Add S3 config parameters to flux/dlrm/retinanet yaml files
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 203
- Forks
- 67
- Avg merge
- 20m
- Merged PRs (30d)
- 8
Description
The v3.0 workloads {flux|dlrm|retinanet}_{b200|mi355} are currently missing S3 storage parameters. We would need to create new {flux|dlrm|retinanet}_{b200|mi355} workload yaml files for the currently support S3 libraries {s3dlio|minio|s3torch} and add the following fields (similar to unet3d).
storage:
storage_type: s3
storage_root: <s3-bucket-name>
storage_library: <{s3dlio|minio|s3torch}>
storage_options:
endpoint_url: <S3_URL>
region: us-east-1
s3_force_path_style: true
checkpoint:
checkpoint_folder: s3://<s3-bucket-name>/s3dlio/llama3-8b
FYI: @russfellows
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 comparing the existing unet3d workload YAML with the v3.0 flux, dlrm, and retinanet workload files. Create the requested B200 and MI355 variants for s3dlio, minio, and s3torch, adding the specified storage and checkpoint fields, then verify each YAML file has the expected S3 configuration.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- yaml
- Domain
- cloud, machine-learning
- Issue type
- Feature
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 68/100