New Tesla GPU Smashes World Record, Says Nvidia

Tuesday was a big day for Nvidia, as the company announced that its new Tesla M2090 GPU (opens in new tab) is now rolling out into server configurations including HP's ProLiant SL390 G7 4U server (opens in new tab). The company also boasted that the GPU has achieved the fastest-ever performance in a key measure of scientific computation.

"In the latest version of AMBER 11, one of the most widely used applications for simulating behaviors of biomolecules, four Tesla M2090 GPUs coupled with four CPUs delivered record performance of 69 nanoseconds of simulation per day," Nvidia said in a press release. "The fastest AMBER performance recorded on a CPU-only supercomputer is 46 ns/day."

Tagged as the world's fastest parallel processor for high performance computing, the beastly Tesla M2090 comes packed to the teeth with 512 CUDA parallel processing cores, delivering 665 gigaflops of peak double-precision performance. Other features include a peak single precision floating point performance of 1331 gigaflops, up to 6 GB of GDDR5 memory, and a memory bandwidth of 177 GB/s (with ECC memory error protection off).

In the SL390 G7 4U server, up to eight Tesla M2090 GPUs can be crammed into the half-width 4U chassis. Coupled with two CPUs, HP's new server rig is a monster powerhouse best suited for applications ranging from "quantum chemistry and molecular dynamics to seismic processing and data analytics."

In addition to the new Tesla GPU, Tuesday Nvidia also announced (opens in new tab) the GeForce GTX 560 GPU -- the "little brother" of the previously-released Ti version -- and the release of the new beta GeForce R275 drivers.

According to Nvidia, these new drivers will bring increased performance and enhanced functionality to a "broad spectrum of PC games." This means 3D Vision support for Duke Nukem Forever, PhysX support to Alice: Madness Returns, and Surround support to Dungeon Siege III. Crysis 2 will get a 6-percent performance boost, Bulletstorm a 15-percent boost, and Portal 2 an 8-percent boost, among others.

As for the new GTX 560 GPU, cards will start at $199 USB and available from your favorite manufacturers including ASL, Asus, Colorful, ECS, EVGA, Gainward, Galaxy, Gigabyte, Innovision 3D, Jetway, Leadtek, MSI, Palit, Point of View, PNY, Sparkle, Zotac and others.

Did you catch that? The PC version of Duke Nukem Forever will support Nvidia 3D Vision! Check out the company's teaser trailer pasted below!

  • acadia11
    Yes. The answer to the ultimate question.
    Reply
  • internetlad
    Reply
  • JerseyFirefighter
    dont they break this record every time they release a new product? **yawn**
    Reply
  • JerseyFirefighter
    ^^in the video^^ is that Bill from Bill and Ted..all grown up?
    Reply
  • dread_cthulhu
    $199 USB

    So... I can trade $199 worth of USB devices for this? That would be pretty cool...

    XD
    Reply
  • kalogagatya
    is that salt, pepper, and olive oil??

    so they can use it on whatever he is grilling on those cards? :D
    Reply
  • Marthian
    here's some clarification: the M090 gpu is business oriented, they also say "the fastest-ever performance in a key measure of scientific computation." not video games, and as for such tasks will cost an arm and a leg (which to a business that uses this kind of technology won't really be a big deal).
    as for the GTX 560, I already got a GTX 560 Ti, so no thanks.
    Reply
  • 11796pcs
    I know there are different drivers for professional and comsumer based graphics solutions but this M2090 GPU looks like if given the appropriate drivers could crush a 6990. 6GB of DDR5? Holy crap. Is there something special about how professional cards are build vs consumer based that makes them impossible for playing games? Any clarification would be nice.
    Reply
  • PreferLinux
    Marthianhere's some clarification: the M090 gpu is business oriented, they also say "the fastest-ever performance in a key measure of scientific computation." not video games, and as for such tasks will cost an arm and a leg (which to a business that uses this kind of technology won't really be a big deal).as for the GTX 560, I already got a GTX 560 Ti, so no thanks.11796pcsI know there are different drivers for professional and comsumer based graphics solutions but this M2090 GPU looks like if given the appropriate drivers could crush a 6990. 6GB of DDR5? Holy crap. Is there something special about how professional cards are build vs consumer based that makes them impossible for playing games? Any clarification would be nice.It is not a graphics card. It has no video outputs at all. It uses a GPU for GPGPU only.
    Reply
  • dragonsqrrl
    That's a pretty significant boost in hardware specs and performance off the M2070. Nvidia doesn't seem to specify the GPU used, but I wonder if this is the first non-gaming card based on gf110 instead of gf100. The specs seem to lend themselves toward this conclusion, with all 512SP active. I suppose a massive under-clock of gf100 could have allowed them to accomplish this as well, but this seems unlikely based on theoretical FP64/32 performance. There's roughly a 14% increase in shader horsepower going from the M2070 to the M2090, but there's nearly a 30% increase in FP64/32 performance.

    It would be absolutely awesome if this is in fact based on gf110, because it also means higher performance Quadro cards could be on the way as well.
    11796pcsI know there are different drivers for professional and comsumer based graphics solutions but this M2090 GPU looks like if given the appropriate drivers could crush a 6990. 6GB of DDR5? Holy crap. Is there something special about how professional cards are build vs consumer based that makes them impossible for playing games? Any clarification would be nice.There are both hardware and driver differences between the Tesla and Geforce cards. What defines these differences is the intended purpose for the GPU, which in the case of the Tesla series is almost exclusively GPGPU computing. You'll notice that the Tesla M2000 series doesn't have any monitor output ports because of this, it's meant for installation in rack-mount blade systems and servers.

    Even if you could run games off this card, I can guarantee you it wouldn't outperform the HD6990. In fact the GTX580 would probably outperform it without a problem. The hardware is tweaked and geared towards its intended purpose. What makes a great high performance gaming card is very different from what makes a great high performance compute card. Raw clock speeds and mem bandwidth isn't as important, so these are often lower then their Geforce counterparts. Stability and data integrity are paramount. In terms of GPU architecture one of the most significant differences is FP64 performance, which is severely crippled on gf100/gf110 based Geforce cards (1/8 FP32, as opposed to 1/2 on Tesla and Quadro cards).

    And as you've already pointed out, mem capacity is much higher. Going from 1.5GB to 6GB wouldn't benefit a Geforce card very much, in fact it would probably result in a performance hit due to the lower mem bandwidth, but it can be extremely important when working with large data sets. There's also ECC support, and a number of other hardware differences, but that basically sums it up.

    Software and drivers are completely different. Again, entirely different purpose. No game profiles or any of the things you're used to with gaming cards, so again even if you could run games off a Tesla M2090 it would probably run like crap.
    Reply