faq:cuda_support
This is an old revision of the document!
Which parts of CP2K are CUDA-accelerated?
- FFTs, when compiled with
-D__PW_CUDA
- Anything that uses
dbcsr_multiply
, when compiled with-D__DBCSR_CUDA
. This includes in particular the linear scaling DFT code. - Anything that uses
dgemm
(notpdgem
), when linked against Cray's accelerated LibSci. Currently, this is most useful for theMP2
method.
faq/cuda_support.1393322577.txt.gz · Last modified: 2020/08/21 10:15 (external edit)