Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
hpc-compendium
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Deploy
Releases
Package Registry
Container Registry
Model registry
Operate
Terraform modules
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Terms and privacy
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
ZIH
hpcsupport
hpc-compendium
Commits
fde3928b
Commit
fde3928b
authored
1 year ago
by
Martin Schroschk
Browse files
Options
Downloads
Patches
Plain Diff
Document performance loss and workaround for OpenMPI 4.1.x and OMPIO
parent
ebc298e5
No related branches found
No related tags found
2 merge requests
!887
Automated merge from preview to main
,
!885
Issue 497 - OpenMPI 4.1x. Performance Loss w.r.t. MPI-IO
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
doc.zih.tu-dresden.de/docs/index.md
+1
-0
1 addition, 0 deletions
doc.zih.tu-dresden.de/docs/index.md
doc.zih.tu-dresden.de/docs/jobs_and_resources/mpi_issues.md
+21
-2
21 additions, 2 deletions
doc.zih.tu-dresden.de/docs/jobs_and_resources/mpi_issues.md
with
22 additions
and
2 deletions
doc.zih.tu-dresden.de/docs/index.md
+
1
−
0
View file @
fde3928b
...
@@ -31,6 +31,7 @@ Please also find out the other ways you could contribute in our
...
@@ -31,6 +31,7 @@ Please also find out the other ways you could contribute in our
## News
## News
*
**2023-11-16**
[
OpenMPI 4.1.x - Workaround for MPI-IO Performance Loss
](
jobs_and_resources/mpi_issues/#openmpi-v41x-performance-loss-with-mpi-io-module-ompio
)
*
**2023-10-04**
[
User tests on Barnard
](
jobs_and_resources/barnard_test.md
)
*
**2023-10-04**
[
User tests on Barnard
](
jobs_and_resources/barnard_test.md
)
*
**2023-06-01**
[
New hardware and complete re-design
](
jobs_and_resources/architecture_2023.md
)
*
**2023-06-01**
[
New hardware and complete re-design
](
jobs_and_resources/architecture_2023.md
)
*
**2023-01-04**
[
New hardware: NVIDIA Arm HPC Developer Kit
](
jobs_and_resources/arm_hpc_devkit.md
)
*
**2023-01-04**
[
New hardware: NVIDIA Arm HPC Developer Kit
](
jobs_and_resources/arm_hpc_devkit.md
)
...
...
This diff is collapsed.
Click to expand it.
doc.zih.tu-dresden.de/docs/jobs_and_resources/mpi_issues.md
+
21
−
2
View file @
fde3928b
...
@@ -2,11 +2,30 @@
...
@@ -2,11 +2,30 @@
This pages holds known issues observed with MPI and concrete MPI implementations.
This pages holds known issues observed with MPI and concrete MPI implementations.
## OpenMPI v4.x.x - Performance Loss with MPI-IO-Backend OMPIO
## OpenMPI v4.1.x - Performance Loss with MPI-IO-Module OMPIO
OpenMPI v4.1.x introduced a couple of major enhancements, e.g., the
`OMPIO`
module is now the
default module for MPI-IO on
**all**
filesystems incl. Lustre (cf.
[
NEWS file in OpenMPI source code
](
https://raw.githubusercontent.com/open-mpi/ompi/v4.1.x/NEWS
)
).
Prior to this,
`ROMIO`
was the default MPI-IO module for Lustre.
Colleagues of ZIH have found, that some application suffer a significant performance loss using
`OMPIO`
as MPI-IO module with OpenMPI/4.1.x modules on ZIH systems. At the moment, the root cause is
unclear and needs further investigation.
**A workaround**
for this performance loss is to use "old", i.e.,
`ROMIO`
MPI-IO-module. This
is achieved by setting the environment variable
`OMPI_MCA_IO`
before executing the application as
follows
Workaround is to make use of the "old" IO-backend:
```
console
```
console
export OMPI_MCA_io=^ompio
export OMPI_MCA_io=^ompio
srun ...
```
or setting the option as argument, in case you invoke
`mpirun`
directly
```
console
mpirun --mca io ^ompio ...
```
```
## Mpirun on partition `alpha`and `ml`
## Mpirun on partition `alpha`and `ml`
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment