On Thu, Jun 29, 2006 at 10:10:04AM +0200, Joachim Worringen wrote:
> SPEC benchmarking is an example for this. The advantage of SPEC is that you
> only need one machine to reproduce the results, not a whole cluster.
That's true for SPECcpu. SPEChpc, the MPI codes, require a cluster.
While the cu
Patrick Geoffray wrote:
Recently, I have been thinking about something that you may like. With
motherboards with 4 good PCIE slots coming on the marketing (driven by
SLI and such), it could be doable to have a reasonably sized machine,
let's say 64 nodes, with 4 different interconnects in it. I
Kevin Ball wrote:
- *If* you feel you need to use such a new metric for whatever reason, you
should at least publish the benchmark that is used to gather these numbers to
allow others to do comparative measurements. This goes to Greg.
This has been done. You can find the benchmark used for me
Hi Patrick,
Patrick Geoffray wrote:
Believe it or not, but I really want to do that. I don't think it's
appropriate to compare results from other vendors though: in Europe,
it's forbidden to do comparative advertisement (ie the soap X washes
more white that the brand Y) and I completely agree w
Patrick Geoffray wrote:
White papers are evil by definition. They show what you want to show,
and there is no peer review so you can say what you want.
ACK. At this occasion, I (as Program Co-Chair) can not resist to give a pointer
to the EuroPVM/MPI conference series. Next event is in Septemb
And on we go...
On Wed, 28 Jun 2006, Patrick Geoffray wrote:
> I totally agree that the gap (g) gets important when the Latency (L) is
> small, but only when you send multiple messages in a row. When sending
> one message at a time, it's irrelevant (if the time between messages or
> even the s
- Original Message -
From: "Mark Hahn" <[EMAIL PROTECTED]>
To: "Vincent Diepeveen" <[EMAIL PROTECTED]>
Cc:
Sent: Thursday, June 29, 2006 2:44 AM
Subject: Re: [Beowulf] Three notes from ISC 2006
a huge L3 cache (which most specfp software is some
> a huge L3 cache (which most specfp software is somehow) that it destroyed
spec is, after all, CPU2000, and things have changed quite a lot in 6+ years.
Itaniums were one of the earliest "breaks" of spec cpu2000 codes,
(besides sun's spec-special compiler). if you take it2 results
and remove t
leases.)
Craig
Add to that that the new socket from intel is like 125 watts TDP. That's
just not normal. That's wasting as much as itanium2!
Vincent
----- Original Message - From: "Kevin Ball" <[EMAIL PROTECTED]>
To: "Erik Paulson" <[EMAIL PROTEC
riginal Message - From: "Kevin Ball" <[EMAIL PROTECTED]>
To: "Erik Paulson" <[EMAIL PROTECTED]>
Cc: ; "Patrick Geoffray" <[EMAIL PROTECTED]>
Sent: Wednesday, June 28, 2006 10:29 PM
Subject: Re: [Beowulf] Three notes from ISC 2006
On Wed, 2006-06-28 a
e -
From: "Kevin Ball" <[EMAIL PROTECTED]>
To: "Erik Paulson" <[EMAIL PROTECTED]>
Cc: ; "Patrick Geoffray" <[EMAIL PROTECTED]>
Sent: Wednesday, June 28, 2006 10:29 PM
Subject: Re: [Beowulf] Three notes from ISC 2006
On Wed, 2006-06-28 at 13:
Salut Christian,
Christian Bell wrote:
I agree with you that the inverse of message rate, or the small
message gap in logP-derived models is a more useful way to view the
metric. How much more important it is than latency depends on what
the relative difference is between your gap and your late
On Wed, 2006-06-28 at 13:41, Erik Paulson wrote:
> On Wed, Jun 28, 2006 at 04:25:40PM -0400, Patrick Geoffray wrote:
> >
> > I just hope this will be picked up by an academic that can convince
> > vendors to donate. Tax break is usually a good incentive for that :-)
> >
>
> How much care should
On Wed, 28 Jun 2006, Patrick Geoffray wrote:
High message rate is good, but the question is how much is enough ? At 3
million packet per second, that's 0.3 us per message which all of it is
used by the communication library. Can you name real world applications
that need to send messages every
On Wed, Jun 28, 2006 at 04:25:40PM -0400, Patrick Geoffray wrote:
>
> I just hope this will be picked up by an academic that can convince
> vendors to donate. Tax break is usually a good incentive for that :-)
>
How much care should be given to the selection of the nodes? Performance
is a funct
Kevin Ball wrote:
I have two large concerns.
One is that finding a software stack that works with the latest
interconnect products may or may not correlate well with what end users
are interested in. For some protocols (particularly MPI) this doesn't
I would only care for MPI, at least at
Patrick,
Maybe I'm just too dense to understand. But, you've basically labelled
Greg's post as spam. You've called their metric nonsense. You've
criticized the published number they used that came from your company's
product. For what?
You've provided no new data. No reference to new data.
Mike,
Mike Davis wrote:
Maybe I'm just too dense to understand. But, you've basically labelled
Greg's post as spam.
Yes, I did. Telling me about a new white paper and about something that
I cannot know but I really should does fit my definition of spam. It was
borderline, I recognized that,
On Wed, 28 Jun 2006, Patrick Geoffray wrote:
> High message rate is good, but the question is how much is enough ? At 3
> million packet per second, that's 0.3 us per message which all of it is
> used by the communication library. Can you name real world applications
> that need to send message
Hi Joachim,
Joachim Worringen wrote:
An offer for "getting a secret white paper on request" is marketing, you
are right. But at least the SPEC number was technical content - and we
don't want to analyse every posting sentence-by-sentence, do we?
The SPEC stuff was actually fine. I didn't regi
Patrick,
Thank you for the rapid and thoughtful response,
On Wed, 2006-06-28 at 11:23, Patrick Geoffray wrote:
> Hi Kevin,
>
> Kevin Ball wrote:
> > Patrick,
> >
> >>
> >> From you flawed white papers, you compared your own results against
> >> numbers picked from the web, using older int
Hi Kevin,
Kevin Ball wrote:
Patrick,
From you flawed white papers, you compared your own results against
numbers picked from the web, using older interconnect with unknown
software versions.
I have spent many hours searching to try to find application results
with newer Myrinet and Me
> - *If* you feel you need to use such a new metric for whatever reason, you
> should at least publish the benchmark that is used to gather these numbers to
> allow others to do comparative measurements. This goes to Greg.
This has been done. You can find the benchmark used for message rate
me
Having subscribed to this board for quite some time (a time when I was
in a Beowulf admin class with Jeff Layton, followed Greg's work on the
Legion project up the road in Charlottesville, bought our first cluster
from Doug Eadline and Paralogic, and ran into Robert Brown at Linux
Expo), I do n
Patrick Geoffray wrote:
Greg Lindahl wrote:
On Wed, Jun 28, 2006 at 07:28:53AM -0400, Patrick Geoffray wrote:
I have keep it quiet even when you where saying things driven by
marketing rather than technical considerations (the packet per
second nonsense),
Patrick, that "packet per second non
Patrick,
>
> From you flawed white papers, you compared your own results against
> numbers picked from the web, using older interconnect with unknown
> software versions.
I have spent many hours searching to try to find application results
with newer Myrinet and Mellanox interconnects. I
Vincent Diepeveen wrote:
"Microsoft is usually at the extreme of the marketing spectrum"
Is this your official companies statement about microsoft?
I am not an officer of the company that employs me, thus I have no
official voice. I cannot sign contract and my expression has no legal
bindin
Vincent Diepeveen wrote:
Not at all good marketing that third remark.
Because if there was really something interesting to report,
then it would already have been reported by the *official* marketing
department.
No. Marketing effort implies coordination, that's why most announcements
are emb
]>
To: "Chris Dagdigian" <[EMAIL PROTECTED]>
Cc:
Sent: Wednesday, June 28, 2006 4:13 PM
Subject: Re: [Beowulf] Three notes from ISC 2006
Chris,
Chris Dagdigian wrote:
In short, this was appropriate (and interesting). We've all seen vendor
spam and disguised marketing and this
[EMAIL PROTECTED]>
Cc:
Sent: Wednesday, June 28, 2006 4:13 PM
Subject: Re: [Beowulf] Three notes from ISC 2006
Chris,
Chris Dagdigian wrote:
In short, this was appropriate (and interesting). We've all seen vendor
spam and disguised marketing and this does not rise anywhere close to
t
ally having a big L2 or L3 matters a lot there and quite little in real
world where the matrix calculations
performed don't fit in your L2 :)
- Original Message -
From: "Joe Landman" <[EMAIL PROTECTED]>
To:
Sent: Wednesday, June 28, 2006 3:45 PM
Subject: Re: [Beowulf] T
Joe Landman wrote:
Greg Lindahl wrote:
On Wed, Jun 28, 2006 at 08:28:06AM -0400, Mark Hahn wrote:
the "I know something that I can't tell" bit was childish though ;)
Indeed, it was. I plead jet-lag.
No. It was good marketing. Anyone on the list not at least a little
curious what it is th
Chris,
Chris Dagdigian wrote:
In short, this was appropriate (and interesting). We've all seen vendor
spam and disguised marketing and this does not rise anywhere close to
that level.
I disagree on the level. I use the rule that a vendor should never
initiate a thread, only answer someone el
Greg Lindahl wrote:
On Wed, Jun 28, 2006 at 07:28:53AM -0400, Patrick Geoffray wrote:
I have keep it quiet even when you where saying things driven by
marketing rather than technical considerations (the packet per
second nonsense),
Patrick, that "packet per second nonsense" is the technical r
Greg Lindahl wrote:
> On Wed, Jun 28, 2006 at 08:28:06AM -0400, Mark Hahn wrote:
>
>> the "I know something that I can't tell" bit was childish though ;)
>
> Indeed, it was. I plead jet-lag.
No. It was good marketing. Anyone on the list not at least a little
curious what it is that Greg can'
On Wed, Jun 28, 2006 at 08:28:06AM -0400, Mark Hahn wrote:
> the "I know something that I can't tell" bit was childish though ;)
Indeed, it was. I plead jet-lag.
-- greg
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest
> My $.02 of course
me too. the "I know something that I can't tell" bit was childish though ;)
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowul
On Wed, Jun 28, 2006 at 07:28:53AM -0400, Patrick Geoffray wrote:
> I have keep it quiet even when you where saying things driven by
> marketing rather than technical considerations (the packet per
> second nonsense),
Patrick, that "packet per second nonsense" is the technical reason our
intercon
... it was a short note written by a list regular with a good signal
to noise ratio. The whitepaper contents sound on-topic for many
people on this list and the "email me for a copy" is exactly what I
see myself and many other employed-by-industry types do when we want
to share something
Greg Lindahl wrote:
Second, we have a new whitepaper about performance of the Intel
Woodcrest CPU and InfiniPath interconnect on real applications, email
me for a copy.
Third, MH MHH MH. (That's the sound I make when I
can't tell you something.)
Since when is Beowulf a plac
First off, HP has posted the highest SPEC200fp peak result for an x86
cpu, 3048. First over 3000, too. It uses a combination of the Intel
and PathScale compilers.
http://www.spec.org/cpu2000/results/res2006q2/cpu2000-20060612-06162.asc
Interestingly, this number is better than all currently publi
41 matches
Mail list logo
