llnl / llnl/mpi-tools

Gyan output is empty apart from the headers

Open
#4 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C
Stars
14
Forks
7
PR merge metrics
No merged PRs in 30d

Description

I'm starting my job as follows
```
LD_PRELOAD=~/mpi-tools/mpi_t/gyan/libgyan.so mpirun -np 8 -hostfile hosts -mca plm_rsh_no_tree_spawn 1 -bind-to socket -map-by slot -x HOROVOD_HIERARCHICAL_ALLREDUCE=1 -x HOROVOD_FUSION_THRESHOLD=16777216 -x NCCL_MIN_NRINGS=4 -x LD_LIBRARY_PATH -x PATH -mca pml ob1 -mca btl ^openib -x TF_CPP_MIN_LOG_LEVEL=0 python -W ignore train_imagenet_resnet_hvd.py --synthetic -b 256 --num_batches 200 --clear_log
```

Output is
```
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
Performance profiling for the complete MPI job:
----------------------------------------------------------------------------------------
Variable Name Type Minimum(Rank) Maximum(Rank) Average
----------------------------------------------------------------------------------------
----------------------------------------------------------------------------------------
```

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the documented mpirun command with Gyan and the provided training workload, then trace how Gyan collects and prints MPI performance variables. Check why each rank emits only the headers and no measurements; done means the complete MPI job output contains the expected profiling values rather than empty tables.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
distributed-systems, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.