Skip to content

Update hpctests to obey UCX_NET_DEVICES when RoCE devices present #421

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 7 commits into from
Aug 7, 2024
Merged

Update hpctests to obey UCX_NET_DEVICES when RoCE devices present #421

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
0951242
Turn off higher priority MPI net devices
bertiethorpe Aug 7, 2024
9bedf4c
Update pingmatrix.sh.j2
bertiethorpe Aug 7, 2024
4c7b9bd
Merge branch 'main' into fix/hpctests_ucx_net_devices
sjpb Aug 7, 2024
eba9883
Update pingmatrix.sh.j2
bertiethorpe Aug 7, 2024
edd730e
Update pingpong.sh.j2
bertiethorpe Aug 7, 2024
d346871
Replace j2 comments with bash
bertiethorpe Aug 7, 2024
055a31c
Update pingpong.sh.j2
bertiethorpe Aug 7, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion ansible/roles/hpctests/templates/pingmatrix.sh.j2
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,4 +16,8 @@ echo UCX_NET_DEVICES: $UCX_NET_DEVICES
module load {{ hpctests_pingmatrix_modules | join(' ' ) }}

mpicc -o nxnlatbw mpi_nxnlatbw.c
mpirun nxnlatbw

# mpirun flags force using UCX TCP transports, overriding higher
# priority of OpenMPI btl/openib component, which is also using RDMA
# https://wiki.stackhpc.com/s/985dae84-7bd8-4924-94b7-9629a7827100
mpirun -mca pml_ucx_tls any -mca pml_ucx_devices any nxnlatbw
6 changes: 5 additions & 1 deletion ansible/roles/hpctests/templates/pingpong.sh.j2
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,4 +16,8 @@ echo UCX_NET_DEVICES: $UCX_NET_DEVICES
module load {{ hpctests_pingpong_modules | join(' ' ) }}

#srun --mpi=pmi2 IMB-MPI1 pingpong # doesn't work in ohpc v2.1
mpirun IMB-MPI1 pingpong

# mpirun flags force using UCX TCP transports, overriding higher
# priority of OpenMPI btl/openib component, which is also using RDMA
# https://wiki.stackhpc.com/s/985dae84-7bd8-4924-94b7-9629a7827100
mpirun -mca pml_ucx_tls any -mca pml_ucx_devices any IMB-MPI1 pingpong
Loading