Exploit Low CPU Usage

How can we use extra CPU available for using threaded implementation of MPI?

A. Overlapping communication/computation
Paper Idea: Quantify this

B. Virtual Machines: Creating large virtual testbeds

C. ?

Graphics Processing Units

Assuming GPU can positively affect prince/performance curve of HPC machines, what the best way to use it?



How do we offload compute kernels onto GPU?
How to abstract GPU from the user?
Same questions for FPGA's too.

To start with run NAS benchmarks on GPU.
Or run HPL on GPU.

Q. How to use thread contention in multi-threading MPI apps to offload data to GPU?