Locality-Aware Task Scheduling and Data Distribution for OpenMP Programs on NUMA Systems and Manycore Processors

Listing 1

Parallel map implemented using OpenMP tasks.

/* Allocate and initialize data */

for(int i=0; i<N; i++) {

    list[i] = malloc(sizeof(int) * SZ);

    initialize(i, list[i], SZ);

/* Work in parallel */

for(int i=0; i<N; i++) {

#pragma omp task input(list[i][0:SZ-1])

    map(list[i], SZ);

#pragma omp taskwait