Take a parallel Bandits algorithm with the problem it was used on, implement it in OMP, and run on different processors in different configurations (num. of CPUs, num. of threads, hyper-threading, AVX, etc.). We aim to benchmark trade-offs in running time and energy used and describe them in 4 pages paper.
.
├── README.md (This file.)
├── doc (Papers write-ups, presentations, experiments results, etc.)
├── etc (Other resources e.g. notes, papers, lectures, etc.)
└── src (All the code lives here.)