Re: [Beowulf] gpgpu

Yes, Firestream has a great paper performance, but how can you get it? But for the costs, if you don't mind to use some un-professional components, you can try their gaming cards, much cheaper. We bought NVidia's last flagship card 8800Ultra for 600 Euro, what's a crazy price, and now you can buy

RE: [Beowulf] Stroustrup regarding multicore

On Thu, 28 Aug 2008, Lux, James P wrote: Again, if you're building software for a real estate office it is straightforward for the programmer to just learn what your business is and in any case high performance isn't needed. If you are trying to simulate supernova explosions and you can barely

RE: [Beowulf] Stroustrup regarding multicore

Again, if you're building software for a real estate office it is straightforward for the programmer to just learn what your business is and in any case high performance isn't needed. If you are trying to simulate supernova explosions and you can barely cram the magnetohydrodynamics in a piss poo

Re: [Beowulf] gpgpu

In message from "Li, Bo" <[EMAIL PROTECTED]> (Thu, 28 Aug 2008 14:20:15 +0800): ... Currently, the DP performance of GPU is not good as we expected, or only 1/8 1/10 of SP Flops. It is also a problem. AMD data: Firestream 9170 SP performance is 5 GFLOPS/W vs 1 GFLOPS/W for DP. It's 5 times sl

Re: [Beowulf] Stroustrup regarding multicore

"Lux, James P" <[EMAIL PROTECTED]> writes: > But there's a lot of "hidden process" in building high quality > software that the consumer (the physicist in this example) doesn't > need to be aware of. An engineer building a bridge needs to know > the properties of steel, and have some understandin

[Beowulf] Re: POST card recomendation?

"Billy Crook" <[EMAIL PROTECTED]> wrote: > 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

RE: [Beowulf] Stroustrup regarding multicore

-Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Perry E. Metzger Sent: Thursday, August 28, 2008 8:04 AM To: Steve Herborn Cc: Beowulf@beowulf.org Subject: Re: [Beowulf] Stroustrup regarding multicore "Steve Herborn" <[EMAIL PROTECTED]> writes: > Howeve

Re: [Beowulf] Stroustrup regarding multicore

I agree entirely with Perry here. I'd take it further: even in the case of giving the machinist instructions, "12x12 with holes here and here", it would help if the machinist has some sense of what you are building. Will the product be hot enough so that metal expands and contracts? humidity? shoul

Re: [Beowulf] Stroustrup regarding multicore

On 27 Aug 2008, at 11:47 am, Vincent Diepeveen wrote: There is an unwritten recruitment rule, certainly in my field of science, that the programmer "must understand the science", and actually being able to write good code is very much a secondary requirement. I couldn't disagree more.

Re: [Beowulf] Stroustrup regarding multicore

"Robert G. Brown" <[EMAIL PROTECTED]> writes: >> 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. > > The problem here or anywhere is money. I'm reminded of an analog

Re: [Beowulf] Stroustrup regarding multicore

"Steve Herborn" <[EMAIL PROTECTED]> writes: > However, that being said I would think that it is usually easier to teach a > Scientist to code, then a coder the PhD level of the science. I think either is fine -- you wind up with someone who knows both. The problem is when you try to segregate t

Re: [Beowulf] Stroustrup regarding multicore

Matt Lawrence <[EMAIL PROTECTED]> writes: > Developing software is so much more than just coding. Well said, and too often not understood. Writing the code is the smallest part of the process. Perry ___ Beowulf mailing list, Beowulf@beowulf.org To chan

Re: [Beowulf] Stroustrup regarding multicore

As i'm typing as fast as RGB, though not at the same quality, yet perhaps with a wider scope, seems i'm allowed by the generosity of Greg to post this. With respect to your viewpoints on how scientific world works, it definately reminds me how the average guy might be. Between big extremes; o

Re: [Beowulf] Stroustrup regarding multicore

On Wed, 27 Aug 2008, Perry E. Metzger wrote: Having non-computer scientists specify what computer scientists should do is somewhat like having a layman tell a surgeon what to do to a patient without the surgeon being let in on what the underlying complaint was. The surgeon might expertly remove

Re: [Beowulf] Stroustrup regarding multicore

On Wed, 27 Aug 2008, Perry E. Metzger wrote: About 120WPM, though I might be faster at this point. I don't know how fast RGB types. I don't type. I have perfected a neural interface that slips into a small socket at the base of my brain. Unfortunately, this damaged some very important fronta

Re: [Beowulf] Stroustrup regarding multicore

On Wed, 27 Aug 2008, Greg Lindahl wrote: 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. Can I quote that? R

Re: [Beowulf] Stroustrup regarding multicore

Greg Lindahl wrote: > Spoken like a computer scientist. > > But this topic has slid off topic for this newsgroup. Don't you have > work to do? Or do you type as fast as rgb? > > -- greg I agree. This is akin to a religious war which no one will win. -- Prentice

Re: [Beowulf] Stroustrup regarding multicore

On Wed, 27 Aug 2008, Perry E. Metzger wrote: 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 pr

[Beowulf] Infiniband Subnet Manager

Since an infiniband fabric needs a subnet mananger, should the master node have an IB HCA and be connected to the IB network in order to run the subnet manager? My logic behind this is that the master node will be full enterprise-level hardware (redundant every thing), and should never go down or

Re: [Beowulf] Configuring Infiniband on Linux

Anand Vaidya wrote: > > 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