Power-Efficient Computing: Experiences from the COSA Project

Figure 11

MPI processes versus OpenMP threads performance on a single whole node.