OpenMP program is slower than sequential one

by

The random number generator rand(3) uses global state variables (hidden in the (g)libc implementation). Access to them from multiple threads leads to cache issues and also is not thread safe. You should use the rand_r(3) call with seed parameter private to the thread:

long i;
unsigned seed;

#pragma omp parallel private(seed)
{
    // Initialise the random number generator with different seed in each thread
    // The following constants are chosen arbitrarily... use something more sensible
    seed = 25234 + 17*omp_get_thread_num();
    #pragma omp for
    for (i = 0; i <= 1000000000; i++) {
       double x = rand_r(&seed);
    }
}

Note that this will produce different stream of random numbers when executed in parallel than when executed in serial. I would also recommend erand48(3) as a better (pseudo-)random number source.

More Related Contents:

  1. Is it worth using parallel OpenMp sort with an array of 4000 elements? [closed]
  2. Optimal number of threads per core
  3. Reducing on array in OpenMP
  4. Cholesky decomposition with OpenMP
  5. How does the omp ordered clause work?
  6. Difference between section and task openmp
  7. Measure execution time in C++ OpenMP code
  8. MPI: blocking vs non-blocking
  9. Julia: How to copy data to another processor in Julia
  10. No performance gain after using openMP on a program optimize for sequential running
  11. Processing vec in parallel: how to do safely, or without using unstable features?
  12. Accelerating bitmap grayscale conversion, is OpenMP an option in C#?
  13. GNU parallel not working at all
  14. What is the purpose of the “PAUSE” instruction in x86?
  15. Reductions in parallel in logarithmic time
  16. OpenMP and Python
  17. How does OpenMP handle nested loops?
  18. OpenMP: What is the benefit of nesting parallelizations?
  19. omp parallel vs. omp parallel for
  20. openMP nested parallel for loops vs inner parallel for
  21. multiple targets from one recipe and parallel execution
  22. How to parallelize this array sum using OpenMP?
  23. Why is the != operator not allowed with OpenMP?
  24. Is there a good way to extract chunks of data from a java 8 stream?
  25. Sharing a result queue among several processes
  26. OpenMP time and clock() give two different results
  27. Multithreaded & SIMD vectorized Mandelbrot in R using Rcpp & OpenMP
  28. MPI_Type_create_subarray and MPI_Gather
  29. .net Core Parallel.ForEach issues
  30. Tensorflow and Multiprocessing: Passing Sessions

Leave a Comment