General-Purpose Computation on Graphics Hardware
Those of you following the various parallel computing blogs from our team or who have played with Visual Studio 2010 have probably noticed a heavy focus on single box parallelism (and mostly on client machines).
For a future version of Visual Studio we are exploring stronger investment for the Windows HPC Server (for an example of what we've done for VS2010 see here and here). We are also considering exploiting the compute power of modern graphics cards. For relevant background reading, please visit the posts on my blog: GPGPU and DirectCompute.