I want to solve Poisson problem in parallel. I use red-black ordering in order to introduce parallelism.
Here is the "Red" part of the code:
I am trying to use MKL libraries and OpenMP in a MSVS C++ application on Windows7. The application shows affinity for all 24 processors (2 nodes, 6 processors, HyperThreaded). omp_get_num_procs() also shows 24 processors. When I run the program only 1 node and 6 processors are accessed. This is confirmed when I use "KMP_AFFINITY=verbose,none". It ouputs "OMP: Info #179: KMP_AFFINITY: 1 packages x 6 cores/pkg x 1 threads/core (6 total cores)". I get no compiler or linker complaints.
In one of our programs, we have this piece of code:
Ziya Ma, Director of Big Data Technologies at Intel, stops by to talk about how open source solutions are enabling enterprise to take advantage of the new concepts coming out of big data. She highlights how Intel is a leading contributor within the overall open source community and is working to accelerate the delivery of different vertical analytics solutions.
Does the Intel compiler/ OpenMP runtime lib. support OpenMP 4.0 for accelerators?