| Author |
Message |
Anne & Lynn Wheeler
Guest
|
 Posted:
Fri Aug 12, 2005 10:24 pm Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
prep@prep.synonet.com writes:
| Quote: | Clusters had been around for over 10 years then. Both Vaxen with VMS
and Tops-20 CFS.
|
clusters date back to the 60s & 70s.
i worked on one of the largest clusters in the late 70s at the HONE
system (which provided online infrastructure for all the US sales,
marketing, and field people ... and was also cloned at a number
of places around the world providing worldwide support for sales
and marketing)
http://www.garlic.com/~lynn/subtopic.html#hone
but there were lots of others ... the airline control program for the
airline res systems starting in the 60s and in the same timeframe,
there were the also custom modified 360s for FAA air traffic control
system.
i remember visiting nasa/houston in the late 60s (i think part of a
share meeting held in houston spring of 68). they had five 360/75s in
some sort of cluster providing ground support for missions.
--
Anne & Lynn Wheeler | http://www.garlic.com/~lynn/ |
|
| Back to top |
|
 |
Dan Koren
Guest
|
| Posted:
Sat Aug 13, 2005 7:29 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message
news:ddi7on$kic$1@gemini.csx.cam.ac.uk...
| Quote: |
In article <3m3krfF158slqU1@individual.net>,
Del Cecchi <cecchinospam@us.ibm.com> writes:
|
|> >>NUMA and clusters. That is about the time frame they became
commercial.
|> >>Sysplex. etc.
|
|> > Clusters had been around for over 10 years then. Both Vaxen with VMS
|> > and Tops-20 CFS.
|
|> OK, just NUMA then. And give or take a few years.
Well, only if you disregard companies like
Kendall Square Research as non-commercial.
|
How many systems did KSR manage
to sell before they went under?
dk |
|
| Back to top |
|
|
George Coulouris
Guest
|
| Posted:
Sat Aug 13, 2005 8:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
On 2005-08-13, Dan Koren <dankoren@yahoo.com> wrote:
| Quote: |
How many systems did KSR manage
to sell before they went under?
dk
|
At least four :)
http://www.tc.cornell.edu/CTC-Main/Services/Education/Topics/Parallel/
Concepts/_jointprojunix.htm
mind the wrap.
--
George Coulouris
not speaking for ncbi
remove s to reply |
|
| Back to top |
|
|
Nick Maclaren
Guest
|
| Posted:
Sat Aug 13, 2005 1:45 pm Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
In article <42fd5b00$1@news.meer.net>, Dan Koren <dankoren@yahoo.com> wrote:
| Quote: | |
|> >>NUMA and clusters. That is about the time frame they became
commercial.
|> >>Sysplex. etc.
|
|> > Clusters had been around for over 10 years then. Both Vaxen with VMS
|> > and Tops-20 CFS.
|
|> OK, just NUMA then. And give or take a few years.
Well, only if you disregard companies like
Kendall Square Research as non-commercial.
How many systems did KSR manage
to sell before they went under?
|
That is the point. KSR (like Tera) can be fairly regarded as either
commercial or non-commercial; the new Cray has only just broken into
the normal commercial market.
Regards,
Nick Maclaren. |
|
| Back to top |
|
|
Jeff Kenton
Guest
|
| Posted:
Sun Aug 21, 2005 4:15 pm Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
Dan Koren wrote:
| Quote: | "Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message
|
...
| Quote: | Well, only if you disregard companies like
Kendall Square Research as non-commercial.
How many systems did KSR manage
to sell before they went under?
|
Ship? Or sell? I think that was their downfall -- shipped more systems than
they actually got paid for.
jeff |
|
| Back to top |
|
|
Joe Pfeiffer
Guest
|
| Posted:
Sun Aug 21, 2005 9:05 pm Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
Jeff Kenton <jeffrey.kenton@comcast.net> writes:
| Quote: | Dan Koren wrote:
"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message
...
Well, only if you disregard companies like
Kendall Square Research as non-commercial.
How many systems did KSR manage
to sell before they went under?
Ship? Or sell? I think that was their downfall -- shipped more systems
than they actually got paid for.
|
IIRC, their *real* mistake was selling systems with a promise to buy
any unused time on them at a rate that equalled the payments on the
machines. The SEC Was Not Happy...
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer
skype: jjpfeifferjr |
|
| Back to top |
|
|
Rupert Pigott
Guest
|
| Posted:
Mon Aug 22, 2005 12:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
Ketil Malde wrote:
[SNIP]
| Quote: | The fun thing about PPro was that it was hampered by a perception of
being too slow on 16-bit code, and consequently on the current
Windows - and it wasn't tremendously successful. (It was a great CPU
to run Linux on, though.) IIRC, one of the improvements in the PII
was faster 16-bit execution.
|
AFAIR there was a bit of a gotcha there. The PII's ran their
cache at 1/2 core clock, whereas the PPro's ran their cache
at core clock. Eventually the improvements in PII clock speed
made the point moot though.
I had a couple of codes get bitten by the cache downgrade, and
I did not see any significant wall-clock benefit in the 16 bit
improvements. YMMV, I used Pro's for doing real work, not
running pessimal benchmarks. :)
Cheers,
Rupert |
|
| Back to top |
|
|
Dan Koren
Guest
|
| Posted:
Mon Aug 22, 2005 12:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
"John Mashey" <old_systems_guy@yahoo.com> wrote in message
news:1124657185.934655.105240@o13g2000cwo.googlegroups.com...
| Quote: |
Joe Pfeiffer wrote:
IIRC, their *real* mistake was selling systems with a promise to buy
any unused time on them at a rate that equalled the payments on the
machines. The SEC Was Not Happy...
That wasn't the only one. When I was at SGI, I used to run into KSR at
universities, and I heard often about a specific kind of deal as
follows:
a) KSR would sell a system to a university at price $X, but if the
university did performance studies and wrote papers, they would get a
"research grant" of $Y.
KSR would book this as $X revenue.
b) However, on the assumption that all this would happen, the
university would actually pay $X-$Y upfront.
I heard these stories in the usual way, where several university
departments were pooling resources to get systems, with some people
wanting SGI Challenges (basically to get work done), and others
attracted by the KSR deals and wanting to write research papers. The
former tended to tell us about the KSR deals.
An (generally competent) analyst friend had written a favorable review
of KSR, and I called him to understand why [prepatory to shorting KSR],
but he was out, and I had to go to Europe for 2 weeks, and by the time
I got back, the news had gotten out, and KSR stock was way down, and it
was too late. Sigh.
I did talk to the analyst later, and he ruefully said he'd learned a
good lesson: he'd called many KSR customers, who told him that they
actually worked ... but he hadn't asked anybody about the financing.
There was a lot of this stuff going on in the early/mid 1990s, during
that bizarre period when VCs had funded a horde of mini-super and super
computing companies, for no earthly good reason :-) KSR was one ofthe
more extreme cases, which was too bad, since they at least had
interesting ideas.
|
Perhaps fittingly, the standard acronym for
KSR's architecture is "COMA". It stands for
"cash only memory architecture" ;-)
dk |
|
| Back to top |
|
|
Dan Koren
Guest
|
| Posted:
Mon Aug 22, 2005 12:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
"George Coulouris" <coulouris@ncbi.nlm.nih.gov> wrote in message
news:slrndfr0kt.5ht.coulouris@lacerta.nlm.nih.gov...
Thanks.
I'm glad you did not
say "wrap the mind".
dk |
|
| Back to top |
|
|
Dan Koren
Guest
|
| Posted:
Mon Aug 22, 2005 12:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
"Joe Pfeiffer" <pfeiffer@cs.nmsu.edu> wrote in message
news:1bd5o7cm2t.fsf@viper.cs.nmsu.edu...
| Quote: | Jeff Kenton <jeffrey.kenton@comcast.net> writes:
Dan Koren wrote:
"Nick Maclaren" <nmm1@cus.cam.ac.uk> wrote in message
...
Well, only if you disregard companies like
Kendall Square Research as non-commercial.
How many systems did KSR manage
to sell before they went under?
Ship? Or sell? I think that was their downfall -- shipped more systems
than they actually got paid for.
IIRC, their *real* mistake was selling systems with a promise to buy
any unused time on them at a rate that equalled the payments on the
machines. The SEC Was Not Happy...
|
IIRC the matter that caught the SEC's eye was a
little more egregious than that. However, I do
not recall all the fine print (which made quite
interesting reading at the time) and you may be
right.
A question just popped through my head: if KSR
(or any computer vendor for that matter) had
signed instead an agreement to buy back the
heat generated by the systems and resell it,
would that also be considered unlawful? In an
economy where fuel costs keep rising, it may
not be such a bad idea! ;-)
Some of you probably know that one/some of
the Cray buildings in Eagan and/or Mendota
Heights near Minneapolis were completely
heated by the computers' cooling systems.
dk |
|
| Back to top |
|
|
John Mashey
Guest
|
| Posted:
Mon Aug 22, 2005 12:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
Joe Pfeiffer wrote:
| Quote: | IIRC, their *real* mistake was selling systems with a promise to buy
any unused time on them at a rate that equalled the payments on the
machines. The SEC Was Not Happy...
|
That wasn't the only one. When I was at SGI, I used to run into KSR at
universities, and I heard often about a specific kind of deal as
follows:
a) KSR would sell a system to a university at price $X, but if the
university did performance studies and wrote papers, they would get a
"research grant" of $Y.
KSR would book this as $X revenue.
b) However, on the assumption that all this would happen, the
university would actually pay $X-$Y upfront.
I heard these stories in the usual way, where several university
departments were pooling resources to get systems, with some people
wanting SGI Challenges (basically to get work done), and others
attracted by the KSR deals and wanting to write research papers. The
former tended to tell us about the KSR deals.
An (generally competent) analyst friend had written a favorable review
of KSR, and I called him to understand why [prepatory to shorting KSR],
but he was out, and I had to go to Europe for 2 weeks, and by the time
I got back, the news had gotten out, and KSR stock was way down, and it
was too late. Sigh.
I did talk to the analyst later, and he ruefully said he'd learned a
good lesson: he'd called many KSR customers, who told him that they
actually worked ... but he hadn't asked anybody about the financing.
There was a lot of this stuff going on in the early/mid 1990s, during
that bizarre period when VCs had funded a horde of mini-super and super
computing companies, for no earthly good reason :-) KSR was one ofthe
more extreme cases, which was too bad, since they at least had
interesting ideas. |
|
| Back to top |
|
|
Jeff Kenton
Guest
|
| Posted:
Mon Aug 29, 2005 12:15 am Post subject:
Re: What was new&important in computer architecture 10 years |
|
|
All true, John. Their big problem, from the SEC point of view, was booking
the full price as revenue when the machine shipped even thought nobody really
expected full payment (any payment?) for a large fraction of those machines.
jeff -- consultant at KSR at the end (yes I got paid on that last day)
John Mashey wrote:
| Quote: | Joe Pfeiffer wrote:
IIRC, their *real* mistake was selling systems with a promise to buy
any unused time on them at a rate that equalled the payments on the
machines. The SEC Was Not Happy...
That wasn't the only one. When I was at SGI, I used to run into KSR at
universities, and I heard often about a specific kind of deal as
follows:
a) KSR would sell a system to a university at price $X, but if the
university did performance studies and wrote papers, they would get a
"research grant" of $Y.
KSR would book this as $X revenue.
b) However, on the assumption that all this would happen, the
university would actually pay $X-$Y upfront.
I heard these stories in the usual way, where several university
departments were pooling resources to get systems, with some people
wanting SGI Challenges (basically to get work done), and others
attracted by the KSR deals and wanting to write research papers. The
former tended to tell us about the KSR deals.
An (generally competent) analyst friend had written a favorable review
of KSR, and I called him to understand why [prepatory to shorting KSR],
but he was out, and I had to go to Europe for 2 weeks, and by the time
I got back, the news had gotten out, and KSR stock was way down, and it
was too late. Sigh.
I did talk to the analyst later, and he ruefully said he'd learned a
good lesson: he'd called many KSR customers, who told him that they
actually worked ... but he hadn't asked anybody about the financing.
There was a lot of this stuff going on in the early/mid 1990s, during
that bizarre period when VCs had funded a horde of mini-super and super
computing companies, for no earthly good reason :-) KSR was one ofthe
more extreme cases, which was too bad, since they at least had
interesting ideas.
|
|
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in