... | @@ -2,16 +2,16 @@ |
... | @@ -2,16 +2,16 @@ |
|
|
|
|
|
- not yet decided
|
|
- not yet decided
|
|
|
|
|
|
Changelog for ELPA 2023.05.001
|
|
Changelog for ELPA 2023.11.001
|
|
- enable gpu-streams per default for NVIDIA and AMD GPUs
|
|
- enable gpu-streams per default for NVIDIA and AMD GPUs
|
|
- Updated / improved documentation and man pages
|
|
- Updated / improved documentation and man pages
|
|
- Fixed compilation error on AMD GPUs
|
|
- Fixed compilation error on AMD GPUs
|
|
- Fixed SVE 256 compute kernels
|
|
- Fixed SVE 256 compute kernels
|
|
- Allow (currently in parts of ELPA) to use NVIDIA NCCL for device to device
|
|
- Allow (currently in parts of ELPA) to use NVIDIA NCCL for device to device
|
|
commpunication
|
|
communication
|
|
- Speed up of GPU version of hermitian_multiply by up to an factor of 4
|
|
- Speed up of GPU version of hermitian_multiply by up to an factor of 4
|
|
- significantly faster full-to-tridiagonal step in ELPA 1stage GPU
|
|
- significantly faster full-to-tridiagonal step in ELPA 1stage GPU
|
|
- significatnly faster ELPA 2stage solver on Intel GPUs
|
|
- significantly faster ELPA 2stage solver on Intel GPUs
|
|
- Consistent enabling/disabling of SKEW_SYMMETRIC in header files
|
|
- Consistent enabling/disabling of SKEW_SYMMETRIC in header files
|
|
- new setup_gpu API function
|
|
- new setup_gpu API function
|
|
|
|
|
... | | ... | |