[Training v3.0 consolidation] Confirm computation time step for DLRM
Open
Nobody has claimed this yet.
bug
Future
PASS Score Blocker
Training
- Dominant language
- Python
- Stars
- 203
- Forks
- 67
- Avg merge
- 20m
- Merged PRs (30d)
- 8
Description
DLRM computation time requires a throughput per accelerator of 15 GB/s+ with a compute time of 0.00038 (GB200) and 0.00056 (MI300X).
This issue captures the need to double check the computation time for the benchmark
Contributor guide
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 locating the DLRM benchmark entry point and the code or configuration that defines computation time and accelerator throughput. Verify the 0.00038 GB200 and 0.00056 MI300X values against the 15 GB/s+ requirement; done means the values are confirmed or the issue identifies the needed correction.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 48/100