0 Replies Latest reply on Feb 12, 2016 12:25 PM by juaninf

    Intel Xeon Phi performance doesnot work?

    juaninf

      I have a processor Xeon which has a co-processor Xeon Phi. I would like to know if vectorization on Xeon Phi has better or equal perfomance to the processor Xeon. For that I run this code

       

       

           #include <stdio.h>

           #include <omp.h>

           //usingnamespace std;

           int main() {

           const int SIZE = 300000;

           int x[SIZE];

           int y[SIZE];

           int z[SIZE];

           // Initialize loop

           int i;

           for (i=0; i < SIZE; i++){

             x[i] =i;

             y[i] = i * 2;

             z[i] = 0;

           }

           // // Main loop

           #pragma omp simd

           for (i=0; i < SIZE; i++){

              z[i] =x[i] + y[i];

           }

           return 0;

          }

       

      I am compiling that using with mpi and with flags qopenmp and openmp in this way:

       

           $ mpiicc -o example_openmp.mic example1.c -lm -openmp -mmic

           $ mpiicc -o example_qopenmp.mic example1.c -lm -qopenmp -mmic

           $ mpiicc -o example_openmp example1.c -lm -openmp

           $ mpiicc -o example_qopenmp example1.c -lm -qopenmp

       

      The times of the four executables are

       

           $ time ./example_openmp.mic

       

           real    0m0.151s

           user    0m0.000s

           sys    0m0.030s

       

           $ time ./example_qopenmp.mic

       

           real    0m0.033s

           user    0m0.010s

           sys    0m0.010s

       

           $ time ./example_openmp

       

           real    0m0.008s

           user    0m0.000s

           sys    0m0.005s

       

           $ time ./example_qopenmp

       

           real    0m0.007s

           user    0m0.001s

           sys    0m0.003s

       

      In other words, run the executables in mic has a bad performance respect to the host Xeon. My question is How Do I get the efficiency of the intel coprocessors?