GROMACS ON GPUs
GPUs have the possibility to perform an enormous number of Floating Point OPerations (FLOPs). However, they achieve this high performance by losing generality -- there are only certain types of calculations which would be well-suited to GPUs. The calculations in FAH could possibly take advantage of GPUs and we have been investigating this with our collaborators Prof. Pat Hanrahan (Stanford University, Computer Science Dept) and Prof. Eric Darve (Stanford University, Mechanical Engineering Dept) and their groups. Prof. Hanrahan's group has pioneered methods for programming GPUs with the Brook streaming computer language and Prof. Darve's group has written many applications using Brook. It is unclear which GPUs will be supported and we don't want to specific list any GPUs at the moment. However, it will likely require the very latest GPUs from NVIDIA or ATI. There is also a possibility that only NVIDIA GPUs will work (they use 32 bit floating point, whereas ATI uses 24 bit).
July 2005 We have a working version of GROMACS on Brook, but are tuning performance.
August 2005 Vishal has made great progress in rewriting the GROMACS inner loops in order to take advantage of more memory. Now, we're tweaking for performance.
September 2005 Vishal has found that the new changes aren't quite there yet. We're also benchmarking/testing/evaluating new hardware that has come out.
October 2005 With the release of the R520 card, it looks like we're getting close. Our unoptimized code is already looking good and we are working to optimize it.
December 2005 We are working on optimizing the newest code. We are seeking help from GPU manufactures to tune our code on their hardware.
