemarketing tactic
https://www.garlic.com/~lynn/2009c.html#53 How to defeat new telemarketing tactic
https://www.garlic.com/~lynn/2009c.html#55 Who will give Citigroup the KNOCKOUT blow?
https://www.garlic.com/~lynn/2009c.html#61 Accounting for the "greed factor"
https://www.garlic.com/~lynn/2009c.html#65 is it possible that ALL banks will be nationalized?
https://www.garlic.com/~lynn/2009c.html#67 How to defeat new telemarketing tactic
https://www.garlic.com/~lynn/2009d.html#0 PNC Financial to pay CEO $3 million stock bonus
https://www.garlic.com/~lynn/2009d.html#7 Are Ctibank's services and products so vital to global economy than no other banks can substitute it?
https://www.garlic.com/~lynn/2009d.html#16 The Formula That Killed Wall Street
https://www.garlic.com/~lynn/2009d.html#18 HSBC is expected to announce a profit, which is good, what did they do differently?
https://www.garlic.com/~lynn/2009d.html#22 Is it time to put banking executives on trial?
https://www.garlic.com/~lynn/2009d.html#28 I need insight on the Stock Market
https://www.garlic.com/~lynn/2009d.html#37 NEW SEC (Enforcement) MANUAL, A welcome addition
https://www.garlic.com/~lynn/2009d.html#42 Bernard Madoff Is Jailed After Pleading Guilty -- are there more "Madoff's" out there?
https://www.garlic.com/~lynn/2009d.html#46 Bernanke Says Regulators Must Protect Against Systemic Risks
https://www.garlic.com/~lynn/2009d.html#62 Is Wall Street World's Largest Ponzi Scheme where Madoff is Just a Poster Child?
https://www.garlic.com/~lynn/2009d.html#73 Should Glass-Steagall be reinstated?
https://www.garlic.com/~lynn/2009d.html#74 Why is everyone talking about AIG bonuses of millions and keeping their mouth shut on billions sent to foreign banks?
https://www.garlic.com/~lynn/2009d.html#77 Who first mentioned Credit Crunch?
https://www.garlic.com/~lynn/2009e.html#8 The background reasons of Credit Crunch
https://www.garlic.com/~lynn/2009e.html#15 The background reasons of Credit Crunch
https://www.garlic.com/~lynn/2009e.html#23 Should FDIC or the Federal Reserve Bank have the authority to shut down and take over non-bank financial institutions like AIG?
https://www.garlic.com/~lynn/2009e.html#30 Timeline: 40 years of OS milestones
https://www.garlic.com/~lynn/2009e.html#36 Architectural Diversity
https://www.garlic.com/~lynn/2009e.html#53 Are the "brightest minds in finance" finally onto something?
https://www.garlic.com/~lynn/2009e.html#70 When did "client server" become part of the language?
https://www.garlic.com/~lynn/2009f.html#31 What is the real basis for business mess we are facing today?
https://www.garlic.com/~lynn/2009f.html#35 US banking Changes- TARP Proposl
https://www.garlic.com/~lynn/2009f.html#38 On whom or what would you place the blame for the sub-prime crisis?
https://www.garlic.com/~lynn/2009f.html#41 On whom or what would you place the blame for the sub-prime crisis?
https://www.garlic.com/~lynn/2009f.html#43 On whom or what would you place the blame for the sub-prime crisis?
https://www.garlic.com/~lynn/2009f.html#49 Is the current downturn cyclic or systemic?
https://www.garlic.com/~lynn/2009f.html#56 What's your personal confidence level concerning financial market recovery?
https://www.garlic.com/~lynn/2009f.html#65 Just posted third article about toxic assets in a series on the current financial crisis
https://www.garlic.com/~lynn/2009g.html#1 Future of Financial Mathematics?
https://www.garlic.com/~lynn/2009g.html#27 Flawed Credit Ratings Reap Profits as Regulators Fail Investors
https://www.garlic.com/~lynn/2009g.html#31 OODA-loop obfuscation
https://www.garlic.com/~lynn/2009g.html#34 Board Visibility Into The Business
https://www.garlic.com/~lynn/2009g.html#37 Future of Financial Mathematics?
https://www.garlic.com/~lynn/2009g.html#52 Future of Financial Mathematics?
https://www.garlic.com/~lynn/2009g.html#53 We Can't Subsidize the Banks Forever
https://www.garlic.com/~lynn/2009g.html#61 Prosecute Bank Execs
https://www.garlic.com/~lynn/2009h.html#3 Consumer Credit Crunch and Banking Writeoffs
https://www.garlic.com/~lynn/2009h.html#10 China's yuan 'set to usurp US dollar' as world's reserve currency
https://www.garlic.com/~lynn/2009h.html#29 Analysing risk, especially credit risk in Banks, which was a major reason for the current crisis
https://www.garlic.com/~lynn/2009h.html#40 Analysing risk, especially credit risk in Banks, which was a major reason for the current crisis
https://www.garlic.com/~lynn/2009n.html#13 UK issues Turning apology (and about time, too)
https://www.garlic.com/~lynn/2009n.html#17 UK issues Turning apology (and about time, too)
https://www.garlic.com/~lynn/2009n.html#20 UK issues Turning apology (and about time, too)
https://www.garlic.com/~lynn/2009n.html#21 UK issues Turning apology (and about time, too)
https://www.garlic.com/~lynn/2009n.html#47 Opinions on the 'Unix Haters' Handbook'
https://www.garlic.com/~lynn/2009n.html#56 Opinions on the 'Unix Haters' Handbook'
https://www.garlic.com/~lynn/2009n.html#58 Opinions on the 'Unix Haters' Handbook'
https://www.garlic.com/~lynn/2009n.html#62 Opinions on the 'Unix Haters' Handbook'
https://www.garlic.com/~lynn/2009n.html#68 Opinions on the 'Unix Haters' Handbook'
https://www.garlic.com/~lynn/2009o.html#23 Opinions on the 'Unix Haters' Handbook'
https://www.garlic.com/~lynn/2009p.html#23 Opinions on the 'Unix Haters' Handbook
https://www.garlic.com/~lynn/2009p.html#25 Opinions on the 'Unix Haters' Handbook
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 15:42:02 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
the smaller size of the following generation(s) (after baby boomer bubble) is somewhat independent of they apparently have lower education/skill level. in the wake of the 1990 census ... there was report that half of 18yr olds were functionally illiterate. the consensus seems to be that not only does the following generation have fewer individuals ... but they are much lower educated/skilled ... and therefor tend to have much lower paying jobs (which implies smaller GNP and smaller base for tax revenues; both fewer workers *AND* lower wages ... potentially drastically smaller tax revenues)
recent post on subject:
https://www.garlic.com/~lynn/2009s.html#43 Larrabee delayed: anyone know what's happening?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 16:45:20 -0500Mark Crispin <mrc@panda.com> writes:
i mistyped ... that should have been spring 2008 (actually 15feb2008) ... not "last spring".
wiki page
https://en.wikipedia.org/wiki/Comptroller_General_of_the_United_States
small time stuff had to do with administration ... major stuff was all with regard to congress and appropriations (ongoing refrain that nobody in congress for the last 50yrs was capable of even middle school math).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 17:34:41 -0500Mark Crispin <mrc@panda.com> writes:
misc (I had worked on a couple projects with) vera watson (RIP on
annapurna)
https://www.garlic.com/~lynn/2000.html#18 Computer of the century
https://www.garlic.com/~lynn/2002g.html#60 Amiga Rexx
https://www.garlic.com/~lynn/2007f.html#10 Beyond multicore
https://www.garlic.com/~lynn/2007f.html#12 FBA rant
also
http://www.mcjones.org/System_R/SQL_Reunion_95/sqlr95-Vera.html
quote from above:
Vera had married John McCarthy, the father of Lisp and of artificial
intelligence. John called my office to tell me that he had just learned
of this mess. I remember going in to a meeting of the department. I even
remember the conference room in Building 28 where we met. I told this
story roughly like this that Vera was lost. They did send up others to
try to find her. They were able to see the bodies in the snow way below
but it was not considered safe to descend, and even if you could descend
to the bodies, there was no way to bring the bodies back out without
bringing in helicopters and things like that, which were not considered
justified. So there's a memorial at the base camp at Annapurna today to
Vera, among others who've died on that assault; it's a serious mountain.
... snip ...
I had been blamed for computer conferencing on the internal (corporate)
network in the late 70s and early 80s ... misc. past posts about
internal network
https://www.garlic.com/~lynn/subnetwork.html#internalnet
somewhat as a result, there was a researcher paid to sit in the back of my office for nine months to take notes on how I communicated. They also got copies of all my (incoming & outgoing) email as well as logs of all instant messages. They would also go with me to meetings. This became a research report. It was also stanford phd thesis joint between language and computer (john ... oops, brain check, winograd was adviser on computer side) as well as some number of papers and books. misc. past posts mentioning computer mediated conversation (some of it related to the mention study and nine months of my life)
https://www.garlic.com/~lynn/subnetwork.html#cmc
misc. other posts mentioning original relational/sql
https://www.garlic.com/~lynn/submain.html#systemr
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Larrabee delayed: anyone know what's happening? Newsgroups: comp.arch Date: Mon, 04 Jan 2010 17:40:35 -0500Stephen Fuld <SFuld@alumni.cmu.edu.invalid> writes:
eventually morphing into current InfiniBand
https://en.wikipedia.org/wiki/InfiniBand
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 22:26:01 -0500Michael Wojcik <mwojcik@newsguy.com> writes:
one of (os/360) optimization from real storage heritage was that the channel programs were built in application space ... and then passed to the kernel (excp/svc0) and could do i/o directly into/out-of application space (i/o channel programs are run using real addresses).
in the move of os/360 to virtual memory ... these (application) channel programs would be specifying (application space) virtual address (not real). this represented same problem that cp67 had when virtual machine tried would do a sio specifying channel program; the channel program specified virtual machine "virtual addresses". cp67 handled the process by CCWTRANS routine that created a "shadow" copy of the (virtual) channel program ... which had the real addresses ... and it was the "shadow" channel program that was executed. The initial morphing of os/360 to virtual memory environment borrowed cp67's CCWTRANS to do equivalent shadow channel program creation in EXCP/svc0 processing.
cms under cp67 did similar channel programs ... which had required similar translation/copy process.
so one of the things i did for cp67/cms was implementation of
page-mapped filesystem ... could have all the benefits of "locate" mode
operation ... with elimination of all the translation/copy gorp. Also
there were other efficiencies possible because of the filesystem was
structured to correspond with the virtual address structure. ... misc.
past posts mentioning memory mapped work
https://www.garlic.com/~lynn/submain.html#mmap
a couple old emails about migrating a bunch of stuff, including
the memory-mapped filesystem from cp67/cms to vm370/cms:
https://www.garlic.com/~lynn/2006v.html#email731212
https://www.garlic.com/~lynn/2006w.html#email750102
https://www.garlic.com/~lynn/2006w.html#email750430
later in the 80s, with some tests of moderate i/o intensive vm370/cms application on 3380 disks ... running cms "locate" mode with standard filesystem versus page-mapped filesystem (everything identical, application, cms, 3380 disks, etc) ... i could get approx. three times thruput increase using page-mapped filesystem (vis-a-vis standard filesystem). This representing significantly more than simple elimination of channel program translation overhead.
there were a couple issues left-over from tss/360 experience (on 360/67) with its page-mapped filesystem. tss/360 just did the straight-forward memory mapping of the whole file ... which resulted in severe page fault behavior (no hints on what had already be used or what was about to be used). There is a some trade-off having application use read/write metaphor ... even in the paged-mapped environment ... where the application read/write can be used as indicators as to what data no longer is being used and what data will be used. There was also sometimes issue with some highly multithreaded applications becoming disabled when there was page fault for portions of paged mapped data.
one of the other features of the AIX VMM was part of it was initially built on 801/rios transaction/database memory. unix filesystem metadata was memory mapped into a transaction memory virtual segment. changes to the metadata in the transaction/database segment ... where captured and journaled (aka JFS). There were lots of early claims that 1) transaction memory implementation was faster than explicit log calls and 2) it was simpler than requiring log calls to be retrofitting to unix filesystem.
it is relatively trivial to show that it is simpler. however, when palo alto was looking at making JFS "portable" (to platforms w/o 801/rios transaction memory) ... they had to go thru and put in the necessary transaction log/journal explicit calls. They claimed that it was only a rather modest effort to retrofit the explicit calls ... but they were also able to show that it was faster (on identical hardware).
misc. past posts mentioning 801, romp, rios, iliad, etc
https://www.garlic.com/~lynn/subtopic.html#801
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Mon, 04 Jan 2010 23:32:29 -0500glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:
there is some efficiency issue with synchronous copies and asynchronous copies. processor doing synchronous copies can be processor bottleneck (also larger buffers that are copied passing thru the cache can trash useable cache information ... resulting in increased cache miss rate ... which may be as bad or worse than the direct processor cycles for the original copy).
outboard caching as part of asynchronous i/o ... can be completely transparent and have no effect on processor buffer copy overhead/thruput issues.
rdbms in unix environment will do something similar to cics/bdam & locate mode ... but with its own cache, asynch i/o, and possibly raw disks. disk controllers that do transparent caching can be problem for proper commit semantics ... where the rdbms needs to know that the transfer operation has made it to some permanent location (standard unix filesystem semantics may do lazy writes .. comparable to possible lazy writes by cached controllers) ... in this case it isn't just an efficiency issue ... but also a correctness issue.
most of the high-thruput rdbms also implement some amount of their own
multi-threading ... and make use of compare&swap instruction (or
instruction with similar semantics). charlie had invented compare&swap
when he was doing work on cp67 fine grain multiprocessing locking
... and then it took some doing to get it incorporated into 370 ..
recent post
https://www.garlic.com/~lynn/2010.html#0 Problem with XP scheduler?
801/rios (rs/6000) didn't have any compare&swap kind of operation ... so migrating various of these rdbms to rs/6000 & aixv3 represented something of a challenge. Eventually there was a hack done ... a fast-path emulation in the supervisor call interrupt routine. on non-multiprocessor systems ... atomic compare&swap is only necessary that the operation not be interrupted. A aixv3 compare&swap macro was eventually provided that did supervisor call ... which entered the supervisor call interrupt routine disabled for interrupts ... the fastpath code simualted compare&swap semantics and immediately returned to the application.
since 801/rios didn't have provisions for cache coherency and/or any plans for multiprocessor configurations ... disabled for interrupts was sufficient to achieve correct atomic compare&swap semantics.
the executive we reported to when we started ha/cmp
https://www.garlic.com/~lynn/subtopic.html#hacmp
transferred over to head up somerset ... which was going to adapt
801/rios to single chip implementation with cache coherency support and
being able to be used in multiprocessor configurations (among other
things). misc. past posts mentioning 801, romp, rios, iliad, power,
power/pc, etc
https://www.garlic.com/~lynn/subtopic.html#801
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Larrabee delayed: anyone know what's happening? Newsgroups: comp.arch Date: Tue, 05 Jan 2010 09:43:22 -0500a little SCI drift from long ago and far away
We had been trying to get HARRIER (9333) to interoperate with FCS
... instead it morphed into SSA. old reference to Jan92 meeting:
https://www.garlic.com/~lynn/95.html#13
Both SCI and FCS groups were in to SCSI committee with proposals for serial SCSI.
random sample from little later ... note that oo-operating systems were
something the rage; apple had pink (some of which morphed into taligent)
and sun had done doe/spring.
Date: Sat, 3 Feb 1996 10:17:19 -0800
From: wheeler
Subject: SCI
convex exemplar already does this ... SCI but with pa/risc chips.
sequent is exactly using the 4-way ... but with some industrial
strength re-engineering to the components.
a year ago ... sparc10 engineer had me in to talk about possibly
doing scale-up/commercializing sun's doe/spring oo-operating system.
the original SCI was conceived by Gustavson at SLAC and pushed thru
IEEE committee ... where it became more of a general purpose camel
than a race-horse. the sparc10 guy is now at xxxx and has developed an
optimized subset of the SCI protocol (looks a lot more like
Gustavson's original proposal) ... and in cmos chip can get about five
times the thruput of the dolphin GaAs chip. In practical terms, the
implication is that the subset chip could be used to cost-effectively
provide memory cache consistency between 100s-1000 of workstations
spread around a building ... compared to 10-256 processors in a
central complex (i.e. the dolphin GaAs sci design point).
At the time, I asked why they couldn't get hennessy (i.e. lots of the
mips/sgi scale-up and the dash/flash projects at stanford). The comment
was that DOE (the gov agency, not the sun oo-opsys) has hennessy off
trying to bail-out the DOE/intel teraflop computer.
in any case, we saw the doe/spring (oo-opsys) group at sun the other
afternoon ... looks like they've effectively been all moved over to
java ... and as they mentioned ... their work on doe/spring had only been to
scale-up into the tens of processors.
as per the attached ... with motorola effectively migrating all the
88k to the power/pc ... data general was hung-out for a processor. the
choices were pretty much to make the translation to power/pc ... or go
to intel.
note that SCI is NOT a one gigabyte per second protocol. The dolphin
GaAs chip is a one gigabyte implementation. SCI in fact is effectively
a set of synchronous bus-protocols redone with encapsulated packets and
made asynchronous. SCI can run over serial copper at 250mbits/sec,
slow fiber at 1gbit/sec ... and/or scale-up above 10gbytes/sec.
One of the problems the guy at motorola phoenix claimed as an SCI
problem was that it bottlenecked ... he effectively described an
implementation mapped to something that is akin to FDDI ring
architecture. While entry-level versions of SCI can be done that way
... SCI can also map to 10gbyte/sec, scallable, non-blocking cross-bar
switch (i.e. all attachments capable of concurrent full 10gbyte/sec
thruput).
In any case, the we have some opportunity to see the sequent machine
soon ... and everybody is going to have a hard-time beating their
price/performance & scale-up (as per our deployment platform analysis).
The board cache description is a little off ... it should be shared
4mbyte L2 (i.e. avg. 1mbyte/processor cache). The detailed
cycle-by-cycle hardware simulator that Oracle has developed for its
development platforms has been used to convince some vendors that
1mbyte is sort of entry ante cache size for Oracle ... and that 4mbyte
is still on the sweet part of the curve.
quad board with 4mbyte L2 cache should be a screamer for oracle
... and the sci scale-up to 64boards (256 processors) in a complex
will be a really tough price/performance point for everybody else.
... snip ... top of post, old email index.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Tue, 05 Jan 2010 14:33:29 -0500hancock4 writes:
this was raised recently in this thread by comments about the pain
various customers had in migrating from earlier machines to 360 (and
never wanting to ever do that again).
https://www.garlic.com/~lynn/2009s.html#60 360 programs on a z/10
also mentioned in the thread ... that lesson/concept got temporarily
lost in the future system detour
https://www.garlic.com/~lynn/submain.html#futuresys
there was a corollary ... that being the only vendor getting the single most important thing right ... it might be able to get lots of other things wrong ... and still dominate the market.
part of the issue back then was that software was somewhat simpler and tended to have machine/architecture dependeancies exposed. there has been some software progress in 40yrs being able to better abstract hardware features.
other posts in this thread:
https://www.garlic.com/~lynn/2009r.html#52 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#57 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#68 360 programs on a z/10
https://www.garlic.com/~lynn/2009r.html#69 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#4 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#9 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#13 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#16 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#17 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#20 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#24 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#26 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#28 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#29 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#30 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#34 360 programs on a z/10
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Tue, 05 Jan 2010 14:40:05 -0500John Levine <johnl@iecc.com> writes:
i've actually done paged-mapped filesystem ... that gains significant benefit in a virtual memory environment ... but still leveraged read/write application semantics ... analogy is large buffers are used to provide "window" of the data ... w/o actually requiring to map the whole file at the same time. in smaller real storage environment ... the read/write semantics gave hints to page manager about what was still being used and what wasn't (applications could use either the whole file specification or window specification as optimization option)
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 10:40:58 -0500glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:
controller disk cache option were introduced for 3880 ... an 8mbyte, 4k block 3880-11 (ironwood) cache ... and a 8mbyte 3380 track 3880-13 (sheriff) cache.
the 4k block ironwood cache was oriented towards block being accessed and then later still being in the cache when it was needed again.
the track sheriff cache was oriented towards sequential reads. sheriff publicity had blurbs about having 90% "hit" rates. The scenario was 3380 track formated with 10 4k records ... the first read on track would be a miss ... and then the next 9 sequential reads would be hits. I pointed out trivial change to file i/o operation specification to have ten record buffer (and the underlying system operation effectively would do full track read and do the buffer in system storage) ... would drop the sheriff hit rate to zero (for the same operation) ... aka it was even possible to specify buffering option on the JOB control DD card (not even requiring changes to the software).
For random access stuff ... the limited size of the sheriff cache (8mbyte) compared to size of the disk storage behind (640mbytes/drive and multiple drives) resulted in very low hit rate of randomlly accessed records. Even when 3880-13 was upgraded to 3880-23 with 32mbyte cache, it still would have very low hit rate for randomly accessed stuff.
4k record, ironwood cache option ... was somewhat more oriented to page pools ... and possibly other random oriented operations, not sequential reads. I pointed out the (a different kind of ) configuration problem with this ... which got a little more complex. In the timeframe that these 3880 controller caches were introduce ... a configuration would be 3081 processor with 32mbyte of real storage ... and a couple 3880s possibly with 8mbyte cache each. That would result in possibly 16mbyte of disk (page) cache ... and 32mbyte of system page space. There would be page fault, and page read ... it wouldn't be in disk cache so would be pulled from disk ... with copying being left in disk cache. The page is then in 32mbyte system storage. For almost all possible page use scenarios ... there was no scenario where a page wouldn't be in main memory and but would be in ironwood cache (every page in irownwood cache would always, also be in system memory).
I called this the "dup/no-dup" (duplicate/no-duplicate) scenario. To make ironwood cache useful ... the way the disk cache was used needed to be redone. I had originally implemented such a strategy for "page migration" and 2305 paging drums ... when the amount of real storage was comparable (or larger) than the "high-speed" secondary storage. If top-level of next level cache was much larger than the top level ... then percent of duplicates in the next level cache was small ... and didn't require additional management. When the size of the next level cache was about the same (or smaller) ... then the next level cache would be nearly all duplicates and of little additional use. In scenario with high percentage of duplicates ... it would be necessary to switch to a "no-duplicate" management strategy for the next level cache to be useful.
A "no-duplciate" management strategy can also be viewed as making the next level storage and adjunct to the higher level storage ... rather than straight cache. In the ironwood case ... switching to no-duplicate involved doing (destructive cache) read ... if it was in cache ... it would read it from cache ... and remove it from the cache. If it wasn't in the cache ... it would basically do a read directly from disk, bypassing cache. Page-out operations would write to cache and leave it there. That would mean page would never be in both system memory and cache memory at the same time (no-duplicate).
The ironwood analysis caused heartburn because they assumed it would be useful w/o any additional work (be totally transparent but still provide benefit) ... but the page scenario case was that it would never be useful using transparent strategy because all pages would be "duplicates" of what was in system storage. To make ironwood cache useful ... had to have capability of switching between "duplicate" and "no-duplicate" strategies.
misc. past posts mentioning ironwood, sheriff and/or dup/no-dup
strategy:
https://www.garlic.com/~lynn/93.html#13 managing large amounts of vm
https://www.garlic.com/~lynn/2000d.html#13 4341 was "Is a VAX a mainframe?"
https://www.garlic.com/~lynn/2001.html#18 Disk caching and file systems. Disk history...people forget
https://www.garlic.com/~lynn/2001b.html#61 Disks size growing while disk count shrinking = bad performance
https://www.garlic.com/~lynn/2001c.html#17 database (or b-tree) page sizes
https://www.garlic.com/~lynn/2001d.html#49 VTOC position
https://www.garlic.com/~lynn/2001d.html#68 I/O contention
https://www.garlic.com/~lynn/2001l.html#53 mainframe question
https://www.garlic.com/~lynn/2001l.html#54 mainframe question
https://www.garlic.com/~lynn/2001l.html#55 mainframe question
https://www.garlic.com/~lynn/2001l.html#63 MVS History (all parts)
https://www.garlic.com/~lynn/2002b.html#10 hollow files in unix filesystems?
https://www.garlic.com/~lynn/2002b.html#20 index searching
https://www.garlic.com/~lynn/2002d.html#55 Storage Virtualization
https://www.garlic.com/~lynn/2002e.html#11 What are some impressive page rates?
https://www.garlic.com/~lynn/2002f.html#20 Blade architectures
https://www.garlic.com/~lynn/2002o.html#3 PLX
https://www.garlic.com/~lynn/2002o.html#52 ''Detrimental'' Disk Allocation
https://www.garlic.com/~lynn/2003b.html#7 Disk drives as commodities. Was Re: Yamhill
https://www.garlic.com/~lynn/2003c.html#36 "average" DASD Blocksize
https://www.garlic.com/~lynn/2003f.html#5 Alpha performance, why?
https://www.garlic.com/~lynn/2003i.html#72 A few Z990 Gee-Wiz stats
https://www.garlic.com/~lynn/2003m.html#6 The real history of comp arch: the short form
https://www.garlic.com/~lynn/2003o.html#62 1teraflops cell processor possible?
https://www.garlic.com/~lynn/2004g.html#13 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#17 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#18 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#20 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#21 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#22 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004h.html#19 fast check for binary zeroes in memory
https://www.garlic.com/~lynn/2004i.html#1 Hard disk architecture: are outer cylinders still faster than inner cylinders?
https://www.garlic.com/~lynn/2004l.html#29 FW: Looking for Disk Calc program/Exec
https://www.garlic.com/~lynn/2005c.html#27 [Lit.] Buffer overruns
https://www.garlic.com/~lynn/2005m.html#28 IBM's mini computers--lack thereof
https://www.garlic.com/~lynn/2005m.html#30 Massive i/o
https://www.garlic.com/~lynn/2005t.html#50 non ECC
https://www.garlic.com/~lynn/2006.html#4 Average Seek times are pretty confusing
https://www.garlic.com/~lynn/2006c.html#8 IBM 610 workstation computer
https://www.garlic.com/~lynn/2006c.html#46 Hercules 3.04 announcement
https://www.garlic.com/~lynn/2006e.html#45 using 3390 mod-9s
https://www.garlic.com/~lynn/2006e.html#46 using 3390 mod-9s
https://www.garlic.com/~lynn/2006i.html#41 virtual memory
https://www.garlic.com/~lynn/2006j.html#11 The Pankian Metaphor
https://www.garlic.com/~lynn/2006j.html#14 virtual memory
https://www.garlic.com/~lynn/2006s.html#32 Why magnetic drums was/are worse than disks ?
https://www.garlic.com/~lynn/2006v.html#31 MB to Cyl Conversion
https://www.garlic.com/~lynn/2006y.html#35 The Future of CPUs: What's After Multi-Core?
https://www.garlic.com/~lynn/2007c.html#0 old discussion of disk controller chache
https://www.garlic.com/~lynn/2007c.html#12 Special characters in passwords was Re: RACF - Password rules
https://www.garlic.com/~lynn/2007c.html#23 How many 36-bit Unix ports in the old days?
https://www.garlic.com/~lynn/2007e.html#10 A way to speed up level 1 caches
https://www.garlic.com/~lynn/2007e.html#38 FBA rant
https://www.garlic.com/~lynn/2007e.html#42 FBA rant
https://www.garlic.com/~lynn/2007e.html#60 FBA rant
https://www.garlic.com/~lynn/2007l.html#61 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2008b.html#15 Flash memory arrays
https://www.garlic.com/~lynn/2008d.html#52 Throwaway cores
https://www.garlic.com/~lynn/2008f.html#19 Fantasy-Land_Hierarchal_NUMA_Memory-Model_on_Vertical
https://www.garlic.com/~lynn/2008i.html#41 American Airlines
https://www.garlic.com/~lynn/2008k.html#80 How to calculate effective page fault service time?
https://www.garlic.com/~lynn/2008s.html#39 The Internet's 100 Oldest Dot-Com Domains
https://www.garlic.com/~lynn/2009p.html#11 Secret Service plans IT reboot
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 11:08:02 -0500John Levine <johnl@iecc.com> writes:
I got dragged into some customer performance situations; examp. was
large national retailer with multiple mainframes in shared disk
configuration. They had single shared application (PDS) library shared
across whole complex. It had three (3330) cylinder PDS directory that
involved multi-track search every time any application was loaded
(target was possibly couple dozens of applications being loaded a second
across the whole complex). PDS directory search was taking avg. of
cylinder & half ... or approx. 30tracks. 3330 spun at 3600rpm ... each
PDS directory search (part of finding & loading every application) was
taking .5seconds elapsed time (effectively limiting the number of
application loads/sec to two). misc. past posts about CKD, FBA &
multi-track search
https://www.garlic.com/~lynn/submain.html#dasd
by the early 80s, I was making comments like the relative system thruput
of disks had declined by a factor of ten times over a period of years
and some disk executive took exception and assigned the division
performance group to refute the statements. example comparison table
I used (compared cp67/360-67 & vm370/3081):
https://www.garlic.com/~lynn/93.html#31 Big I/O or Kicking the Mainframe out the Door
after a couple weeks, they came back and said I had slightly understated
the situation (when "RPS-miss" was taken into account). They then turned
the comparison around and used if for a SHARE user group presentation on
how to optimize disks to improve system thruput. a couple different bits
& pieces from that presentation
https://www.garlic.com/~lynn/2002i.html#18
https://www.garlic.com/~lynn/2006o.html#68
misc. other posts mentioning the 360/67 & 3081 comparison
https://www.garlic.com/~lynn/94.html#43 Bloat, elegance, simplicity and other irrelevant concepts
https://www.garlic.com/~lynn/94.html#55 How Do the Old Mainframes Compare to Today's Micros?
https://www.garlic.com/~lynn/95.html#10 Virtual Memory (A return to the past?)
https://www.garlic.com/~lynn/98.html#46 The god old days(???)
https://www.garlic.com/~lynn/99.html#4 IBM S/360
https://www.garlic.com/~lynn/2001d.html#66 Pentium 4 Prefetch engine?
https://www.garlic.com/~lynn/2001f.html#62 any 70's era supercomputers that ran as slow as today's supercomputers?
https://www.garlic.com/~lynn/2001l.html#40 MVS History (all parts)
https://www.garlic.com/~lynn/2001l.html#61 MVS History (all parts)
https://www.garlic.com/~lynn/2001m.html#23 Smallest Storage Capacity Hard Disk?
https://www.garlic.com/~lynn/2002.html#5 index searching
https://www.garlic.com/~lynn/2002b.html#11 Microcode? (& index searching)
https://www.garlic.com/~lynn/2002b.html#20 index searching
https://www.garlic.com/~lynn/2002e.html#8 What are some impressive page rates?
https://www.garlic.com/~lynn/2002e.html#9 What are some impressive page rates?
https://www.garlic.com/~lynn/2004p.html#39 100% CPU is not always bad
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 11:29:01 -05001022 guy <s1022.xoxo@xoxy.net> writes:
vm370 tried to do some interesting formating of dasd to optimize various rotational latency. basically vm370 would do pre-formating ckd dasd to 4k blocks ... and then simulate FBA from then own. vm370 also trivially added support for real FBA devices (compared to problem with the favorite son operating system ... these days all disks are physically FBA ... and the favorite son operating system requires an enormous layer of complexity in disk controllers to simulate CKD).
for 3330 disks ... it was possible to format three 4k blocks per track (with a little left over) ... with 19 tracks per cylinder ... that gave 57 4k blocks to head position (cylinder).
vm370 had support to reoder all queued requests to optimize arm motion (order seek) and rotational delay. for queued 4k requests (read or write) for the same cylinder ... it would attempt to to batch them into single chained I/O operation ... with the order to maximize transfers per rotation. basically the metaphor were there "slots" per rotation and 19 possible records for each slot (corresponding to the rotation position for slot on each track).
the problem was that I/O channel command architecture required the channel program be in mainframe memory and the channel/controller/device command fetch be strictly serialized sequential (no prefetching). For chained request that tried to read "slot 1" from one track and "slot 2" from a different track (in single revolution) ... that command sequence required an (added) "seek head" command. This introduced additional command processing latency to fetch and decode the additional command ... all the time that the disk continues to rotate. So vm370 used special CKD disk format that added short, physical dummy blocks (between the formated 4k data records). The theory being that the rotational latency to skip over the short dummy block ... would be sufficient to accommodate the latency involved in fetching, decoding, and executing the "seek head" command.
misc. past posts mentioning dummy record formating:
https://www.garlic.com/~lynn/2003g.html#22 303x, idals, dat, disk head settle, and other rambling folklore
https://www.garlic.com/~lynn/2004d.html#64 System/360 40 years old today
https://www.garlic.com/~lynn/2004d.html#65 System/360 40 years old today
https://www.garlic.com/~lynn/2004d.html#66 System/360 40 years old today
https://www.garlic.com/~lynn/2006t.html#19 old vm370 mitre benchmark
https://www.garlic.com/~lynn/2008s.html#52 Computer History Museum
https://www.garlic.com/~lynn/2009p.html#12 Secret Service plans IT reboot
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 11:37:39 -0500Michael Black <et472@ncf.ca> writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 12:39:09 -0500Pat Farrell <pfarrell@pfarrell.com> writes:
started being standard ... if it had intermediate electronic stroage (say track cache) in the controller ... or much more tightly coupled to primary processor storage.
DBMS transaction logs tended to be write-only (except in case of failure and recovery). for high transaction systems ... where log starts to be bottleneck ... tended towards full-track log records and disks where the arm remained in position with little or no activity on the disk. This eliminate arm motion latency. But there would still be avg. rotational delay latency.
before full-track (3880-13 controller) cache ... there was gimick for IMS log around this. Now normally, log has a set of tracks that are written sequentially until the end ... and then log restarts from the front (overlaying oldest records). recovery requires determining which is the track with the tail (newest) track and track for the head (oldest) track.
the gimick was to format track with 4k blocks and then set up the channel program ... instead of search for the first record on the track to start write of full track of 4k blocks ... the search was for any record on the track to start write for 4k blocks. then during recovery, the "head" & "tail" log record for each track had to be determined (in addition to "head" & "tail" of the whole log ... basically use the information in each record for determining log "head" & "tail" for also determining the "head" & "tail" record on each track).
as mentioned in the previous post ... application doing full track read/write ... eliminated the 3880-13 full track cache claims about "90% hit rate" ... aka sequently reading individual 4k blocks versus reading full track of 4k blocks.
other post in locate mode threaed:
https://www.garlic.com/~lynn/2010.html#42 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#43 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#46 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#48 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#49 locate mode, was Happy DEC-10 Day
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 13:09:18 -0500John Levine <johnl@iecc.com> writes:
also amount of electronic storage has increased significantly .. to the point that electronic storage started being used for caching as mechanism to compensate for disk i/o thruput bottlenecks. part of my pontification on the subject in the 70s ... was having started "dynamic adaptive resource management" as undergraduate in the 60s ... along with something I called "scheduling to the bottleneck". To be able to (dynamically) "schedule to the bottleneck" ... required (dynamically) determining primary bottlneck(s).
One of the things in the 60s database ... with exposed record numbers (whether absolute or relative) as part of the (application) data ... had side effect of conserving real storage (as well as disk operation efficiency).
with the original relational/sql implementation in the 70s ... some
past posts
https://www.garlic.com/~lynn/submain.html#systemr
there was some contention between IMS (60s database with "exposed" record numbers) and System/R. System/R had abstracted away "exposed" record numbers with the concept of primary key field (and implicit index as part of the internal operation that provided the mapping from key to disk record).
The IMS group would claim that the internal/implicit (relational) index typically doubled total disk space (compared to same implementation in IMS) and added 4-5 (or more) disk i/os for every record access (to move thru various levels of the index). System/r countered that the "exposed" record numbers significantly increased the application and administrative overhead. Going into the 80s ... there was enormous increase in system memory (allowing much of RDBMS index to be cached in storage, eliminated many of the additional disk i/os), significant increase in disks space, and significant decrease in price/mbyte of disk space (mitigating the additional index disk space requirements). At the same time peole overhead was becoming relatively more expensive (compared to declining hardware costs) and DBMS skills were becoming scarce as demand went up. As a result there was some tipping in various trade-offs from the 60s "physical database" to the 70s "logical database" (tradeoff of people time/skills against hardware costs ... as well as resources that could compensate for disk thruput limitations).
That doesn't mean that IMS has gone completely away ... large, high
value applications ... that have kept a relatively static information
structure (over the years ... minimizing all the infrastructure gorp
needed to deal with information restructure in 60s dbms technology)
are still around. a couple recent posts mentioning IMS still being
around:
https://www.garlic.com/~lynn/2009r.html#20 70 Years of ATM Innovation
https://www.garlic.com/~lynn/2009r.html#21 Small Server Mob Advantage
One of the sporadic complaints about current RDBMS paradigm is it did do
trade-off from generalized relational for collecting all fields
associated with primary key in the same physical record (and having an
implicit common homogeneous, relational structure for all records in
table). This was something of a trade-off ... to show RDBMS transaction
thruput better approaching that of IMS for financial/account operations
... like an ATM cash machine. A few posts mentioning row/column table
limitation/tradeoff:
https://www.garlic.com/~lynn/2000b.html#64 definitions
https://www.garlic.com/~lynn/2004e.html#49 c.d.theory glossary (repost)
https://www.garlic.com/~lynn/2005.html#24 Network databases
https://www.garlic.com/~lynn/2005d.html#57 Thou shalt have no other gods before the ANSI C standard
https://www.garlic.com/~lynn/2005f.html#23 the relational model of data objects *and* program objects
https://www.garlic.com/~lynn/2009s.html#56 DEC-10 SOS Editor Intra-Line Editing
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 13:37:06 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
disclaimer ... at the same time I was doing various pieces of work on
system/r (original relational/sql implementation)
https://www.garlic.com/~lynn/submain.html#systemr
I also was doing some of the work for a much more generalized implementation that supported/allowed arbitrary (possibly non-homogeneous) relationship between items (and used a different kind of abstraction to eliminate the record pointer paradigm).
I've done periodic re-implementations of such things over the
years ... and use a recent one for maintaining information from which
I generate my rfc index
https://www.garlic.com/~lynn/rfcietff.htm
as well as managing information i use for generating various merged
taxonomies & glossaries
https://www.garlic.com/~lynn/index.html#glosnote
there are a couple articles mentioned in the first part of the
index
https://www.garlic.com/~lynn/index.html
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 13:57:13 -0500Mensanator <mensanator@aol.com> writes:
in some sense it was targeted as follow-on (and much less expensive) to
the 2321 datacell (drastically reducing cost per megabyte ... allowing
lower value data to be brought online) ... 2321 reference:
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_2321.html
3850MSS did something similar but in much larger and much more expensive
package ... 3850MSS reference
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_3850.html
there was problem that the floppy material had tendency to stretch under constant rotation (which messed up the whole thing).
other posts in this locate mode thread:
https://www.garlic.com/~lynn/2010.html#42 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#43 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#46 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#47 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#48 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#49 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#51 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#52 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#53 locate mode, was Happy DEC-10 Day
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Daylight Savings Time again Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 14:25:42 -0500Olafur Gunnlaugsson <oli@audiotools.com> writes:
DOJ, UBS and Swiss Seek Stay in $14b Tax Haven Court Proceedings
http://www.popdecay.com/2009/07/13/doj-ubs-and-swiss-seek-stay-in-14b-tax-haven-court-proceedings/2510
there was all sort of stuff about turning over information for the largest 500-1000 accounts as first step and then follow-on steps for the other 50k accounts. Then there was IRS saying that people that came forward voluntarily (before the IRS got the information from the swiss), wouldn't be criminally prosecuted.
UBS statement of Mark Branson before the permanent Subcommittee of Investigation
http://www.ubs.com/1/e/investors/releases?newsId=145699
US crossborder-case Key developments
http://www.ubs.com/1/e/index/crossborder/johndoesettlement/key_developments.html
UBS: Banking's Secret no More
http://www3.gsionline.com/legalcurrents/Article_20090226_E1.asp?contactid=LearnWLCB
Tax haven crackdown showing cracks
http://www.qwealthreport.com/blog/tax-haven-crackdown-showing-cracks/
Reaction to UBS Debacle: Tough New Foreign Tax Laws
http://www.law.com/jsp/law/international/LawArticleIntl.jsp?id=1202435186330
... lots of other stuff found with search engine
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Daylight Savings Time again Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 14:36:06 -0500re:
swiss banks also "outsourced" some kinds of transactions to a neighboring state ... so that they technically didn't occur in switzerland. many were of the nature that it got the neighboring state put on the treasury's money laundering black list.
a couple years ago we were at a conference of EU financial (large banks, stock exchanges, investment institutions, etc) CEOs that was held there (it looked like part of effort to clean up their image and get off the treasury's money laundering black list) and stayed at an inn (parts of it dated back to 13th century). In some discussions with the innkeaper ... he made a joke about getting constant flow of people from the US with business cards that read "dept. of money laundering".
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 14:46:32 -0500Walter Bushell <proto@panix.com> writes:
with fedora11 (ext3 filesystem) thru 100mbit ethernet router ... sustained transfer would peak at 10mbyte/sec. going to fedora12 (ext4 filesystem with contiguous allocation support) and upgrading router to gigabit (all my enet adapters already supported gigabit) ... it now sustains at least 20mbyte/sec and sometimes transfers show sustained 40mbyte/sec.
a couple past posts mentioning ext4 filesystem
--
https://www.garlic.com/~lynn/2009e.html#55 Gone but not forgotten: 10 operating systems the world left behind
https://www.garlic.com/~lynn/2009h.html#63 Operating Systems for Virtual Machines
40+yrs virtualization experience (since Jan68), online at home since Mar1970
locate mode, was Happy DEC-10 Day
From: Anne & Lynn Wheeler <lynn@garlic.com>
Subject: Re: locate mode, was Happy DEC-10 Day
Newsgroups: alt.sys.pdp10,alt.folklore.computers
Date: Wed, 06 Jan 2010 14:59:44 -0500
scott@slp53.sl.home (Scott Lurndal) writes:
Because copying the data pollutes the processor caches with use-once
data. Because the processor could be doing work for another thread
instead of copying data.
previous comment (in this thread) related to cache-bypass
https://www.garlic.com/~lynn/2010.html#43 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#47 locate mode, was Happy DEC-10 Day
if you don't have a cache-bypass move instruction ... copying a megabyte of data is going to take 2mbytes of cache lines ... megabyte for the source location and megabyte for the target location ... little or none of it is possibly going to be actually used by the processor (and therefor would actually be required in the cache ... and the cache gets really polluted and copy has forced out stuff that might actually be useful for processor performance).
misc. other past posts mentioning cache-bypass:
https://www.garlic.com/~lynn/95.html#15 multilevel store
https://www.garlic.com/~lynn/2001b.html#37 John Mashey's greatest hits
https://www.garlic.com/~lynn/2004g.html#17 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2004g.html#18 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2006c.html#42 IBM 610 workstation computer
https://www.garlic.com/~lynn/2007r.html#44 complicated address generation unit?
https://www.garlic.com/~lynn/2008d.html#1 What happened to resumable instructions?
https://www.garlic.com/~lynn/2008m.html#24 Some confusion about virtual cache
https://www.garlic.com/~lynn/2008r.html#44 another one biting the dust?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 16:06:57 -0500"1022 guy" <s1022.xoxo@Use-Author-Supplied-Address.invalid> writes:
disk revolve around 3600rpm, 60times/sec, 1/60th sec. per revolutioin.
channel programs can be "chained" together so that transition from record transfer to record transfer is handled asynchronously in the channel.
creation of new channel programs can go own asynchronously with the execution of existing channel programs ... on 3330 with 4k blocks per track-revolution ... could easily create a single channel program for half dozen to dozen 4k transfers (2-4 revolutions).
on cp67, I got the pathlength to 1) take page fault, 2) execute page replacement algorithm, 3) queue page transfer request, 4) dispatch a different task, 5) take resulting I/O interrupt, 6) update finished request, 7) redispatch original task ... to about 500 instructions (part of I/O interrupt processing was pro-rated ... since a single I/O interrupt might signal the completion of a I/O operation with half dozen transfers).
example is little hard to see with disk ... where there is going to be unproductive time with arm transfer. easier to see in example of a paging drum with fixed-head per track (and no arm motion). 2301 paging drums were available on 360/67 ... and finally tuned ... so there were no idle transfer time per revolutions ... would do 300 page transfers per second (no missed records and no lost time for arm motion). Pathlength for sustaining that page transfer rate might take 10-15 percent of the processor (including all the page fault handling, task switching, etc).
Part of it, was building the channel programs could be overlapped with the channel execution of the channel programs. Part of it was I had reduced the pathlength for the whole thing by better than factor of ten times ... and part of it was some tricks involved with taking an interrupt (signaling the end of previous channel program) and (re)driving the next channel program (which was synchronous operation) ... being able to do that in tens of instructions (and part of it was redrive might only be required once every 15-20 transfers ... aka that many in a single channel program on 2301 fix-head device).
for 370, there was a 2305, fixhead-disk ... physically a disk ... head for every track on every platter. 2305 had something called "multiple exposures" ... aka processing could schedule up to eight independent channel programs simulataneously for the 2305 ... which mitigated the synchronous requirement to take an interrupt and get the next channel program started (while the disk is rotating and not miss a record) ... since each 2305 could have seven other I/O requests already queued up in the hardware.
some recent posts discussing needing efficient I/O redrive ... to minimize
device idle time:
https://www.garlic.com/~lynn/2009o.html#19 What happened to computer architecture (and comp.arch?)
https://www.garlic.com/~lynn/2009q.html#74 Now is time for banks to replace core system according to Accenture
https://www.garlic.com/~lynn/2009r.html#52 360 programs on a z/10
https://www.garlic.com/~lynn/2009s.html#16 Why Coder Pay Isn't Proportional To Productivity
https://www.garlic.com/~lynn/2009s.html#29 channels, was Larrabee delayed: anyone know what's happening?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 16:19:05 -0500Eric Chomko <pne.chomko@comcast.net> writes:
there was something called "Boca West" ... but it was a Series/1 group with offices in Palo Alto Square (corner of page mill & el camino) ... a Boca effort from the 70s.
Series/1 had a "formal" operating system called RPS ... joke was that it was done by some of the OS/360 MFT people from kingston that had retired to Boca (looking to reinvent MFT). The folklore was that the alternative Series/1 system, EDX ... was done by a summer employee physics grad student at san jose research.
Later, there have been some comments ... that OS/2 also bears some similarties with MFT.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 15:38:28 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
there has been some churn in business news shows ... something about some statement about Bernanke comments about fed-rate and the financial mess.
the issue about the fed regulations and the fed-rate ... was it affected mortages by regulated depository financial institutions ... which also had motivation to qualify borrowers.
the problem with the 1percent, interest-only, no-documentation, no-down loans ... was that they were being done by unregulated load originators who found that they didn't have to care about borrowers qualifications. Being able to "buy" triple-A ratings for everything they sold ... met 1) they had unlimited source of funds (totally divorced from needing deposits as source of funds and 2) the could unload every loan they made regardless of quality.
In fact, big untapped market were borrowers that couldn't get mortgages from anywhere else. Low-income, first time buyers fell into that category ... but it turns out that was minor percentage of the opportunity. It was speculators ... who would buy up everything ... especially the big ticket mc'mansions (much, much larger part of the market than the low-income 1st time buyers). With annual real-estate inflation running 10-20 percent in several parts of the country (once the process got going ... largely fueled also by the speculators), the no-down, 1percent, interest only ARMs had much lower carrying cost than the annual inflation (with the speculators planning on flipping the property before the rate adjusted).
The big failing was in the normal process ... the seculators wouldn't have been given the loans at all ... but in the unregulated loan originators able to buy triple-A ratings ... all that changed.
the interesting aspect is how did several trillion of these (mortgage-backed) toxic CDOs eventually find their way to off-balance books of the four largest, too big to fail, bailed-out financial institutions? (amount is so large, that when the treasury got around to figuring it out ... it realized that the appropriated TARP funds, for buying these assets, wasn't going to make dent in the problem ... and invented a different way for using the funds).
One of the enablers was repeal of Glass-Steagall ... that allowed regulated financial institutions to have unregulated investment banking arms ... who could, in turn actually buy the toxic CDOs (for carrying off-balance). It seems like big motivation was that all the players got to take a percentage of the size of the deals (regardless of risk, eventual loss or profit, etc) ... and that (personal financial) motivation more than offset the possibility that the deals could take down the company (or the country).
I heard reference to joke during the height ... likening the situation to musical chairs ... which institutions would be still holding the toxic CDOs when the bubble burst.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 17:29:44 -0500"John Varela" <OLDlamps@verizon.net> writes:
trying to offset ... I would post in an PC internal discussion list,
"quantity 1" clone prices from sunday mercury news ... past reference:
https://www.garlic.com/~lynn/2001n.html#81 a.f.c history checkup... (was What specifications will the standard year 2001 PC have?)
https://www.garlic.com/~lynn/2001n.html#82 a.f.c history checkup... (was What specifications will the standard year 2001 PC have?)
boca business hired dataquest to due some indepth surveys ... and some other studies of the PC market ... including a couple hour video taped round table of silicon valley "experts". I knew the dataquest people ... and they asked me if I would be one of the "experts" ... they promised to garble my introduction so BOCA didn't realize they had a ringer. I also cleared participation with my line-of-business executives (also not to tell boca what was going on) ... this was all before gartner bought dataquest.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 20:42:30 -0500Peter Flass <Peter_Flass@Yahoo.com> writes:
cp67 that was installed at the univ. jan68 ... did fifo and single page transfer per IO/channel program. one of the things i did as undergraduate in the 60s was rewrite code to do ordered seek queueing ... and where possible, multiple page transfers per channel program. prior to the multiple chained requests, 2301 fixed-head drum would peak-out at about 80 transfers per second (every transfer included half avg. rotational delay). after the chained request, could get 2301 fixed-head drum to peak out at 300 transfer per second (basically hardly missing record). The ordered seek increased heavy load nominal thruput on 2314 disk by about 50%.
there were some fancy tricks for implementation done by LA SE for
supporting (very large number of) ATM cash machines ... a couple refs:
https://www.garlic.com/~lynn/2010.html#11 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#19 Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#52 locate mode, was Happy DEC-10 Day
https://www.garlic.com/~lynn/2010.html#53 locate mode, was Happy DEC-10 Day
there was some layout where the account transaction location could be calculated from the account number. before the transaction was "released" for execution, there was infrastructure to start the arm in motion to the desired arm position. If the next arm position was some distance off, transaction would be delayed (proportional to the arm motion distance) ... based on some probability (heavy transaction load) that a new transaction would arrive (during the delay) requiring a record between the current arm position and the next arm position (being able to pick up a transaction almost for free ... in terms of disk arm thruput).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Wed, 06 Jan 2010 20:53:26 -0500re:
i.e.
Bernanke Won't Rule out Interest Rate Hike
http://www.cbsnews.com/stories/2010/01/04/business/main6052099.shtml
from above:
Bernanke said the direct links were weak between super-low interest
rates and the rapid rise in house prices that occurred at roughly the
same time. The stance of interest rates during that period "does not
appear to have been inappropriate," he said.
... snip ...
i.e. the financial mess was in large part caused by unregulated loan originators have no motivation to pay any attention to borrowers qualification (which in turn was result of being able to package up the mortgage and "pay" for triple-A ratings, regardless of actual quality ... being able to unload every mortgage they made ... as fast as they could make them).
the FEDs problem was that their regulation and the FED interest rates ... effectively had no impact on what the unregulated load originators were doing.
misc. past posts mentioning Bernanke statements:
https://www.garlic.com/~lynn/2007t.html#50 Newsweek article--baby boomers and computers
https://www.garlic.com/~lynn/2008e.html#42 Banks failing to manage IT risk - study
https://www.garlic.com/~lynn/2008e.html#65 Banks failing to manage IT risk - study
https://www.garlic.com/~lynn/2008h.html#90 subprime write-down sweepstakes
https://www.garlic.com/~lynn/2008i.html#30 subprime write-down sweepstakes
https://www.garlic.com/~lynn/2008i.html#77 Do you think the change in bankrupcy laws has exacerbated the problems in the housing market leading more people into forclosure?
https://www.garlic.com/~lynn/2008j.html#38 dollar coins
https://www.garlic.com/~lynn/2008j.html#76 lack of information accuracy
https://www.garlic.com/~lynn/2008m.html#16 Fraud due to stupid failure to test for negative
https://www.garlic.com/~lynn/2008m.html#26 Fraud due to stupid failure to test for negative
https://www.garlic.com/~lynn/2009d.html#22 Is it time to put banking executives on trial?
https://www.garlic.com/~lynn/2009d.html#30 I need insight on the Stock Market
https://www.garlic.com/~lynn/2009d.html#36 Bernanke Says Regulators Must Protect Against Systemic Risks
https://www.garlic.com/~lynn/2009d.html#40 Bernanke Says Regulators Must Protect Against Systemic Risks
https://www.garlic.com/~lynn/2009d.html#45 Bernanke Says Regulators Must Protect Against Systemic Risks
https://www.garlic.com/~lynn/2009d.html#46 Bernanke Says Regulators Must Protect Against Systemic Risks
https://www.garlic.com/~lynn/2009f.html#43 On whom or what would you place the blame for the sub-prime crisis?
https://www.garlic.com/~lynn/2009f.html#47 TARP Disbursements Through April 10th
https://www.garlic.com/~lynn/2009f.html#49 Is the current downturn cyclic or systemic?
https://www.garlic.com/~lynn/2009f.html#53 What every taxpayer should know about what caused the current Financial Crisis
https://www.garlic.com/~lynn/2009f.html#56 What's your personal confidence level concerning financial market recovery?
https://www.garlic.com/~lynn/2009f.html#65 Just posted third article about toxic assets in a series on the current financial crisis
https://www.garlic.com/~lynn/2009g.html#3 Do the current Banking Results in the US hide a grim truth?
https://www.garlic.com/~lynn/2009g.html#27 Flawed Credit Ratings Reap Profits as Regulators Fail Investors
https://www.garlic.com/~lynn/2009g.html#34 Board Visibility Into The Business
https://www.garlic.com/~lynn/2009g.html#37 Future of Financial Mathematics?
https://www.garlic.com/~lynn/2009g.html#53 We Can't Subsidize the Banks Forever
https://www.garlic.com/~lynn/2009h.html#3 Consumer Credit Crunch and Banking Writeoffs
https://www.garlic.com/~lynn/2009h.html#25 The Paradox of Economic Recovery
https://www.garlic.com/~lynn/2009h.html#29 Analysing risk, especially credit risk in Banks, which was a major reason for the current crisis
https://www.garlic.com/~lynn/2009h.html#49 IBM to Build Europe, Asia 'Smart Infrastructure'
https://www.garlic.com/~lynn/2009i.html#60 In the USA "financial regulator seeks power to curb excess speculation."
https://www.garlic.com/~lynn/2009j.html#35 what is mortgage-backed securities?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Oldest Instruction Set still in daily use? Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 21:05:24 -0500some past posts mentioning various faa.gov web pages (some gone 404 and have to go to the "way back" machine);
for instance:
Host and Oceanic Computer System Replacement Equipment for the National
Airspace System
https://web.archive.org/web/20011221011933/http://www.faa.gov/aua/ipt_prod/enroute/hocsrfct.htm
current faa history page (with several references):
http://www.faa.gov/about/history/
I wasn't ever involved in projects ... one of the "modernization efforts" in the early 90s was being done by somebody we knew pretty well. He sort of had a different job during the day ... but was spending a lot of spare minutes programming in ADA for the modernization effort.
also since we were doing ha/cmp ... and some of the "modernization efforts" involved rs/6000 triple replicated configurations ... we got to see participate in some of the analysis. a basic premise was that there would be lots of system infrastructure to mask all possible failures from the application code. the application programmers could then take the functional description and do a straight-forward implementation w/o having to ever worry about contingencies. however, there turned out to be some classes of failures/glitches that were at the application/business level (not at the system level) ... and as a result were getting lost.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Wed, 06 Jan 2010 23:29:16 -0500Andrew Swallow <am.swallow@btopenworld.com> writes:
i guess there might have been some chance of having been influenced by the example of boca series/1 and RPS (having been done in house). there are a number of issues doing things by contracts with outside entities (one such is that it cuts down internal political competitors)
boca played politics ... like later forcing the rs/6000 to use the
microchannel adapter cards designed for PS2 (totally different cost &
performance trade-offs ... joke was that it would slow down rs/6000 to
thruput of ps2) ... recent posts/reference:
https://www.garlic.com/~lynn/2009j.html#64 A Complete History Of Mainframe Computing
https://www.garlic.com/~lynn/2009l.html#15 SNA: conflicting opinions
https://www.garlic.com/~lynn/2009p.html#16 WSJ.com - IBM Puts Executive on Leave
https://www.garlic.com/~lynn/2009r.html#15 Small Server Mob Advantage
one was that the ps2 16mbit token/ring microchannel card that had lower thruput than the 4mbit t/r isa/at card that the workstation division had done for the pc/rt.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Thu, 07 Jan 2010 10:05:32 -0500Anne & Lynn Wheeler <lynn@garlic.com> writes:
the claim was that the vm370-based ATM cash machine implementation on 370/158 was outperforming similar implementation on TPF running on 370/168 (TPF was/is IBM's superfast transaction system and 168 was about 2.5 times the processor speed of 158).
the claim was that while short pathlength was useful for handling large number of ATM cash machine transactions ... more important was the optimization of disk arm ... including anticipation of requests that might not have yet come in. some of this could be based on long-term patterns of ATM cash machine use ... like specific machines near lunch rooms at lunch hrs ... and the account numbers would tend to be from specific branches (the vm370-based implementation used a lot more information about ATM cash machine use patterns for scheduling transactions as part of optimizating disk arm patterns).
TPF wiki page ...
https://en.wikipedia.org/wiki/Transaction_Processing_Facility
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Thu, 07 Jan 2010 10:53:20 -0500John Levine <johnl@iecc.com> writes:
"stand-alone" seek had always been standard ... at its peek, 360/67 at the science center had 45 2314 drives (five standard 8-drive 2314s and a "short" five-drive 2314 ... science center was on 4th flr of 545 tech sq ... but machine room was on 2nd flr).
standard CKD/dasd channel program was:
seek
searchid equal
tic *-8
read/write
... standard process was for the operating system to unchain the seek
operation ... and execute it all by itself. then when the arm was in
position, it would rechain the seek and reexecute the whole operation.
under heavy load, there could easily be half dozen stand-alone seeks going on concurrently.
standard disk formating was to have an ID field as part of each record. for vtoc or pds directory it could be something symbolic meaningful (and there would be multitrack search looking for the symbolic value anywhere on track). for more mundane formating ... it would be something like "CCHHR" (aka slightly simulating fba ... it would be the physical record position). the searchid-equal would do compare every time a record id field came under the head. if it failed ... it would drop thru to the next channel command (and the tic *-8 would branch back to repeat the operation); if it succeeded, it would branch to +8. A standard search would signal exception and stop if it made whole revolution w/o a match. A "multi-track" search would keep repeating the search on the following tracks (on same cylinder) until it reached the end of the cylinder.
reference standard 8-drive 2314 ... actually 9-drives ... but only
8 drives could be addressed at any time
http://www-03.ibm.com/ibm/history/exhibits/storage/storage_2314.html
CKD DASD Command Codes:
https://www.garlic.com/~lynn/gcard.html#26
misc. past posts mentioning CKD, FBA, and/or multi-track search
https://www.garlic.com/~lynn/submain.html#dasd
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: What happened to Eric Loriaux' web site Newsgroups: bit.listserv.ibm-main Date: 7 Jan 2010 09:07:31 -0800shmuel+ibm-main@PATRIOT.NET (Shmuel Metz , Seymour J.) writes:
gives last update for the above as 30mar2008 ... and doesn't
show anything past 31may2008 (or for 2009). somethings similar
for portal.loriaux.com
https://web.archive.org/web/*/http://portal.loriaux.com
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@garlic.com (Lynn Wheeler) Date: 7 Jan, 2010 Subject: Post Office bank account 'could help 1m poor' Blog: Payment Systems NetworkPost Office bank account 'could help 1m poor'
In the US, there has been estimates that as much as 1/3rd are "unbanked" ... because their business is below the existing financial industry cost structure.
there has been lots of concern for more than the last decade ... that WalMart (&/or M'soft) would get into financial services. WalMart has been notorious for drastically reducing cost of doing business and being able to undercut competitors (by offering comparable services for lower price) ... and there is concern that WalMart could achieve something similar for financial services ... including being able to acquire nearly all of the current "unbanked".
Testimony on the floor for the 1999 bank modernization act (GLBA, also repealed Glass-Steagall) included statements that a major purpose for the act was to prevent institutions that weren't already banks from becoming banks ... specifically WalMart and M'soft.
In the past decade, there was lots of furor when it looked like WalMart might require an ILC charter ... allowing it to do financial operations. WalMart had releases that ILC would only be used to become its own merchant acquiring financial institution (thereby effectively reducing its interchange fees) ... but there was still lots of concern that WalMart could use the ILC charter to branch into issuing and other financial services.
misc. past posts mentioning WalMart & ILC charter
https://www.garlic.com/~lynn/2007i.html#42 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2007i.html#47 John W. Backus, 82, Fortran developer, dies
https://www.garlic.com/~lynn/2008c.html#7 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008c.html#12 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008c.html#25 Toyota Sales for 2007 May Surpass GM
https://www.garlic.com/~lynn/2008e.html#65 Banks failing to manage IT risk - study
https://www.garlic.com/~lynn/2009h.html#19 Does anyone know of merchants who have successfully bypassed interchange costs
https://www.garlic.com/~lynn/2009i.html#77 Financial Regulatory Reform - elimination of loophole allowing special purpose institutions outside Bank Holding Company (BHC) oversigh
https://www.garlic.com/~lynn/2009j.html#1 Is it possible to have an alternative payment system without riding on the Card Network platforms?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Thu, 07 Jan 2010 15:12:36 -0500jayarelim@HOTMAIL.COM (J R) writes:
YES CARD vulnerability reference ... basically compromise POS terminal
(or other swipe mechanism to skim the data ... effectively same kind of
exploit used to skim magstripe data) ... and then "trivially" create
counterfeit YES CARD ... original reference gone 404 ... but can be
found at the wayback machine referencing presentation at cartes2002:
https://web.archive.org/web/20030417083810/http://www.smartcard.co.uk/resources/articles/cartes2002.html
about the same time there was presentation on the vulnerabilities at the
ATM integrity task force meetings (prompting somebody in the audience to
comment that they managed to spend billions of dollars to prove that
chips are less secure than magstripe) ... a couple recent posts
with references:
https://www.garlic.com/~lynn/2009q.html#78 70 Years of ATM Innovation
https://www.garlic.com/~lynn/2009r.html#16 70 Years of ATM Innovation
lots of past posts mentioning YES CARD:
https://www.garlic.com/~lynn/subintegrity.html#yescard
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Thu, 07 Jan 2010 15:19:36 -0500Howard Brazee <howard.brazee@cusys.edu> writes:
there was actually a rather large deployment in the NE about the time of the cartes2002 presentation (and the atm integrity task force meetings) ... which then seemed to disappear w/o a trace. There has been some concern expressed about the much larger deployment costs for the US ... but it may actually not so much be about the cost of a single deployment ... but that there may have to be a large number of deployments.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Thu, 07 Jan 2010 16:03:57 -0500Howard Brazee <howard.brazee@cusys.edu> writes:
the compromise of terminal or machine to skim data ... whether magstripe or chip ... was nearly identical. the cost of magstripe cards is several cents less than chipcards used for YES CARDs ... but that is relatively minor compared to the compromise effort to skim&collect the data ... as well as the avg. fraud ROI per counterfeit card.
as referenced in the cartes2002 presentation ... it was trivial to create a counterfeit YES CARD ... and the technology and description was readily available on the internet thru the later half of the 90s.
after having done work with small client/server startup (the startup also had invented this technology called "SSL" that they wanted to use) for payment transactions and what is now commonly called "electronic commerce" ... in the mid-90s we were invited to participate in the x9a10 financial standard working group ... which had been given the requirement to preserve the integrity of the financial infrastructure for all retail payments. The YES CARD kind of exploit was one of the early, easily identifed vulnerabilities by the x9a10 standard working group (long before any kind of actual deployment of that technology)
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: 1964 CTSS film Newsgroups: alt.folklore.computers Date: Thu, 07 Jan 2010 16:10:27 -0500this was just posted over in alt.os.multics
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Thu, 07 Jan 2010 22:20:26 -0500Charles Richmond <frizzle@tx.rr.com> writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: y2k10 problem with credit cards in Germany Newsgroups: bit.listserv.ibm-main Date: 8 Jan 2010 08:21:40 -0800ray.pearce@MACRO4.COM (Ray Pearce) writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Fri, 08 Jan 2010 12:13:07 -0500E99071@JP.IBM.COM (Timothy Sipples) writes:
in states ... there has tended to still be a bunch of stuff done in the
overnight batch window ... some recent posts about doing optimization
work on 450+k statement cobol program that overnight ran on 40+ mainframe fully
tricked-out CECs (>$1B).
https://www.garlic.com/~lynn/2009d.html#5 Why do IBMers think disks are 'Direct Access'?
https://www.garlic.com/~lynn/2009e.html#76 Architectural Diversity
https://www.garlic.com/~lynn/2009f.html#55 Cobol hits 50 and keeps counting
https://www.garlic.com/~lynn/2009g.html#20 IBM forecasts 'new world order' for financial services
https://www.garlic.com/~lynn/2009s.html#9 Union Pacific Railroad ditches its mainframe for SOA
several places in the financial industry spent billions in the 90s on failed straight-through processing efforts (to replace overnight batch window) ... they were planning on using large number of parallel "killer micros" and some COTS libraries. Problem was that they didn't actually size the overhead of the COTS libraries (some vague anticipation that more micros would offset the increased overhead).
it turned out that the COTS libraries had factor of 100 times increase
in overhead (compared to batch COBOL), totally swamping anticipated
thruput improvement with large numbers of killer micros. some past
references to the billions spent on failed straight-through processing
implementation:
https://www.garlic.com/~lynn/2009h.html#1 z/Journal Does it Again
https://www.garlic.com/~lynn/2009h.html#2 z/Journal Does it Again
https://www.garlic.com/~lynn/2009i.html#21 Why are z/OS people reluctant to use z/OS UNIX?
https://www.garlic.com/~lynn/2009l.html#57 IBM halves mainframe Linux engine prices
https://www.garlic.com/~lynn/2009m.html#22 PCI SSC Seeks standard for End to End Encryption?
https://www.garlic.com/~lynn/2009m.html#81 A Faster Way to the Cloud
https://www.garlic.com/~lynn/2009o.html#81 big iron mainframe vs. x86 servers
https://www.garlic.com/~lynn/2009q.html#67 Now is time for banks to replace core system according to Accenture
https://www.garlic.com/~lynn/2009q.html#68 Now is time for banks to replace core system according to Accenture
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: y2k10 problem with credit cards in Germany Newsgroups: bit.listserv.ibm-main,alt.folklore.computers Date: Fri, 08 Jan 2010 12:32:17 -0500carney@ILLUSTRO.COM (Chuck Arney) writes:
i included a much earlier post (from somebody working in Houston
... also mentioned PARs problem and shuttle problem)
https://www.garlic.com/~lynn/99.html#email841207
from 1984 in an internal discussion group (on the corporate internal network) about the looming y2k problems. Above '84 post includes description of 1970 problem when some calculations rolled over 1969 to 196A.
misc. past posts mentioning internal network (larger than the
arpanet/internet from just about the beginning until late '85
or possibly earlay '86).
https://www.garlic.com/~lynn/subnetwork.html#internalnet
a few other references to the 84 post:
https://www.garlic.com/~lynn/2000.html#94 Those who do not learn from history...
https://www.garlic.com/~lynn/2003p.html#21 Sun researchers: Computers do bad math ;)
https://www.garlic.com/~lynn/2006r.html#16 Was FORTRAN buggy?
https://www.garlic.com/~lynn/2006u.html#14 Year-end computer bug could ground Shuttle
https://www.garlic.com/~lynn/2006u.html#35 Friday fun - Discovery on the pad and the software's not done
https://www.garlic.com/~lynn/2009.html#16 Date arithmetic and Zune bug
https://www.garlic.com/~lynn/2009f.html#60 Cobol hits 50 and keeps counting
https://www.garlic.com/~lynn/2009i.html#27 My Vintage Dream PC
https://www.garlic.com/~lynn/2009n.html#53 Long parms...again
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Fri, 08 Jan 2010 13:07:50 -0500Ahem A Rivet's Shot <steveo@eircom.net> writes:
list of some unixes
http://www.levenez.com/unix/
Micro implementations of Unix
http://www.robotwisdom.com/linux/nonnix.html
misc. other ...
http://www.vintage-computer.com/vcforum/showthread.php?11151-PC-IX-Images
http://www.softpanorama.org/People/Torvalds/Finland_period/xenix_microsoft_shortlived_love_affair_with_unix.shtml
http://linuxfinances.info/info/unixlist.html
long ago and far away ... there used to be conferences in silicon valley ... where there would be unannounced products and people from different companies would play with each others toys; with increasing commercialization ... that has gone by the wayside.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Fri, 08 Jan 2010 17:56:59 -0500"John Varela" <OLDlamps@verizon.net> writes:
... 30s in eastman georgia.
has pointer to ...
http://www.stuckeys.com/history.php
http://www.stuckeys.com/history_detailed.php
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Fri, 08 Jan 2010 22:21:38 -0500re:
bill moyers journal going on in real time ... interviewing mother jones
journalists
http://www.pbs.org/moyers/journal/01082010/watch.html
opened the show with statement that goldman sachs paid less than 1% effective tax rate for 2009 ... and then some comments about the journalists visiting several locations ... including visiting the goldman sachs branch office in washington DC, "the us treasury".
some more reference ...
http://www.pbs.org/moyers/journal/01082010/profile.html
http://www.pbs.org/moyers/journal/finance/index.html
http://www.pbs.org/moyers/journal/community/index.html
http://www.pbs.org/moyers/journal/01082010/campaignfinance.html
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sat, 09 Jan 2010 09:40:15 -0500Peter Flass <Peter_Flass@Yahoo.com> writes:
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sat, 09 Jan 2010 10:17:57 -0500re:
global warming & more energy in the air predicts more water evaporation ... when higher humidity air hits the cold front from the artic express ... it would tend to produce higher snow fall levels.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sat, 09 Jan 2010 11:28:56 -0500jmfbahciv <jmfbahciv@aol> writes:
there was still a lot of reasons why a virtual page might have to reside below the 16mbyte line. originally, the proposal to bring a page "above the line" to "below the line", was to write it out to disk (it was possible to use an i/o address field that was full-word ... originally designed for 24bit real storage address ... but now useable for >16mbyte addresses).
I provided them with a hack that used two dummied up page table entries,
with the from & to page numbers and then did a 4k move instruction. It
could be used to bring virtual pages down below the line and also to
push/shuffle other pages above the line (to make room). old email
discussing the hack (and avoid the operation to disk)
https://www.garlic.com/~lynn/2006t.html#email800121
in this post
https://www.garlic.com/~lynn/2006t.html#15 more than 16mbyte support for 370
misc. other posts discussing the hack:
https://www.garlic.com/~lynn/2000d.html#82 "all-out" vs less aggressive designs (was: Re: 36 to 32 bit transition)
https://www.garlic.com/~lynn/2003f.html#4 Alpha performance, why?
https://www.garlic.com/~lynn/2004f.html#38 Infiniband - practicalities for small clusters
https://www.garlic.com/~lynn/2005.html#34 increasing addressable memory via paged memory?
https://www.garlic.com/~lynn/2005p.html#19 address space
https://www.garlic.com/~lynn/2005q.html#30 HASP/ASP JES/JES2/JES3
https://www.garlic.com/~lynn/2005u.html#44 POWER6 on zSeries?
https://www.garlic.com/~lynn/2006l.html#2 virtual memory
https://www.garlic.com/~lynn/2006w.html#23 Multiple mappings
https://www.garlic.com/~lynn/2007b.html#34 Just another example of mainframe costs
https://www.garlic.com/~lynn/2008f.html#12 Fantasy-Land_Hierarchal_NUMA_Memory-Model_on_Vertical
https://www.garlic.com/~lynn/2009g.html#71 308x Processors - was "Mainframe articles"
https://www.garlic.com/~lynn/2009g.html#74 308x Processors - was "Mainframe articles"
the original design was just pulling pages below the line and descarding (as necessary) other pages to make room (as opposed to pushing them above the line) ... which really messes up the page replacement algorithm.
minor email from long ago & far away discussing just one simple example
how one of the proposals would unnatrually skew page replacement (in
a different way).
Date: 08/21/81 12:22:01
From: wheeler
co-incidence? i was out until tuesday of last week . . . just was
getting around to reading document in detail this morning . . . my
first take was a question about why & how are you going to move page
to <16meg??? Are you going to use the VFAULT flag as indication that
request is coming from PRG fault handling as opposed to other places
in CP??? If you in fact alwas attempt to satisfy page fault request
from >16meg. then you will "skew" real storage utilization
i.e.:
assume that 90% of page fetches are because of page faults and only
10% are because of cp fetches. Complete partioning of real storage for
page faults will result in the storage >16meg to be "scanned" 9 times
more often compared to <16meg & their reference bits will be reset 9
times more often. Easy to show that resulting effect could be that (at
least) 90% of the paging is done from >16meg, over utilizing that area
compared to <16meg. Actual effectiviness of the <16meg. area would be
as bad as some of the other simpler proposals for utilizing >16 meg.
area.
... snip ... top of post, old email index.
the issue was to ordinarily treat above & below the line areas as equivalent ... then there would be some extra pressure (on below line area) for pages that had to be brought below the line ... which then had to be compensated for (including possibly pushing replaced pages to above the line instead of discarding).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sat, 09 Jan 2010 11:49:23 -0500greymaus <greymausg@mail.com> writes:
larger amount of snow in more temperate regions ... interaction between stronger artic winds (possibly from global warming) and warmer air from the south with more moisture (also from global warming). overall colder would tend to mean less snow ... since there should be less moisture in the air.
corresponding to more snow in temperate regions ... is that the warmer artic regions is increasing melting on surface of the ice. melted water on the surface correspondingly decreases the whiteness/reflectivity ... increasing the heat absorbtion and melting rate.
few weeks of additional snow cover (and higher reflection) in temperate regions doesn't seem to offsetting the lower reflection in the artic. Part of the issue is snow cover (& higher reflection) in temperate regions correspond to normally shorter (as well as overcast) days & shorter amounts of sun (aka change in higher reflection rate times normal number of sun hrs) while artic melt is during longer days and longer amounts of sun (change in lower reflection rate times normal number of sun hrs).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: locate mode, was Happy DEC-10 Day Newsgroups: alt.sys.pdp10,alt.folklore.computers Date: Sat, 09 Jan 2010 12:22:42 -0500re:
This sort of discusses a scenario for >16m similar to later 3090
"expanded store". In the 3090 scenario ... physical packaging put some
amount of electronic store far enough away that affected instruction
latency. So instead of treating longer latency like normal memory,
they put it on a wide/fast special bus and used software to shuffle
pages back and forth between regular memory and "expanded store"
(logically an electronic paging device but using synchronous move
instruction instead of asynchronous i/o) The bus was optimized for 4k
move.
Date: 09/27/81 16:11:51
From: wheeler
oh yes, & in conjunction with the I/O question. Even tho TRANS
defer/lock/etc. works, it means all pages involved in I/O have to be
brought into the <16meg area. For normal CMS type file I/O that isn't
too bad, but all CMS modules & MVS virtual paging is also done via
I/O. That would imply nearly everyone's virtual memory has to be
intialized via activity in <16meg. In total that would imply very few
pages going to >16meg. area (at least initially). The load on the
system would have to be great enuf that the pages are stolen and then
used again before they would have a chance of getting into >16meg.
This is an area (virtual I/O) where it would probably be necessary
that pages be moved from the <16meg area to the >16meg area in order
that you get some sort of utilization. Suggestion that all pages get
paged into the <16meg area and then when replaced get moved to the
>16meg area is probably as good a scheme for handling that situation
as any.
Problem with that algorithm is the implication that the pages in the
>16meg area have lower utilization that pages in the <16meg area,
Therefor any page in the <16meg area can cause the replacement of any
page in the >16meg area. By implication that means that all pages in
the <16meg area have preferrence over all pages in the >16meg area.
Now pages may get moved to the >16meg area because they weren't used
for awhile, but what happens when a page starts getting used again
after it has been moved??? The page replacement algorithm is
supposedly an attempt to simulate LRU. Again I reference the problem
of attempting to judge between the relative priority of two items when
they are maintained on two different lists (like E1 & E2). Each
individual list is ordered by LRU criteria but there is no mechanism
which judges the relative LRU criteria between two items (pages) on
different lists. It is possible that the page in the <16meg area has
been used further in the past than every page in the >16meg area.
A possible solution is that when pages are moved to the >16meg area,
you invalidate them. Then when a page is moved from the <16meg area to
the >16meg area, it is only allowed to replace invalid pages. Ok, now
what happens when a page in the >16meg area is referenced? Is it
brought to the <16meg area or is it just validated in place???? If it
is just validated in place, there is no mechanism to replace it (i.e.
only time a page from the >16meg area is replaced is if it is not
valid & being replaced by a page from the <16meg area). Therefor it
must be moved from the >16meg area to the <16meg area for the
algorithm to work (otherwise you either don't replace it or get into
the same situation as you were in if you weren't going thru the
invalidation/validation cycle). In this situation the >16meg area just
becomes a fancy paging device.
Suggest you reconsider using the 20 line PSA subroutine.
... snip ... top of post, old email index.
the "20 line PSA subroutine" refers to much earlier instruction based
proposal (as alternative to moving above/below line by paging i/o
to/from disk).
https://www.garlic.com/~lynn/2006t.html#email800121
in this post
https://www.garlic.com/~lynn/2006t.html#15 more than 16mbyte support for 370
for other drift about 3090 "expanded store" ... when there was effort to add HIPPI to 3090 (HIPPI was 100mbyte/sec copper parallel standardization version from cray channel) ... standard 3090 I/O didn't have the bandwidth. What they did, was cut into the side of the expanded store bus ... however, since the processor only had 4k move instructions to interface to the bus ... a "peek/poke" interface was created to program the HIPPI I/O (i.e. peek/pok operations were done by executing "expanded store" 4k move instructions ... to reserved addresses).
past posts mentioning 3090 "expanded store" (and shuffling
pages to/from "expanded store").
https://www.garlic.com/~lynn/95.html#15 multilevel store
https://www.garlic.com/~lynn/2001m.html#25 ESCON Data Transfer Rate
https://www.garlic.com/~lynn/2001m.html#53 TSS/360
https://www.garlic.com/~lynn/2002c.html#53 VAX, M68K complex instructions (was Re: Did Intel Bite Off More Than It Can Chew?)
https://www.garlic.com/~lynn/2002e.html#26 Crazy idea: has it been done?
https://www.garlic.com/~lynn/2002e.html#32 What goes into a 3090?
https://www.garlic.com/~lynn/2003j.html#2 Fix the shuttle or fly it unmanned
https://www.garlic.com/~lynn/2003p.html#41 comp.arch classic: the 10-bit byte
https://www.garlic.com/~lynn/2003p.html#46 comp.arch classic: the 10-bit byte
https://www.garlic.com/~lynn/2005.html#17 Amusing acronym
https://www.garlic.com/~lynn/2005h.html#13 Today's mainframe--anything to new?
https://www.garlic.com/~lynn/2005j.html#13 Performance and Capacity Planning
https://www.garlic.com/~lynn/2006l.html#43 One or two CPUs - the pros & cons
https://www.garlic.com/~lynn/2006r.html#35 REAL memory column in SDSF
https://www.garlic.com/~lynn/2006r.html#36 REAL memory column in SDSF
https://www.garlic.com/~lynn/2006s.html#16 memory, 360 lcs, 3090 expanded store, etc
https://www.garlic.com/~lynn/2006s.html#17 bandwidth of a swallow (was: Real core)
https://www.garlic.com/~lynn/2007c.html#23 How many 36-bit Unix ports in the old days?
https://www.garlic.com/~lynn/2007f.html#18 What to do with extra storage on new z9
https://www.garlic.com/~lynn/2007g.html#67 Unusual Floating-Point Format Remembered?
https://www.garlic.com/~lynn/2007s.html#9 Poster of computer hardware events?
https://www.garlic.com/~lynn/2008f.html#6 Fantasy-Land_Hierarchal_NUMA_Memory-Model_on_Vertical
https://www.garlic.com/~lynn/2008f.html#8 Fantasy-Land_Hierarchal_NUMA_Memory-Model_on_Vertical
https://www.garlic.com/~lynn/2008i.html#10 Different Implementations of VLIW
https://www.garlic.com/~lynn/2008i.html#11 Different Implementations of VLIW
https://www.garlic.com/~lynn/2009g.html#9 Architectural Diversity
https://www.garlic.com/~lynn/2009i.html#39 My Vintage Dream PC
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sat, 09 Jan 2010 15:30:31 -0500Peter Flass <Peter_Flass@Yahoo.com> writes:
more evaporation, more moisture in the air ... the higher moisture content hitting cold front from the "artic express" results in more snow.
more energy in the atmosphere would tend to contribute to more weather of various kinds.
re:
https://www.garlic.com/~lynn/2010.html#82 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#83 360 programs on a z/10
https://www.garlic.com/~lynn/2010.html#85 360 programs on a z/10
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sun, 10 Jan 2010 11:39:00 -0500weather station last night had the "artic express" dipping far south and then across the gulf states ... but said it would be moving back north to pattern blowing across the great lakes.
search engine for "artic express"
a reference to "artic express", "siberian express" (& "pineapple
express")
http://weblogs.wgntv.com/chicago-weather/tom-skilling-blog/2009/01/arctic-express-vs-siberian-exp.html
a few random other:
http://blog.memphisweather.net/2010/01/arctic-express-headed-to-mid-south.html
http://www.thedenverchannel.com/weather/22130027/detail.html
http://weatherblog.dallasnews.com/archives/2010/01/arctic-express-will-deliver-sub-freezing.html
http://www.katv.com/blogs/wx/arctic_express_rolls_into_arkansas_to_open_2010.html
http://weblogs.marylandweather.com/2009/01/arctic_express_headed_this_way.html
some of the heaviest snowfall in the midwest seems to be when there is strong, warm, heavy moisture wind pattern blowing northwest from gulf ... sliding against a southwest "artic express" cold front. this particular gulf air pattern, when it hits the rockies, can rotate counter clockwise and then blow south along the eastern slope of the rockies, dumping large amounts of snow along the whole path.
global warming pumping more energy into the atmosphere can account for increased/changed wind patterns ...
Wind Patterns Could Mask Effects Of Global Warming In Ocean
http://www.sciencedaily.com/releases/2008/02/080207101333.htm
http://www.physorg.com/news121601825.html
others
http://www.liv.ac.uk/researchintelligence/issue34/warming.htm
http://www.sciencecentric.com/news/article.php?q=08020746
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sun, 10 Jan 2010 13:03:03 -0500for the fun of it ... at least one local condition contra-indicator ... period of 1998-2009 for apple blossom "degree-days"
theoretically, with general global warming ... degree-day accumulation should result in full-bloom occurring earlier in the season.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: How long before Microsoft goes the way of DEC (and in part, IBM)? Newsgroups: alt.folklore.computers Date: Sun, 10 Jan 2010 15:18:09 -0500Michael Wojcik <mwojcik@newsguy.com> writes:
besides all the legal stuff, including novell claiming it has the rights
to unix ... not SCO ... and that anything SCO may have collected
... actually belongs to Novell.
https://en.wikipedia.org/wiki/SCO_v._Novell
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: 360 programs on a z/10 Newsgroups: alt.folklore.computers Date: Sun, 10 Jan 2010 15:21:58 -0500traditional articles are that winds tend to be west to east and that west coasts of continents tend to have more modereated weather because of a combination of ocean effects and the west to east winds. typical examples are seattle and british isles ... tending to have more temperate weather than can be expected by their latitude (ocean gulf stream also cited for british isles). when the effects let up &/or change ... there is surprise at the colder &/or hotter than normal temperature (prediction that small changes in atlantic gulf stream could result in significantly colder winters and hotter summers in british isles).
the articles on global warming putting more energy in the air and stronger winds ... have scenario where the coastal effects may extended further inland ... as the strength of wind goes up.
a pre-global warming localized scenario is the air conditioning effect on san francisco. heated air in south valley creates a ground level, low pressure (hotter the air, more air is rising and the stronger the ground level low pressure) ... and because of the hills on both sides of the valley ... the low pressure sunction effect extends north to break in the hills at golden gate bridge ... causing huge amounts of cooler ocean air to be sucked in. The hotter the air in south valley, the stronger the sunction effect pulling in cooler ocean air (helping moderate san fran weather). In changing weather conditions ... it sometimes can take a couple days for the air conditioning effect to go into operation ... and san fran will experience temperatures comparable to the rest of the region.
ocean west to east coastal winds also have "rain shadow" effect ...
moist ocean air hits range of cooler hills/mountains ... dumping lots of
the moisture ... but it also results in semi-arid/arid "rain shadow" on
the back side. This can be seen in san jose (on back side of santa cruz
mountains), neveda (on back side of sierras), columbia basin (on back
side of cascades) ... and normally very moderate snow fall in denver (on
back side of rockies). Heavy Denver snow fall frequently is the humid,
moist gulf counter rotation ... that dumps large amounts of snow along
the eastern slopes of the rockies.
https://www.garlic.com/~lynn/2010.html#88 360 programs on a z/10
past post mentioning denver, "rain shadow", and counter rotating gulf
pattern
https://www.garlic.com/~lynn/2006y.html#2 The Future of CPUs: What's After Multi-Core?
in any case, one scenario for the apple blossom degree days is possible
stronger ocean air effects ... moderating highs during the spring (as
opposed to overall planet cooling).
https://www.garlic.com/~lynn/2010.html#89 360 programs on a z/10
past post mentioning the golden gate effect and global warming pumping
more energy into the atmosphere
https://www.garlic.com/~lynn/2008s.html#46 Globa Warning. was Re: : Is SUN going to become x86'ed ??
misc. other past posts mentioning break in hills at golden gate
moderating san francisco temperatures:
https://www.garlic.com/~lynn/2001b.html#68 weather biasing where engineers live (was Re: Disk power numbers)
https://www.garlic.com/~lynn/2001i.html#0 YKYGOW...
https://www.garlic.com/~lynn/2002o.html#11 Home mainframes
https://www.garlic.com/~lynn/2006.html#24 IBM up for grabs?
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Larrabee delayed: anyone know what's happening? Newsgroups: comp.arch Date: Sun, 10 Jan 2010 15:56:51 -0500Kai Harrekilde-Petersen <khp@harrekilde.dk> writes:
a couple more items from long ago and far away (following also
includes Dolphin/Convex/Exemplar announcement):
Date: January 23, 1991
Subject: DOLPHIN SERVER ABANDONS ECL 88000 RISC PLAN
Computergram - Norsk Data A/S affiliate Dolphin Server Technology
A/S is re-focussing its effort to build a 1,000 MIPS, multi-processing
server based on an ECL version of the Motorola 88000 RISC chip. The
project, known as Orion, was originally slated for completion late in
1992, but the ECL CPU development effort has run into trouble - similar
problems have already bedevilled other RISC projects, most recently MIPS
Computer Systems' R6000 ECL part. Dolphin and Motorola collaborated on
the design of the ECL part, and National Semiconductor was to fabricate
it. Central to Dolphin's long- term plan has been the use of SCI, the
Scalable Coherent Interface bus architecture, which is similar to the
Futurebus+ system in concept, but has attracted a good deal less
attention. Dolphin already has the 88000-based Triton 88 server under
its belt, and now plans to introduce an interim Triton SCI system early
in 1992. It will be a 300 MIPS multi-processor system combining SCI,
cache and memory components from the Orion, with Motorola's much
previewed 88110 RISC chip. The Orion - now not expected until 1993 -
will use Motorola's post-88110 100MHz BiCMOS technology, rather than the
Dolphin-designed CPU. Dolphin, which says it has peeped behind the
curtain and seen what Motorola is up to, has both feet firmly in the
Motorola camp, and expects single-chip, multi-processors with 100m
transistors clocking at 300MHz from the firm by the late 1990s, with a
4,000 MIPS part by the year 2000. Dolphin is awaiting final ratification
of an SCI standard from the IEEE - expected later this year - and will
then go straight into production of the Triton SCI. Dolphin is
implementing SCI in a Token Ring-like formation, which it claims, offers
up to five times the throughput of Futurebus+. On the Triton SCI Dolphin
will offer bridges to VME-based systems, to other types of SCI systems,
and may also develop links to Futurebus+. Enhancements planned for the
Triton 88 this year include the addition of Unix System V.4, Novell and
Banyan Vines networking support, increased storage options and a new
plug-in CPU board with up to five 88000 processors. Following its OEM
deal with Thomson-CSF SA subsidiary Cetia SA, Dolphin says it is now
finalising a European distribution channel, and will also make a UK
announcement soon. Dolphin claims an installed base of 225 Triton 88s.
... snip ... top of post, old email index.
above mentions SCI in T/R-like ... offers five times the throughput of Futurebus+
following very long & heavily "snipped" ... totally unrelated, SLACVM
was the original webserver outside of CERN:
https://ahro.slac.stanford.edu/wwwslac-exhibit
Date: Tue, 23 Jun 1992 22:11 -0800 (PST)
From: DBG@SLACVM.SLAC.Stanford.EDU
Subject: Some online SCI documents
To: wheeler
Current status: the base standard is now approved by the IEEE as
IEEE Std 1596-1992. It originally went out for official ballot in
late January 91. Voters approved it by a 92% affirmative vote that
ended April 15. Final corrections and polishing were done, and
the revised draft was recirculated to the voters again and passed.
Draft 2.00 was approved by the IEEE Standards board on
18 March 1992. Pre-publication copies of the standard are
available from the IEEE Service Center, Piscataway, NJ, (800)678-4333.
Commercial products to support and use SCI are already in final design
and simulation, so the support chips should be available soon, 3Q92.
-
SCI-related documents are available electronically via anonymous FTP
from HPLSCI.HPL.HP.COM, except for a few documents which are paper only.
Online formats are Macintosh Word 4 (Compacted,self expg) and PostScript.
The PostScript includes Unix compressed and uncompressed forms.
Paper documents can be ordered from Kinko's 24hr copy Service, Palo Alto,
California, (415)328-3381. Various payment forms can be arranged.
Newcomers should order the latest mailing plus the package NEW, which
contains the most essential documents from previous mailings.
SCI depends on the IEEE 1212 CSR Architecture as well, so you will
also need a copy of that, which is available from the IEEE Service Ctr.
-
Send your name, mailing address, phone number, fax number, email
address, to me and I will put you on a list of people to be notified
when new mailings are available; you will also be listed in an
occasional directory of people who are participating in or observing
SCI development.
-
Contact:
-
David B. Gustavson
IEEE P1596 Chairman
Stanford Linear Accelerator Center
Computation Research Group
P.O.Box 4349, Bin 88
Stanford, CA 94309
415-926-2863 or dbg@slacvm.slac.stanford.edu
An SCI Extensions Study Group has been formed to consider what SCI-related
extensions to pursue and how to organize them into standards.
Related standards projects:
1212: Control and Status Register Architecture. This specification
defines the I/O architecture for SCI, Futurebus+ (896.x) and SerialBus
(P1394). Chaired by David V. James, Apple Computer, dvj@apple.com,
408-974-1321, fax 408-974-0781. An approved standard as of December
1991. Being published by the IEEE.
P1596.1: SCI/VME Bridge. This specification defines a bridge
architecture for interfacing VME buses to an SCI node. This will provide
early I/O support for SCI systems via VME. Products are likely to be
available in 1992. Chaired by Bjorn Solberg, CERN, CH-1211 Geneva 23,
Switzerland. bsolberg@dsy-srv3.cern.ch, ++41-22-767-2677, fax
++41-22-782-1820.
P1596.2: Cache Optimizations for Large Numbers of Processors using the
Scalable Coherent Interface. Develop request combining, tree-structured
coherence directories and fast data distribution mechanisms that may be
important for systems with thousands of processors, compatible with the
base SCI coherence mechanism. Chaired by Ross Johnson, U of Wisconsin,
ross@cs.wisc.edu, 608-262-6617, fax 608-262-9777.
P1596.3: Low-Voltage Differential Interface for the Scalable Coherent
Interface. Specify low-voltage (less than 1 volt) differential signals
suitable for high speed communication between CMOS, GaAs and BiCMOS
logic arrays used to implement SCI. The object is to enable low-cost
CMOS chips to be used for SCI implementations in workstations and PCs,
at speeds of at least 200 MBytes/sec. This work seems to have converged
on a signal swing of 0.25 V centered on +1 V. Chairman is Stephen
Kempainen,National Semiconductor, 408-721-2836, fax 408-721-7218.
asdksc@tevm2.nsc.com
P1596.4: High-Bandwidth Memory Interface, based on SCI Signalling
Technology. Define a high-bandwidth interface that will permit access to
the large internal bandwidth already available in dynamic memory chips.
The goal is to increase the performance and reduce the complexity of
memory systems by using a subset of the SCI protocols. Started by Hans
Wiggers of Hewlett Packard, current chairman is David Gustavson,
Stanford Linear Accelerator Center, 415-961-3539, fax 415-961-3530.
P1596.5: Data Transfer Formats Optimized for SCI. This working group has
defined a set of data types and formats that will work efficiently on
SCI for transferring data among heterogeneous processors in a
multiprocessor SCI system. The working group has finished, voting to
send the draft out for sponsor ballot. Chairman is David V. James, Apple
Computer, dvj@apple.com, 408-974-1321, fax 408-974-0781.
CONVEX SELECTS DOLPHIN'S SCI INTERCONNECT TECHNOLOGY FOR USE IN FUTURE
PROCESSORS.
According to a technology transfer agreement announced today, Dolphin
SCI Technology A.S, a subsidiary of Dolphin Server Technology A.S, will
share its Scalable Coherent Interface (SCI) technology with Convex
Computer Corporation for use in Convex' future generation supercomputers
currently under development. The agreement is initially worth several
hundred thousands USD to Dolphin, and includes intentions of future
cooperation between the two companies.
Convex is continously working to develop new machines to strengthen the
company's supercomputer market position. Convex manufactures systems
which solve many of today's most demanding applications such as climate
modelling, genetic sequencing and computational fluid dynamics. Convex
recently announced a relationship with Hewlett-Packard which will result
in Convex's adoption of HP's PA-RISC processor technology to build these
future high performance machines.
The Scalable Coherent Interface is an enabling technology for
multiprocessor systems. With the current rapidly increasing RISC
microprocessor power, even the best of today's interconnect - or "bus" -
systems can only support small multiprocessor configurations. Buses are
inherently bottlenecks, because only one processor "talks" at a time,
and clock rates are limited by the physics of tapped transmission lines
with variable loading. Buses also scale poorly with system size because
propagation delays limit handshake and arbitration speed.
... snip ... top of post, old email index.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main Date: Mon, 11 Jan 2010 10:53:55 -0500Charles.Hardee@CA.COM (Hardee, Charles H) writes:
as previously mentioned, the YES CARD scenario for chipcard resulted in bigger infrastructure vulnerability and more fraud than traditional magstripe.
supposedly the chipcard was hard to counterfeit *AND*
had two-factor authentication (chip/plastic: something you
have and PIN: somthing you know). from three factor
authentication model, misc. posts
https://www.garlic.com/~lynn/subintegrity.html#3factor
• something you have • something you know • something you are
the assumption that multiple factor authentication is more secure than single factor is based on different authentication factors having different vulnerabilities.
the problem with skimming (whether for the YES CARD or magstripe) ... is it is possible to have a single compromise process (end-point skimming compromise) ... invalidating the assumption about different factors having different vulnerabilities. In the case of multi-factor authentication magstripe (plastic/magstripe & PIN) ... a compromised end-point skims both the magstripe information and the PIN (a common vulnerability)
in the YES CARD scenario, a compromised end-point skims the information used by terminals to establish a valid chipcard. the crooks then install the skimmed information (similar to information skimmed for counterfeit magstripe) in a counterfeit YES CARD chip.
once a terminal has accepted the chipcard's validation information, it then asks the chipcard 1) whether the correct PIN has been entered (a YES CARD always answers YES ... so it isn't necessary to even know/skim the PIN), 2) whether the transaction should be offline (YES), and 3) whether the transaction is within the account credit limit (YES).
in counterfeit magstripe scenario, the account number is eventually invalidated at the backend database (and future transactions are rejected). In the counterfeit YES CARD scenario, the terminal doesn't go online to find out about any account number invalidation. the greater counterfeit YES CARD fraud is because infrastructure business rules have been moved into the chipcard (infrastructure relying on the chipcard to decide whether it is online/offline transaction and whether the transaction is within the account's credit limit).
misc. past YES CARD posts
https://www.garlic.com/~lynn/subintegrity.html#yescard
one of the issues with something you are biometrics ... is that nominally biometrics information is reduced to some sort of electronic pattern for matching against value stored in backend database. If that value is compromised (analogous to something you know PIN/passwords) ... it is difficult to issue a new finger or iris. Frequently biometrics are most dependable ... when they involve secure sensors/endpoints ... that possibly are under constant surveillance by armed guards.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Daylight Savings Time again Newsgroups: alt.folklore.computers Date: Mon, 11 Jan 2010 11:27:48 -0500Peter Flass <Peter_Flass@Yahoo.com> writes:
misc. other past posts mentioning boyd
https://www.garlic.com/~lynn/subboyd.html#boyd
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: Anne & Lynn Wheeler <lynn@garlic.com> Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main Date: Mon, 11 Jan 2010 14:53:21 -0500Howard Brazee <howard.brazee@cusys.edu> writes:
the issue with pin/passwords aren't that they are something you know
authentication ... but they are shared-secrets ... some past posts
https://www.garlic.com/~lynn/subintegrity.html#secrets
the issue is that a unique shared-secret is required for every unique security domain ... as countermeasure to cross-domain attacks (say local garage ISP and some online banking).
in YES CARD scenario ... the PIN wasn't a shared-secret ... but was between you and "your" chipcard. the problem was that the chipcard had the YES CARD vulnerability ... and so the whole infrastructure wasn't very secure.
it is possible to have a something you know authentication ... w/o requiring what-ever is used ... is not "shared". In the "non-sharing" scenario ... it would be acceptable to have the same (non-shared) something you know authentication used in multiple different security domains.
something you are, biometric authentication is a problem in the online scenario ... since it can be difficult to assure secure/trusted sensor/end-point (under constant surveillance by trusted, armed guards)
part of the issue is that biometric (electronic pattern recorded in backend database) is also frequently implemented as shared-secret. If all biometric sensors/end-points aren't constantly secured & validated ... then the recording of the biometric electronic pattern could be used to spoof a biometric reading ... by just directly transmitting the pattern. In the case of a password shared-secret compromise ... the password can be replaced with new one ... fingers and iris are a little harder to replace.
for a little more drift ... because of the cross-domain attack scenario, for shared-secrets ... current authentication is extremely institutional-centric (unique cards & passwords per security domain). In theory, a biometric shared-secret implementation would require unique biometric per security domain ... modulo nobody has quite figured out how to implement such a thing. As a result, compensating procedures are required for biometric shared-secrets ... like secure/trusted sensors/end-points under constant surveillance by armed guards.
it is possible to design a single something you have (like a chip) and something you know authentication ... used in multiple different domains ... analogous to the way that same fingerprint should work in multiple different domains. part of the inhibitor to moving from institutional-centric authentication to person-centric authentication ... is when things like institutional-specific business rules are layered ontop of the authentication mechanism (like in the YES CARD vulnerability).
In the 90s, I did quite a bit of work on AADS chip strawman for
enabling migration to a person-centric authentication infrastructure
(not limited just to biometrics)
https://www.garlic.com/~lynn/x959.html#aads
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main Date: 11 Jan 2010 12:00:02 -0800Charles.Hardee@CA.COM (Hardee, Charles H) writes:
the signature isn't a fraud countermeasure ... it is a dispute issue. if you dispute the charge and the merchant doesn't even have signed receipt ... there is nothing demonstrating that you agreed to the charge.
for some low-value purchases, they've eliminated the signature requirement ... the issue is that there aren't going to be a lot of crooked consumers disputing low value charges ... and if they do ... it is trivial amount (convenience offset against crooked consumers). the infrastructure countermeasure against crooked consumers disputing large number of (unsigned) charges ... is they revoke the card.
fraud countermeasure is the name on the piece of plastic and the clerk checks the name against same/similar name on some other piece of authentication (like gov. issued picture document).
there was an issue in the EU at one time regarding a privacy directive ... where electronic payment cards should be as anonymous as cash at point of sale (i.e. no name on the payment card). this somewhat implied that the financial infrastructure improved the authentication mechanisms to the point that anti-fraud measures didn't require clerk matching names on multiple documents.
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main Date: 11 Jan 2010 14:22:53 -0800phil@VOLTAGE.COM (Phil Smith) writes:
chipcards have countermeasures for some random person taking a valid chip and extracting the necessary information ... a random person can copy magstripe information significantly easier.
however, by at least the early 90s, there were cases of compromised end-points recording valid information (done during the process of valid transactions). these operations tended to be more large scale wholesale operations ... getting information for tens of thousand (or millions) ... rather than a few tens.
in the end-point compromises ... the process was esssentially identical for recording magstripe information and recording chipcard authentication information (for YES CARD exploit).
along the way, the criminals added wireless and other remote procedures for retrieving the skimmed/recorded information (again, little or no difference between magstripe and chipcard).
part of the issue in the US was that there was fairly large scale chipcard deployment in the time-frame of cartes2002 (presentation on YES CARD and the YES CARD presentations at the ATM integrity task force meetings) ... and then evaporated w/o a trace (which may have also created some reluctance to try again).
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970
From: lynn@GARLIC.COM (Anne & Lynn Wheeler) Subject: Re: Korean bank Moves back to Mainframes (...no, not back) Newsgroups: bit.listserv.ibm-main Date: 11 Jan 2010 15:19:08 -0800tony@HARMINC.NET (Tony Harminc) writes:
for the most part ... merchant associations don't like the idea of clerks having to be involved in the authentication process ... partly because they have little or no training ... partly because they have little or no authority ... and partly because clerks tend to already have more than enough to deal with.
in general, merchants also don't like signature debit ... since the interchange fees (merchant discount fees, the subtracted from the total for actual paying to the merchant) are much higher
there have been various disputes about the whole signature debit operation ... latest is:
Best Buy Cuts off Visa Contactless with Little Risk to Sales
http://www.digitaltransactions.net/newsstory.cfm?newsid=2418
above mentions problem with it being signature debit interchange fees. somewhat older article ...
Study: Signature Debit Fraud Runs 15 Times Higher Than on PIN Debit
http://www.digitaltransactions.net/newsstory.cfm?newsid=738
part of the interchange fee is supposedly related to fraud level of the corresponding kind of transaction ... and there can be more than an order-of-magnitude difference (in interchange fee) between the transactions with lowest fraud and transactions with highest fraud.
Past merchant class action lawsuit (sometimes referred to as the "Wal-Mart" case) over the high cost of signature debit cards:
MasterCard Puts the 13-Year-Old Wal-Mart Case in the Rear-View Mirror
http://www.digitaltransactions.net/newsstory.cfm?newsid=2256
--
40+yrs virtualization experience (since Jan68), online at home since Mar1970