Well I know this is kind of old, but I am just discovering it and it seems like a good resource. It’s basically a series of videos put together by Microsoft to explain their DirectCompute API. If you’re not familiar with DirectCompute, it’s an API that allows you to write general purpose algorithms and run them on the GPU. Since the GPU is massively parallel, certain algorithms can run orders of magnitude faster than on the CPU.
Currently I am trying to evaluate if DirectCompute is a good option for building a GPU accelerated physics engine. So far I have watched the first 3 lectures, and it appears to be helpful. If you are interested in leveraging DirectCompute in your applications, you should definitely check out this series.