-
# Problem
The usage differences between MPI_COMM_FREE and MPI_COMM_DISCONNECT are implied by the current normative text but an example shows more clearly the range of freedom permitted.
# Propos…
-
Could you show the code about generating mpi_annotation.json? @wangziren1
-
Following on from #309, once we have an MPI set up we should think about writing data directly from each node, so that the full dataset doesn't need to be gathered. Blocked until #4 goes through.
-
Is this supported? We have the ability (through Slurm) to `srun ... dask-mpi...` which will launch the cluster on a queue set aside for interactive work. From the command line it's a single command.…
-
-
## Background information
### What version of Open MPI are you using? (e.g., v1.10.3, v2.1.0, git branch name and hash, etc.)
v3.0.1
### Describe how Open MPI was installed (e.g., from a sour…
-
We need to configure cpus-per-task and OMP_NUM_THREADS when running openmc. We also should not run more than 8 tasks-per-node. see https://github.com/radiasoft/sirepo/issues/6220#issuecomment-16996097…
-
Running in Ubuntu 22.04.1 with OpenMPI/4.1.6-GCC-13.2.0, the `make check` gives the following error for the MPI tests:
```
{EESSI 20240402} jmorillo@arriesgado-2 /tmp/eb/easybuild/build/Extrae/4.1.6…
-
### System Info
AWS g5.12xlarge (GPU 4 x NVIDIA A10G)
CPU x86_64
TensorRT-LLM v0.11.0
### Who can help?
@byshiue
### Information
- [X] The official example scripts
- [ ] My own modified scripts…
-
This looks like a new test -- it is failing on pm-cpu with nvidia compiler. Based on the dates of log files, it looks like
the test is hanging.
Note the current MPI count used by default for this …