omp-critical pattern experiment

The master code was run on a single 48-core node of MareNostrum-4 at Barcelona Supercomputing Center for a range of thread counts from 1 to 48. The problem size was specified by Ni=500, Nj=500, Nk=500 and Nt = 20. The run times and the average IPC (Instructions Per Cycle) are shown here. Initially, the run times are increasing with the thread count and the average IPC value is falling rapidly. These values level off from around 8 threads.