5enxia/parallel-krylov

Issues

• Calculation result becomes strange at the time of asynchronous execution with memcpyPeerAsync()

  #14 opened 3 years ago by 5enxia
  0

• Reuse calculated vector x

  #13 opened 3 years ago by 5enxia
  0

• Make interface scipy-like

  #11 opened 3 years ago by 5enxia
  0

• Implement with NCCL

  #12 opened 3 years ago by 5enxia
  0

• CUDA_VISIBLE_DEVICE Operation

  #10 opened 3 years ago by 5enxia
  0

• InfiniBand cache error

  #9 opened 3 years ago by 5enxia
  3

• Doesn't converged, k-skip MrR with GPU + MPI

  #8 opened 3 years ago by 5enxia
  1

• IndexError: Index 204 is out of bounds for axis 0 with size 136

  #7 opened 3 years ago by 5enxia
  1

• The output of the MPI adaptive k skip MrR method does not match a single version.

  #1 opened 5 years ago by 5enxia
  0