Best offers
|
Edge Z30 Midsize Desktop (2.66GHz... | $1099.00 Velocity Micro More info |
|
iMac All-In-One Desktop (3.06GHz... | $1699.00 ABT More info |
|
Pavilion p6210f Mini-Tower Desktop... | $599.98 STAPLES More info |
|
Compaq Presario CQ5210F Mini-Tower... | $399.98 STAPLES More info |
|
iMac All-In-One Desktop (3.06GHz... | $1169.00 MacConnection More info |
Part 1: Building A Balanced Gaming PC
What does it mean to build a truly-balanced PC? How great would it be to piece together a machine bottlenecked by neither CPU or GPU? We set forth to measure the perfect balance in seven different games and four resolutions in this first of several parts. Read More
-
System Builder Marathon, Sept. '09: AMD System Value Compared
Our response to reader-demand for AMD systems focuses on the company’s penchant for gaming value across three budget classes. How do these fully-optimized systems compare to each other in ultimate performance and value? Read More
-
System Builder Marathon, Sept. '09: $2,500 Performance PC
Reader suggestions and previous test results defined most of this month’s highest-priced build. Will the extra planning and testing pay off in clear overclocking and performance superiority? We use Radeon HD 4890s, SSDs, and 8GB of RAM to find out! Read More
- sli cuda
- laptop amd 780 chipset support memory
- k9a2 platinum power supply
- k9a2 platinum and 9800gx2
- samsung 750 spinpoint f1
- building graphics processing units supercomputer
- amd chipset 780
- amd 780 chipset drivers
- best cuda applications
- k9a2 platinum memory
- k9a2 sli platinum
- power supply for msi k9a2 platinum
- phenom 9850 cpu temp
- graphic card as supercomputer
- case power supply 1500w
Partners
The Games selection
kids :
Bob
Throw bubbles so as to make the ones that appear in the game disappear. For this, use the Right / Left arrow keys to duck or move about, and the...
|
crazy :
PC Breakdown
What is worst than a Fatal Error occuring during a game you did not save? Unleash your rage at your PC in this game. Blow it to pieces, it feels so...
|
Sponsored links
$5K PC Takes On $4.6M Supercomputer
Next news- Email |
- Print |
- Comments (35) |
- Share
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.
Source : Tom's Hardware US
Sponsored links
Related forums topics
- AMD & ATI making a come back?
- AMD dominates Supercomputers, used in more systems!!!
- Isnt the final or next true step for cpu's organic?
- AMD Phenom FX Deneb Core - Kentsfield Beater
- Intel's 'Overspeed protection' will limit Core i7 overclocks!!!!!!
- Most Overclockable Processor?
- Optical Motherboards
- AMD dead or Alive
- Diamond based CPUs? (81Ghz??)
- Fluid Flight Sim PC and more. Help please on best equipment
- quick stupid question
- Please Advise! Cas Latency confusion! :(
- The Scientists\' Opinions on Gaming Physics
- I need some help [non-hardware]
sweet
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.
Ok, now we know how well 8 GPU cores handle supercomputer-tasks. Lets see how the $4.6M supercomputer can handle Crysis!
It's about time someone realizes we build for shear power!!!
It's about time someone other than folding did this. Only hope others will do the same.
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
8GB of ram on Windows XP? Hopefully it's 64bit XP otherwise they're wasting more than half of that ram...
how does one, let alone 4 9800 gx2s get 12k in 3dmark?
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!
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!
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.
how 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.
Yea! Fastra!
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.
gawd MSI, atleast the $4.6M will last longer then a few months
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.
They 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.
It can. CUDA API will reports 2 GPUs presence when a 9800 GX2 is attached.
It 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).
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.
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?
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.
We as enthusiasts should be proud that our constant upgrading in search of ultimate gaming performance has funded the development of consumer affordable technology so advanced that it can benefit important applications such as this.
The is very interesting.
Intel and AMD are working hard now to make it easier to program to their multi-core monsters, but right now Nvidia has the upper hand.
Lars
Well it is very interesting to know the facts about this Fastra motherboard
I believe there are quite few other board out there that haven't been tested for various combination. Is impresive how technology has evolved over time. With such tech so advance we might some day have a Super Desktop PC in our Home.
Why the inherent bias against AMD? They chose AMD to save money? So instead of paying an outrageous $5500 for a Core2Quad system, they paid a more reasonable $5300 for a Phenom CPU + mobo? LMAO. As the above poster said, AMD have had an advantange on memory and system bandwidth ever since the day the Athlon 64 first came out. This is exactly why there are so many more supercomputers running Opterons than Xeons...
This disussion it not so much about CPU architecture as it is using the GPU obvious vector adavntages over the CPU.
I agree that AMD firestream API might be a little smarter at using CPUs and GPUs for the same job.
Looking forward to some major progress in this field really soon.
Lars
As someone already pointed out why did they not buy a Tesla
e.g. NVIDIA Tesla™ D870 $5000
Achieve up to 700 GFLOPS of performance (1 TeraFLOP peak) with both C870 GPUs
As someone already pointed out why did they not buy a Tesla
e.g. NVIDIA Tesla™ D870 $5000
Achieve up to 700 GFLOPS of performance (1 TeraFLOP peak) with both C870 GPUs
They wanted to compare consumer level (gaming)... The average joe doesnt buy a Tesla... They buy 8800GT's.
Why the inherent bias against AMD? They chose AMD to save money? So instead of paying an outrageous $5500 for a Core2Quad system, they paid a more reasonable $5300 for a Phenom CPU + mobo? LMAO. As the above poster said, AMD have had an advantange on memory and system bandwidth ever since the day the Athlon 64 first came out. This is exactly why there are so many more supercomputers running Opterons than Xeons...
Well, the reason in this case is... currently only MSI K9NA2 Platinum -which is an AMD mainboard- that has 4 PCI-express 16x slots with spacing evenly (2 slots each), which allows 4 9800 gx2 combination. The Intel platform only has 3 PCI-express slots at most (either nvidia or skulltrail).
As someone already pointed out why did they not buy a Teslae.g. NVIDIA Tesla? D870 $5000Achieve up to 700 GFLOPS of performance (1 TeraFLOP peak) with both C870 GPUs
Of course for price reason. With 4x 9800 gx2, this $5300 system has theoretical peak of 4 TFLOPS, compared to $5000 Tesla one.
Quad CrossFire (8X+8X+8X+8X)}; 4X per GPU
Do you think there's enough bandwidth to feed the GPUs?
Also ATI seems to be double precission floating points where as Nvidia does not. So it sucks for Monte Carlo simulations.
Lars
[QUOTE]Well, the reason in this case is... currently only MSI K9NA2 Platinum -which is an AMD mainboard- that has 4 PCI-express 16x slots with spacing evenly (2 slots each), which allows 4 9800 gx2 combination. The Intel platform only has 3 PCI-express slots at most (either nvidia or skulltrail).[/QUOTE]
But the article suggests that it was chosen to save an amount of money equal to < 5% of the total cost of the system, which is just absurd... Aside from the additional PCI-E slot, AMDs integrated memory controller more than makes up for the nominal loss of IPC. Regardless, I have no doubt that there was some serious thought put into the components the team used, and that cost was not a factor.