1. 03 Jan, 2018 1 commit
  2. 01 Dec, 2017 1 commit
  3. 28 Nov, 2017 1 commit
  4. 27 Nov, 2017 1 commit
  5. 26 Oct, 2017 1 commit
    • Lorenz Huedepohl's avatar
      First implementation of an autotuning procedure · 0ebe4802
      Lorenz Huedepohl authored
      To be used like this
      
         class(elpa_t), pointer      :: e
         class(elpa_autotune_t), pointer :: tune_state
      
         e => elpa_allocate()
         call e%set(...)
         [...]
         assert_elpa_ok(e%setup())
      
         tune_state => e%autotune_setup(ELPA_AUTOTUNE_FAST, ELPA_AUTOTUNE_DOMAIN_REAL)
      
         ! Autotuning loop, continues until all combinations have been tried
         do while (e%autotune_step(tune_state))
           ! Do the steps that are representative of your calculation
           call e%eigenvectors(a, ev, z, error)
         end do
      
         ! Fix best parameters, and de-allocate the autotune object
         call e%autotune_set_best(tune_state)
         call elpa_autotune_deallocate(tune_state)
      0ebe4802
  6. 11 Sep, 2017 2 commits
  7. 10 Sep, 2017 1 commit
  8. 09 Sep, 2017 2 commits
  9. 03 Sep, 2017 2 commits
  10. 01 Sep, 2017 2 commits
  11. 24 Aug, 2017 1 commit
  12. 21 Aug, 2017 2 commits
  13. 18 Aug, 2017 2 commits
  14. 17 Aug, 2017 3 commits
  15. 10 Aug, 2017 2 commits
  16. 05 Aug, 2017 1 commit
  17. 03 Aug, 2017 2 commits
  18. 01 Aug, 2017 1 commit
  19. 31 Jul, 2017 1 commit
  20. 30 Jul, 2017 1 commit
    • Lorenz Huedepohl's avatar
      Loop over all possible domain decompositions · fabb1c42
      Lorenz Huedepohl authored
      We got reports from a user that there were troubles with certain domain
      decompositions. So far the tests only looked at (approximately) square
      decompositions in column-major process order.
      
      Now, a new class of tests loops over all possible decompositions
      (row * col) for a given number of total tasks.
      
      So far, we can not confirm that there are any problems, all
      possibilities work as expected.
      fabb1c42
  21. 29 Jul, 2017 1 commit
  22. 27 Jul, 2017 1 commit
  23. 25 Jul, 2017 2 commits
  24. 20 Jul, 2017 1 commit
    • Pavel Kus's avatar
      extending check_correctness · dbef90e4
      Pavel Kus authored
      Previously we checked ortonormality of eigenvectors by comparing
      the matrix S^T * S to identity matrix. The new feature is also
      checking just the diagonal of the matrix S^T * S. By that we get
      the information, how far are the eigenvectors from having length 1.
      If it turns out to be important, we could try to normalize them
      at the end of elpa, which is simple (in contrast with enforcing
      better orthogonality).
      dbef90e4
  25. 18 Jul, 2017 5 commits