Paul, Thank you for your suggestion. I will try different spack install specifications.
Cho ________________________________ From: Grosse-Bley, Paul Leonard <paul.grosse-b...@stud.uni-heidelberg.de> Sent: Friday, June 30, 2023 4:07 AM To: Ng, Cho-Kuen <c...@slac.stanford.edu> Cc: petsc-users@mcs.anl.gov <petsc-users@mcs.anl.gov> Subject: Re: [petsc-users] Using PETSc GPU backend Hi Cho, you might want to specify the GPU architecture to make sure that everything is compiled optimally. I.e. "spack install petsc +cuda cuda_arch=80 +zoltan" Best, Paul On Friday, June 30, 2023 01:50 CEST, petsc-users-requ...@mcs.anl.gov wrote: Date: Thu, 29 Jun 2023 23:50:10 +0000 From: "Ng, Cho-Kuen" <c...@slac.stanford.edu> To: "petsc-users@mcs.anl.gov" <petsc-users@mcs.anl.gov> Subject: [petsc-users] Using PETSc GPU backend Message-ID: <byapr07mb5431a3dacc91a963572d66e4e0...@byapr07mb5431.namprd07.prod.outlook.com> Content-Type: text/plain; charset="iso-8859-1" I installed PETSc on Perlmutter using "spack install petsc+cuda+zoltan" and used it by "spack load petsc/fwge6pf". Then I compiled the application code (purely CPU code) linking to the petsc package, hoping that I can get performance improvement using the petsc GPU backend. However, the timing was the same using the same number of MPI tasks with and without GPU accelerators. Have I missed something in the process, for example, setting up PETSc options at runtime to use the GPU backend? Thanks, Cho