Hello, I am wondering if someone can provide a bit more descriptions on these parameters to optimize performance. As far as I know, when using multiple GPUs, I had to select local-rank for device-id and cuda-aware for mpi-type. When exactly should i be using round-robin and local-rank? And when sh…

Cuda backend parameters

WillT Split this topic 23 May 2021 21:50 5

Topic		Replies	Views
GPU ID and GPU multiply tasking Cases	5	434	6 September 2022
About running PyFR on multiple GPUs General	7	213	17 April 2015
GPU parallelization error, ranks not equal to gpu number Errors cuda , mpi	23	932	20 September 2023
Cuda backend error ‘CudaOutOfMemory’ ‘CudaInvaildDevice’ General	9	474	18 July 2022
Whats the effect of CUDA-aware MPI? General	1	121	2 March 2017