Hi Vincent,
Yes, the libraries can't cover all calculations, or they can do only some
calculations based on GPU.
GPGPU is just a small step for many-core architecture. It equips great power,
but with deadly weakness.
When you have got a complex calculations could be arranged to many pieces and
On Wed, Aug 27, 2008 at 14:00, Perry E. Metzger <[EMAIL PROTECTED]> wrote:
> Cyberguys sells several models -- PCI and Mini PCI, though no PCI-X
> that I've seen there.
That is likely because PCI2.0 and 3.0 cards work in PCI-X slots. A
PCI post-code reader *should* work in a PCI-X slot just the s
On Thu, Aug 28, 2008 at 12:56 AM, Nifty niftyompi Mitch <
[EMAIL PROTECTED]> wrote:
> On Wed, Aug 27, 2008 at 10:42:23AM -0400, Prentice Bisbal wrote:
> >
> > I will soon be setting up my first cluster that uses Infiniband. I've
> > searched the net, but I haven't found any good tutorials or artic
On Wed, Aug 27, 2008 at 11:56, Nifty niftyompi Mitch
<[EMAIL PROTECTED]> wrote:
> One key to Infiniband is that the IB fabric must have
> a subnet manager. You only need one...
And you need only one (Not more than one). That's just as important.
___
Be
The HPC TECH FORUM will have it's first meeting under the new format
Tuesday September 9th at 2:30 in Lanham Maryland.
Please go to http://www.hpctechforum.com for details
Mike Fitzmaurice
___
Beowulf mailing list, Beowulf@beowulf.org
To change your su
I have a SuperMicro "SuperBlade" that came with IB - At this point we
have installed Rocks. I'm in the learning curve for IB in general and
Rocks specifically.
The IB roll was installed but IB does not work as of yet. I'm just in
need of some pointers to get me a leap ahead. I've spent a few d
Perhaps his judgment isn't "universally" objective (not influenced by
personal feelings, interpretations, or prejudice; based on facts), but could
very well could be objective (and his version of reality) in his little part
of the world. Being at a University I've run across more of what he said
t
I have tried CTM on Firestream, it was not a good tool for most programmer.
CAL may be different, maybe. AMD GPU's SP is 5D SIMD based, which means if you
need all ALU working, you should program your codes based on SIMD and then as
what CUDA did. AMD's 4870 was supposed to provide 1.2TFlops, but
On Wed, 27 Aug 2008, Lux, James P wrote:
But is it harder or easier to do that than to teach the Scientist to
code? In the long run, the effort in my approach is better invested,
because it allows you to put 2 or more people to work, rather than just
one. If the Scientist spends their time c
On Wed, Aug 27, 2008 at 10:04:10PM +0200, Vincent Diepeveen wrote:
> I would go even further. It is definately true that from physics some
> scientists learn programming really well, at the best level i'd say.
> That's not their job however.
Vincent,
I know you're an expert on chess, but I did
Greg Lindahl <[EMAIL PROTECTED]> writes:
> On Wed, Aug 27, 2008 at 03:37:54PM -0400, Perry E. Metzger wrote:
>> By the way, this isn't special to computer science. Imagine if laymen
>> were directing the physics research . Progress in physics might be
>> rather slowed down, don't you think.
>
> Sp
Jim,
I would go even further. It is definately true that from physics some
scientists learn programming really well, at the best level i'd say.
That's not their job however. From hardware engineering sometimes
even better low level programmers get on the scene.
The point being that a great
On Wed, Aug 27, 2008 at 03:37:54PM -0400, Perry E. Metzger wrote:
> By the way, this isn't special to computer science. Imagine if laymen
> were directing the physics research . Progress in physics might be
> rather slowed down, don't you think.
Spoken like a computer scientist.
But this topic h
"Lux, James P" <[EMAIL PROTECTED]> writes:
>>It is not uncommon for me to get someone asking me a question of the
>>form "I need to do X, how can I do it", and while explaining to them
>>how to do X, further questioning slowly reveals that they really
>>should be doing Y. It just isn't possible fo
On Wed, 27 Aug 2008, Vincent Diepeveen wrote:
Physics in itself is eating 50%+ of all supercomputer time.
Only 50%?
Comrades, we must work harder! Build more computers!
It's not for lack of trying -- there are any number of projects any one
of which could use 100% of the aggregate cycles on
-Original Message-
From: Perry E. Metzger [mailto:[EMAIL PROTECTED]
Sent: Wednesday, August 27, 2008 11:53 AM
To: Lux, James P
Cc: Steve Herborn; Beowulf@beowulf.org
Subject: Re: [Beowulf] Stroustrup regarding multicore
>"Lux, James P" <[EMAIL PROTECTED]> writes:
>> I would say that you
"David Mathog" <[EMAIL PROTECTED]> writes:
> I have been thinking about getting a POST diagnostic card since my main
> file server acted up the other day and kept hanging before the BIOS came
> up - then just as mysteriously cured itself. (This was a SuperMicro
> H8DC8 motherboard. The power supp
"Lux, James P" <[EMAIL PROTECTED]> writes:
> I would say that you shouldn't be teaching the Scientist to
> code. You should be teaching them to write good requirements
> documents & test cases that someone who is skilled at coding can
> use.
I think it is very hard to do that.
It is not uncommon
On Wed, Aug 27, 2008 at 01:19:43PM -0400, Perry E. Metzger wrote:
> Many smart people use C++, few smart people love C++.
C++ is an octopus constructed by nailing legs onto a dog. Can we
please end this useless discussion? Thanks.
-- greg
___
Beowulf
Ed Hill <[EMAIL PROTECTED]> writes:
> So, I'm curious about what you consider "legitimate"
> and why. And please understand -- I'm not trying to start arguments or
> change any opinions about language choices [ other than perhaps my own
> :-) ].
It is pretty straightforward. C++ is a big mess
I have been thinking about getting a POST diagnostic card since my main
file server acted up the other day and kept hanging before the BIOS came
up - then just as mysteriously cured itself. (This was a SuperMicro
H8DC8 motherboard. The power supply tested OK with two different PS
testers.) There
On Wed, Aug 27, 2008 at 10:42:23AM -0400, Prentice Bisbal wrote:
>
> I will soon be setting up my first cluster that uses Infiniband. I've
> searched the net, but I haven't found any good tutorials or articles on
> configuring Infiniband networking on Linux and/or how the Infiniband
> networking p
In message from "Li, Bo" <[EMAIL PROTECTED]> (Wed, 27 Aug 2008 08:50:11
+0800):
Hello,
IMHO, it is better to call the BLAS or similiar libarary rather than
programing you own functions.
Eh, I would be happy to use GPGPUs only via math libraries - it's
equal for me to "black box" as Vincent sa
-Original Message-
From: Steve Herborn [mailto:[EMAIL PROTECTED]
Sent: Wednesday, August 27, 2008 9:33 AM
To: 'Vincent Diepeveen'; 'Tim Cutts'
Cc: Beowulf@beowulf.org; Lux, James P
Subject: RE: [Beowulf] Stroustrup regarding multicore
Perhaps his judgment isn't "universally" objective (n
Have you tried ACML on Firestream?
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On
> Behalf Of Li, Bo
> Sent: 27 August 2008 01:50
> To: Mikhail Kuzminsky; Vincent Diepeveen
> Cc: Beowulf
> Subject: Re: [Beowulf] gpgpu
>
> Hello,
> IMHO, it is better to call th
Hi Bo,
Thanks for your message.
What library do i call to find primes?
Currently it's searching here after primes (PRP's) in the form of p
= (2^n + 1) / 3
n is here about 1.5 million bits roughly as we speak.
For SSE2 type processors there is the George Woltman assembler code
(MiT) to d
On Aug 26, 2008, at 10:34 PM, Tim Cutts wrote:
On 26 Aug 2008, at 2:29 pm, Perry E. Metzger wrote:
I think part of the issue is that most people doing scientific
computing don't have computer science backgrounds, which is a
shame.
There is an unwritten recruitment rule, certainly in my fie
I will soon be setting up my first cluster that uses Infiniband. I've
searched the net, but I haven't found any good tutorials or articles on
configuring Infiniband networking on Linux and/or how the Infiniband
networking protocol (if that is the correct term) works.
Can someone point me to a good
On Wed, 27 Aug 2008 10:00:19 -0400 "Perry E. Metzger" wrote:
>
> Getting back to the original topic, a set of C++ classes that handled
> APL style array operations and automatically parallelized them on
> vector processors would be quite neat, and would actually be a
> legitimate use of C++ (unlik
On Wed, Aug 27, 2008 at 10:26:17AM -0400, Perry E. Metzger wrote:
>
> I'm specing out a small cluster this week, and I'm looking for
> recommendations on decent, inexpensive server mother boards. Important
> features are handling reasonably recent Intel dual or quad core
I'm not sure why you're e
Hi,
I use to shutdown my Rocks 5 based Beowulf cluster of 4-nodes, when it
is unused.
Today when I restarted my Rocks5 frontend, it showed errors syaing
"Network is unreachable" and saying during startup of sge_master it
shows with comlib error.
When I logon as root and type "network service res
I'm specing out a small cluster this week, and I'm looking for
recommendations on decent, inexpensive server mother boards. Important
features are handling reasonably recent Intel dual or quad core
processors, 8G of ECC memory, decent disk controller, decent onboard
NICs -- onboard IPMI and the li
Alan Louis Scheinine <[EMAIL PROTECTED]> writes:
> My list does not include C++ for scientific programming.
> The idea here is how to get past the initial difficulties
> of using C++. As I've written a few hours ago, simple C
> is efficient for numerical programming, as others have said.
I've of
33 matches
Mail list logo
