Asus Releases Desktop-Sized Supercomputer

angry tapir writes "Asustek has unveiled its first supercomputer, the desktop computer-sized ESC 1000, which uses Nvidia graphics processors to attain speeds up to 1.1 teraflops. Asus's ESC 1000 comes with a 3.33GHz Intel LGA1366 Xeon W3580 microprocessor designed for servers, along with 960 graphics processing cores from Nvidia inside three Tesla c1060 Computing Processors and one Quadro FX5800."

Comments

Hrmm

[acehole]

How many pets would I have to eat to balance out the carbon footprint of this?

I've got a six-pack of kittens at the ready.

Re:Hrmm

[wisty]

The PSU is only 1100W. It's not that intensive - three teslas are like three big graphics cards. 2 or 3 kittens would be sufficient, so you've got enough to share.

Do you have pepper sauce?

Re:Boinc Applications...

[Anonymous Coward]

More importantly can this actually run Crysis 2? Probably not.

Super computer?

[MosesJones]

Ummm isn't this just a ridiculously powerful desktop computer rather than a super computer? The current 500th super computer on the top500 list is this machine which has a Rmax of 17 Tflops and an Rpeak of just over 37.6. Now its impressive that this desktop system has 1/37th of the power of the lowest machine on the super computer list... but does that really make it a super computer? Moore's Law says that it will take around 10 years for this desktop box to evolve to the power of that current bottom top500 box. So in other words its 10 years behind the performance of the current 500th best super computer.

If its because it hits 1 Tflops then in a few years time you'll have mobile phone "super computers" as Moore's Law is still moving onwards.

This is a very very fast desktop computer suited to certain simulation elements which are GPU intensive. Nice box, fast box.... but not a real modern super computer.

But how can you trust the results?

[HalfFlat]

The Tesla c1060 processor boards sound like a very efficient way of packing in compute power, but unless they're neglecting to mention it, the 4GB of GDDR3 RAM each has on board has no error correction. Given the rates of correctable errors observed e.g. here, I could never recommend using it for computing simulations that matter. A flipped bit in a floating point number can have a disproportionate affect on the outcome of calculations that rely upon it, and short of running the whole simulation a second or third time, one couldn't be confident that such an error did not occur.

Large compute-intensive simulations can take weeks, and are used to justify engineering and business decisions that involve the disposition of large amounts of money and other resources — it is important that the computational part of the process can be relied upon.

Re:But how can you trust the results?

[Anonymous Coward]

I'm a student at the University of Washington and once talked to a representative for Cray about using GPU's a a cheaper supercomputer and he told me that they generally have a nontrivial error rate. The issue with using ECC memory is that the GPU's are also libel for errors within their computations, making the ECC RAM pointless. A weird pixel in one frame of a game is no problem, but an error when performing a large simulation creates problems if the algorithm isn't designed to compensate for that noise.

Re:But how can you trust the results?

[mkaushik]

Then you would be happy to know that Nvidia's new Fermi chip supports ECC throughout the architecture.

Re:Eat a vacuum cleaner

[Fred_A]

>Pepper sauce? Pepper sauce?!? Do you have any idea what the carbon footprint of pepper sauce is? My brother ate pepper sauce once. He had to eat a whole zoo full of animals to make up for it! Stay away from the sauce!

But it's *green* pepper !