On Wednesday 14 June 2006 11:10, Mark Hahn wrote:
> when a vendor starts using the phrase "supported platform",
> I always ask "tell me about the bugs in your code which violate
> the ABI". apps should list a required set of ABIs.
I remember the pain one of the people I know in Melbourne went th
On Wed, Jun 14, 2006 at 02:26:06PM -0500, Geoff Jacobs wrote:
> I know many commercial F90 compilers have origins in big iron. Absoft
> came from Cray, Lahey from Fujitsu, Pathscale from SGI. Did Portland
> Group make their own from scratch, or did they make a deal with someone
> else to develop a
Well, the time is here. When MS puts their marketing machine in gear they
will be sending a clear an consistent message to the HPC community. At
this
point I have to ask "What is the Linux Cluster message?"
Over the last 3-4 years I have been involved in trying to help formulate
and broadcast this
I know many commercial F90 compilers have origins in big iron. Absoft
came from Cray, Lahey from Fujitsu, Pathscale from SGI. Did Portland
Group make their own from scratch, or did they make a deal with someone
else to develop an existing compiler on x86?
--
Geoffrey D. Jacobs
Go to the Chinese
On Wed, Jun 14, 2006 at 12:25:04PM -0400, Joe Landman wrote:
> Quadrics/Myrinet/Infinipath folks... care to comment? Somehow I don't
> really think that this makes sense, cutting off a $4B+ market growing at
> 60+%/year in order to service a market of questionable size, growth
> rate, and inte
Vincent Diepeveen wrote:
Then it'll have soon better drivers than linux too for highend network
cards,
who no longer will release.
Quadrics/Myrinet/Infinipath folks... care to comment? Somehow I don't
really think that this makes sense, cutting off a $4B+ market growing at
60+%/year in ord
- Original Message -
From: "Joe Landman" <[EMAIL PROTECTED]>
To: "Thomas H Dr Pierce" <[EMAIL PROTECTED]>
Cc: ; "Robert G. Brown" <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>
Sent: Wednesday, June 14, 2006 2:02 PM
Subject: Re: [Beowulf] MS HPC... Oh dear...
Hi Thomas:
Thomas H Dr Pier
Douglas Eadline wrote:
Of course there is no absolute rule on how to maintain a market, but
stagecoach companies who did not recognize locomotives kind of missed
the technology curve.
Joking aside, there are things Linux can do that Windows (and any closed
source OS for that matter) cannot do.
Of course there is no absolute rule on how to maintain a market, but
stagecoach companies who did not recognize locomotives kind of missed
the technology curve.
Joking aside, there are things Linux can do that Windows (and any closed
source OS for that matter) cannot do. There are market segments
Of course there is no absolute rule on how to maintain a market, but
stagecoach companies who did not recognize locomotives kind of missed
the technology curve.
Joking aside, there are things Linux can do that Windows (and any closed
source OS for that matter) cannot do. There are market segments
Of course there is no absolute rule on how to maintain a market, but
stagecoach companies who did not recognize locomotives kind of missed
the technology curve.
Joking aside, there are things Linux can do that Windows (and any closed
source OS for that matter) cannot do. There are market segments
Hi Thomas:
Thomas H Dr Pierce wrote:
[...]
> Microsoft HPC will work IF the market wants the technology that they can
> supply. The answer is in market segments. And the Supercomputer market
I don't think the technology is what will win people over. Its the
packaging.
[...]
> The Linux HPC
On Tue, 13 Jun 2006 at 10:55pm, Gerry Creager N5JXS wrote
Mark Hahn wrote:
I tend to Tyan for supra-desktop machines, though expect most SuperMicro
and MSI would also do OK.
Tyan or SuperMicro, depending on whether I'm going Opteron (preferred) or
Xeon.
Err, why the distinction? Both v
On Tue, 13 Jun 2006, Vincent Diepeveen wrote:
You and i are like junkies busy downloading all kind of linux distributions
'for free',
sitting and waiting a LONG time (in your case expensive time paid by the
government,
in my case i just CLAIM my time is expensive as i'm running my own company)
Tri-bonded? Have you tried this? What MPI stack and/or other
interconnect software do you plan to use?
My understanding is that kernel support for bonding has been vastly
improved. This is only through anectodtal accounts of others. I have
not yet tried it myself.
Instead of using MPI, I am wr
I'm trying to install MPICH on a small cluster of powermac G4's
running opensuse 10.1 . The machines are in a campus network
environment with no cluster-specific fileserver, central LDAP host
etc. While I have been able to compile and install MPICH, I can't
get the ring up and running (mp
Joe et al.
I'm not sure how many of you will be interested, but I think that the
g03/Centos 3.6 issue is resolved.
In talking to David Hibbs in Australia (who had experienced a similar
issue), I found out that the issue was likely a compiler bug. David had
solved the problem by compiling wit
All,Please accept my apologies if this is not the place but...I have an older IBM Linux cluster that has the following for sale:
(1) Myrinet2000 5 Slot Switch Chassis(19) Myrinet2000 LAN adapters with
cables (M3S-PC164B-2)(3) M3-SW16-8S Line cards(1) M3-Blank
The rest of the cluster has been slow
Gardner
Pomper wrote:
Well, at the risk of the entire
mailing list coming down on my, I must disagree with the prevailing
viewpoint.
it does seem
that as soon
as someone mentions MS or somthing related to MS, most turn away...
give them some credit, they keep most idiots away from linu
Richard Walsh wrote:
All,
Could those of you who have perhaps used or researched the general
purpose use of GPUs (vendors, buyers, builders) to augment the data-
parallel compute power of your clusters add, subtract, and/or comment
on the following summary of the current options in this area?
-CALL FOR PARTICIPATION
HPDC'15
The 15th IEEE International Symposium on
High Performance Distributed Computing
June 19-23 2006
Paris, France
--> http://hpdc.lri.fr
www.hpdc.org
Joe,
Thank you. ulimit -a is unlimited.
I will need to look at the Makefile to check the stuff.
Mike Davis
Joe Landman wrote:
Hi Mike:
You hit a segfault. What options did you use with GO3? The last
build we worked on with a customer that we had to restrict it to -O2
-Mnosse or somethi
Folks,
We are all so technical!
It does not work that way
Microsoft HPC will work IF the market
wants the technology that they can supply. The answer is in market
segments. And the Supercomputer market segment has been the science and
engineering users of the last 30 years. It has no
Lombard, David N wrote:
Joe Landman is quite correct that you can do well by presenting a Linux
cluster with a web-based interface and SMB-based storage. But, let's
face it, those who respond to that pitch will respond all the more to
the full Windows gig.
A web-based interface would p
On Tue, 13 Jun 2006, Geoff Jacobs wrote:
Repeat after me: Wine Is Not an Emulator
I'm not sure about Cedega, but WINE dll hell is just as bad as the real
thing(tm).
Cedega is supra-tree commercial WINE on steroids. Sold almost entirely
to gamers. We have several in household and use cedega
On Tue, 2006-06-13 at 20:53 +0100, James Cownie wrote:
> > Nothing. Unfortunately most folks use statically linked binaries for
> > MPI, so .so's are not a factor. I could be wrong, and maybe there
> > is a
> > way to get statically linked binaries to respect LD_PRELOAD or
> > LD_LIBRARY_PATH,
On Tue, Jun 13, 2006 at 08:53:48PM +0100, James Cownie wrote:
> As I understand it this is what Intel's MPI (for LInux) does, so that
> you can choose the underlying hardware transport without relinking
> your code.
Just like HP-MPI, Scali MPI, Verari MPI, Par-Tec's MPI, and OpenMPI.
Did I l
27 matches
Mail list logo
