Sign in with
Sign up | Sign in

$5K PC Takes On $4.6M Supercomputer

By - Source: Tom's Hardware US | B 35 comments

 

Antwerp (Belgium) - Recent advances in general purpose GPU computing are beginning to shift perceptions in supercomputing applications. Belgian researchers have assembled a relatively simple enthusiast PC with an emphasis on graphics processing capability, which beats a multi-million dollar supercomputer in its target application.

The desktop PC, called Fastra, was built with a focus on the development of new computational methods for tomography. Tomography is a technique used in medical scanners to create three-dimensional images of the internal organs of patients, based on a large number of x-ray photos that are acquired over a range of angles. As these 3D images can be quite large, advanced reconstruction techniques can sometimes require weeks of computation time on a regular PC. Which means that supercomputers are usually required to process computer tomography (CT) images.

The research group Vision Lab at the University of Antwerp came up with a different solution and constructed a PC that integrates four GeForce 9800 GX2 graphic cards (with a total of eight GPU cores) that runs CUDA-optimized tomography applications. The specifications include a MSI K9A2 Platinum motherboard, an AMD Phenom 9850 CPU, 4 x 2 GB Corsair TWINX DDR2 PC6400 memory, a Samsung Spinpoint F1 750 GB hard drive, a Thermaltake Toughpower 1500W Modular power supply unit as well as four MSI 9800GX2 cards. The researchers said that the cost of the system was less than 4000 Euro or about $5300.

It isn’t quite a tricked-out gaming system and the 3DMark06 score is just above what your average PC can manage to come with today (12,603 points). However, it is the CUDA application where this PC really shines. Compared to the 512-processor, $4.6M CalcUA supercomputer purchased in 2005, the PC can be more than a match : The projection of image slices took 23.4 seconds on the supercomputer and 35.1 seconds on the PC. The reconstruction of the slices was displayed after 67.4 seconds on the supercomputer systems and after just 52.2 seconds on the PC. The Vision Lab crew now believes that a real-time construction is possible through GPUs and is now building a cluster of such systems.

While it is an impressive example how GPUs can be applied in non-traditional ways, there are a few notes to be added. Of course, GPUs cannot replace traditional supercomputers, which still can be applied to applications with a broader range. Also, supercomputers usually carry huge memories, often in the Terabyte range, which cannot be matched by today’s GPU clusters. When we talk to scientists working with supercomputers and GPUs, they typically believe that future supercomputers will not completely transition to GPU clusters, but may develop into systems that consist of a traditional supercomputer structure as well as GPU capability.

An interesting side note about the Fastra PC is its motherboard. Eagle-eyed readers may have noticed that the MSI K9A2 Platinum board is not an Nvidia SLI-based board, but uses AMD Crossfire (780 chipset). The simple reason to choose this board may have been cost, but it is unlikely to impact the performance of the system : CUDA does not support SLI at this time, which means that the GPUs have to communicate with each other as well as with the CPU via PCI Express. The researchers claim that they have not seen any impact on performance and the GPUs apparently are scaling well.

Discuss
Ask a Category Expert

Create a new thread in the News comments forum about this subject

Example: Notebook, Android, SSD hard drive

This thread is closed for comments
  • 0 Hide
    Blackopsninja , May 30, 2008 6:23 PM
    sweet
  • 2 Hide
    dittopb , May 30, 2008 6:29 PM
    I think the reason why they choose MSI K9A2 Platinum board is the PCIe slot spacing. I checked Asus and Gigabyte boards with quad PCIe 2.0 support and found to be unevenly spaced and would be hard to fit 4 dual-slotted GPUs.
  • 1 Hide
    hixbot , May 30, 2008 6:42 PM
    Ok, now we know how well 8 GPU cores handle supercomputer-tasks. Lets see how the $4.6M supercomputer can handle Crysis!
  • Display all 35 comments.
  • 0 Hide
    hughyhunter , May 30, 2008 6:56 PM
    It's about time someone realizes we build for shear power!!!
  • 0 Hide
    lopopo , May 30, 2008 8:13 PM
    It's about time someone other than folding did this. Only hope others will do the same.
  • -3 Hide
    dobby , May 30, 2008 10:13 PM
    interesting choice of CPU, i know that this doesnt use the CPU directly for it Processing power, but you would think that they would go the full hog and use a skull trial system
  • 0 Hide
    khaydin , May 30, 2008 11:26 PM
    8GB of ram on Windows XP? Hopefully it's 64bit XP otherwise they're wasting more than half of that ram...
  • 0 Hide
    customisbetter , May 30, 2008 11:43 PM
    how does one, let alone 4 9800 gx2s get 12k in 3dmark?
  • 0 Hide
    hughyhunter , May 31, 2008 12:29 AM
    Quote:
    how does one, let alone 4 9800 gx2s get 12k in 3dmark?

    I agree... I'm interested to hear a reason why. Also... I didnt know there was support for four GX2's. How do you run 4 of them? Their is only 1 connector!
  • 0 Hide
    khaydin , May 31, 2008 3:24 AM
    hughyhunterI agree... I'm interested to hear a reason why. Also... I didnt know there was support for four GX2's. How do you run 4 of them? Their is only 1 connector!


    They're not setup in SLI since the motherboard and CUDA doesn't support SLI. So if they were doing 3dmark only 1 of those 4 would be getting benched.
  • 0 Hide
    navvara , May 31, 2008 3:45 AM
    customisbetterhow does one, let alone 4 9800 gx2s get 12k in 3dmark?


    They are not running SLI. They are just plugged in the PCI-X and accessed by the app indepentently.

  • 0 Hide
    drho , May 31, 2008 5:35 AM


    Higher memory bandwidth with the AMD versus Intel until Intel implements a real memory bus. CPU memory bandwidth to GPU memory can be a significant factor.
  • 0 Hide
    romulus47plus1 , May 31, 2008 5:36 AM
    Yea! Fastra!
  • -2 Hide
    apache_lives , May 31, 2008 7:11 AM
    gawd MSI, atleast the $4.6M will last longer then a few months :D 

    skulltrail with the quad channel memory would have been a wiser and more reliable choice although the cost would have become alot higher, and when CUDA starts to support SLi, perhaps the motherboard choice would come back to bite them.
  • 0 Hide
    martel80 , May 31, 2008 9:35 AM
    navvaraThey are not running SLI. They are just plugged in the PCI-X and accessed by the app indepentently.

    PCI-X is something completely different from PCI-express.
    http://en.wikipedia.org/wiki/PCI-X

    And if CUDA doesn't support SLI, I wonder if this computer even uses both parts of the GX2 since this single card already runs in SLI.
  • 0 Hide
    Anonymous , May 31, 2008 12:13 PM
    It can. CUDA API will reports 2 GPUs presence when a 9800 GX2 is attached.
  • 0 Hide
    navvara , May 31, 2008 1:43 PM
    ismailfaruqiIt can. CUDA API will reports 2 GPUs presence when a 9800 GX2 is attached.


    My point exactly. The reason they use Nvidia video cards on intel chipsets is the extra pci-x (or express who cares about wording) lanes. CUDA supports multiple GPU's (it will detect all compatible Nvidia GPU's attached to a system) but not SLI.

    Although this news is interesting I wonder why they didn't use tesla? It would certainly scale better with tesla solutions being rack mountable (though more expensive to get the same crunching power).
  • 0 Hide
    wild9 , May 31, 2008 6:31 PM
    Quite an astonishing story, in my opinion. A super-computer in your home..now just imagine a whole planet full of these systems working on cancer cures. I'm sure there are some applications that need more memory than a desktop can handle, but this is astounding.
  • -1 Hide
    hughyhunter , May 31, 2008 6:35 PM
    Quote:
    CUDA supports multiple GPU's (it will detect all compatible Nvidia GPU's attached to a system) but not SLI.

    I dont get it... what's the difference?
  • 0 Hide
    Jaevric , May 31, 2008 9:55 PM
    Is it bad that my first thought was "I hope they bought EVGA so they can step-up in a couple of months?"

    Still, definitely really cool.
Display more comments