Skip to content
Snippets Groups Projects
  1. Jun 05, 2016
  2. May 11, 2016
  3. Nov 05, 2015
  4. Aug 09, 2015
  5. Jun 30, 2015
  6. Jun 21, 2015
  7. Jul 17, 2014
  8. Jun 15, 2014
  9. May 29, 2014
  10. Apr 29, 2014
  11. Oct 17, 2013
  12. Sep 16, 2013
  13. Aug 17, 2013
  14. Jul 29, 2013
  15. Jul 07, 2013
  16. Jul 06, 2013
  17. Jun 10, 2013
  18. Mar 20, 2013
  19. Oct 24, 2012
    • dhj's avatar
      Updated benchmark to use equivalent work loads for cpu and gpu. · fa833504
      dhj authored
      * Numpy does element-wise operations by default.  Updated the cpu
        operation to use pure numpy.
      
      * Eliminated the loop which is not necessary to demonstrate
        parallelism on array operations.
      
      * Made the number of workers explicit rather than gpu chosen,
        through local_size variable passed to kernel execution.
      
      * Increased to ~8 million data points to more clearly demonstrate
        the difference between cpu and gpu based computations.
      fa833504
  20. Oct 06, 2012
  21. Aug 11, 2012
  22. Jul 30, 2012
  23. Jul 25, 2012
  24. Jun 18, 2012
  25. Dec 07, 2011
  26. Aug 22, 2011
  27. Aug 08, 2011
  28. Jul 04, 2011
  29. May 07, 2011
  30. Apr 29, 2011
  31. Apr 17, 2011
  32. Feb 25, 2011
  33. Feb 18, 2011
  34. Feb 17, 2011
  35. Jan 19, 2011
Loading