Incorrect timing values with Python Code
Nobody has claimed this yet.
- Dominant language
- C
- Stars
- 112
- Forks
- 35
- PR merge metrics
- No merged PRs in 30d
Description
While using mpiP with a python based application that uses Horovod + NCCL, I am seeing wrong timings reported by mpiP. A sample report along with system details is attached for reference. As per the Linux time command (time mpiexec ....), the application finishes in 6950 seconds whereas mpiP reports show apptime of 13900 seconds for each rank. Does mpiP work correctly with Python + NCCL code or let me know if I am missing something?
[mpiP.zip](https://github.com/LLNL/mpiP/files/6176851/mpiP.zip)
Thanks,
Amit
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by examining the attached mpiP.zip report and system details, then compare the reported apptime with the Linux time command for the mpiexec run. Determine whether the discrepancy is specific to the Python, Horovod, or NCCL application and document the conditions that reproduce it. Done means explaining the timing difference or identifying the missing configuration or fix.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- linux, python
- Domain
- distributed-systems, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100