No performance gain after using openMP on a program optimize for sequential running

by

Even if your program benefits from using OpenMP, you won’t see it because you are measuring the wrong time.

clock() returns the total CPU time spent in all threads. If you run with four threads and each runs for 1/4 of the time, clock() will still return the same value since 4*(1/4) = 1. You should be measuring the wall-clock time instead.

Replace calls to clock() with omp_get_wtime() or gettimeofday(). They both provide high precision wall-clock timing.

P.S. Why are there so many people around SO using clock() for timing?

More Related Contents:

  1. In an OpenMP parallel code, would there be any benefit for memset to be run in parallel?
  2. Cholesky decomposition with OpenMP
  3. Difference between section and task openmp
  4. Reductions in parallel in logarithmic time
  5. How to parallelize this array sum using OpenMP?
  6. optimized sum of an array of doubles in C [duplicate]
  7. Please explain the following parallel code template
  8. Improve INSERT-per-second performance of SQLite
  9. sending blocks of 2D array in C using MPI
  10. Trying to understand gcc option -fomit-frame-pointer
  11. parallel prefix (cumulative) sum with SSE
  12. Why is this code using strlen heavily 6.5x slower with GCC optimizations enabled?
  13. Matrix multiplication: Small difference in matrix size, large difference in timings
  14. Is it faster to count down than it is to count up?
  15. What is the fastest way to swap values in C?
  16. Efficient 4×4 matrix vector multiplication with SSE: horizontal add and dot product – what’s the point?
  17. Measure execution time in C++ OpenMP code
  18. Understanding the collapse clause in openmp
  19. Accelerating bitmap grayscale conversion, is OpenMP an option in C#?
  20. When is assembly faster than C? [closed]
  21. AVX/SSE version of xorshift128+
  22. Why is Parallel.ForEach much faster then AsParallel().ForAll() even though MSDN suggests otherwise?
  23. Best timing method in C?
  24. How does the CPU cache affect the performance of a C program
  25. why does GCC __builtin_prefetch not improve performance?
  26. Concatenating strings in C, which method is more efficient?
  27. Faster approach to checking for an all-zero buffer in C?
  28. OpenMP: What is the benefit of nesting parallelizations?
  29. Why is the != operator not allowed with OpenMP?
  30. How much overhead can the -fPIC flag add?

Leave a Comment