Benchmarks Show the Advantages of GPU Flexibility on Application Performance

In Part V of his series of features on heterogeneous computing performance, Dell’s Dr. Jeff Layton uses a pair of popular benchmarks to illustrate why GPU systems need to be as flexible as possible.

“These two benchmarks illustrate that performance is dependent upon the specific application. In the case of NAMD, you get a bit better performance when each GPU has a dedicated x16 slot. However, in the case of CUDASW++ having a dedicated x16 for each GPU is not necessary and you can still get good performance when two GPUs share a single x16 slot.

Read the Full Story or check out Layton’s presentation video and slides from the recent HPC Advisory Council European Workshop.