Matlab Integrate GPU Support for UberMath Computation

May 23, 2011
By The Architect

GPU computation has been a big event in the high-performace computation world for a year or so now. In case you’ve been living under a rock, GPU supercomputing is the use of the PC graphics card to handle a lot of highly parallel grunt work that multi-core GPUs don’t have the sillicon for.

This makes sense, as a GPU is basically a huge number of very simple CPUs strapped together in a low-latency parallel configuration. For many very parallel problems, such as Smooth Particle Hydronamics, this is ideal. Such simulations require high parallelism to compute simple calculations.

The technology harks back to the Connection Machine back in the 80s, a time when the world was pure and more simple.

http://www.mathworks.com/discovery/matlab-gpu.html

Anyway, the problem with GPUs is coding them up quickly. Parallel programming is a pain in the bum-hole due to the number of bugs that crop up and complexity in finding them. This means less time adding value with smart maths and more time tracing slip-ups.

However, things are changing as MathWorks have finally got round to enabling GPU parallelism support in the 2010b iteration of Matlab. It’s restricted to Nvidia’s CUDA platform right now, but so what, at least it’s there.

Introduction to GPU Computing with MATLAB

So now you can use Matlab without having to code all the slow bits in Fortran or C with CUDA . Here’s some prepared blurb on the features of this newishness:

MATLAB GPU computing capabilities include:

  • Data manipulation on NVIDIA GPUs
  • GPU-accelerated MATLAB operations
  • Integration of CUDA kernels into MATLAB applications without low-level C or Fortran programming
  • Use of multiple GPUs on the desktop (via the toolbox) and a computer cluster (via MATLAB Distributed Computing Server )

Examples and How To

2 Responses to Matlab Integrate GPU Support for UberMath Computation

  1. John Melonakos on May 23, 2011 at 5:46 pm

    Hey, you might want to add another blurb about Jacket which is way better than the PCT-GPU stuff from MathWorks. Learn more here: http://www.accelereyes.com/products/compare.

    BTW, that MATLAB+Tesla image comes from us at AccelerEyes. I put it together using Powerpoint back in 2007, back when we were partners with MathWorks before they decided that Jacket was getting too good for their comfort. Since they don’t have a technical leg to stand on, they exerted their legal forces to make us remove all “L-shaped membrane” logos from our website and that image came down. :)

  2. James Smith on May 23, 2011 at 10:47 pm

    The video doesn’t work. If you could also put up a WebM version that would be great.

Leave a Reply

Your email address will not be published.