Give your project good CARMA, get a CUDA on ARM dev kit!

Subject: General Tech, Graphics Cards, Processors | December 20, 2011 - 04:34 AM |
Tagged: nvidia, CUDA, CARMA, capital letters, arm

Okay so the pun was a little obvious, but NVIDIA has just announced the specifications and name for the development kit used to develop for their ARM-based GPU computing platform. The development kit will provide a method to build and test applications on a platform similar to what will be found in the Barcelona Supercomputing Centre’s upcoming GPU supercomputer until you are ready to deploy the finished application with real data on the real machine. Such is the life of a development units.

View Full Size

Carma: What goes around, comes around... right Intel?

The development kit is quite modest in its specifications:

  • Tegra3 ARM A9 CPU
  • Quadro 1000M GPU (96 CUDA Cores)
  • 2GB system RAM, 2GB GPU RAM
  • 4x PCIe Gen1 CPU to GPU link
  • 1000Base-T networking support
  • SATA, HDMI, DisplayPort, USB.
 
While the specifications are somewhere between a high-end tablet and a modest workstation, the real story is the continued progress by NVIDIA into the High Performance Computing (HPC) market. NVIDIA seems to be certain that they are able to (ARM-)wrestle more market share from Intel and other players such as IBM on the high performance front. Many would probably speculate about NVIDIA’s crushing in towards the home market from both ends, but I expect that creating a compelling ARM product for a desktop PC will never be the problem for NVIDIA: it is a lack of anything compelling to run on it these days for a desktop user.
Source: NVIDIA Blogs
December 20, 2011 | 12:27 PM - Posted by Ben (not verified)

Would this be a good BitCoin generator? Maybe a stack of these in a corner would be pretty efficient.

December 20, 2011 | 02:03 PM - Posted by Scott Michaud

I think we've reached the point where the only efficient BitCoin generator is getting paid for some task or trade with a disproportionately high price.

December 20, 2011 | 12:57 PM - Posted by pdjblum

Scott,

Is it not the case that their HPC aims are centered around their GPU's, and not the Denver ARM project? I thought the latter was to develop ARM CPU parts for servers and desktops, not for HPC computing. Aren't they different things? Can you elaborate or clarify please?

December 20, 2011 | 01:59 PM - Posted by Scott Michaud

The brunt of the computing power would be from the GPUs of course. I should note I was never breifed from NVIDIA or anyone for this, so I am just going by press releases. I was under the impression that Tegra3 would be their development platform and they would at least slightly change the final architecture for the actual HPC environment. Apparently, my intuition might be wrong, it seems like the plan is mostly just to stack a bunch of them together.

Thanks for the comment.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.