fix(slurm): request zero GPUs when export falls back to GPU partition#915
Open
AdamRajfer wants to merge 1 commit into
Open
fix(slurm): request zero GPUs when export falls back to GPU partition#915AdamRajfer wants to merge 1 commit into
AdamRajfer wants to merge 1 commit into
Conversation
PR #901 introduced `export_partition = cpu_partition or cfg.execution.partition`. When cpu_partition is null (existing configs), export falls back to execution.partition — typically a GPU partition like 'batch' — with no --gpus-per-node in the sbatch header, and schedulers that enforce GPU specification on non-CPU partitions reject the submission: sbatch: error: Cannot find GPU specification, you may not submit a job not requesting GPUs in a non-CPU partition, partition: batch The existing `--gpus 0` on the inner srun only applies to the step, not to the allocation. Declare `--gpus-per-node=0` at the sbatch level when falling back so the scheduler accepts an allocation with no GPUs. Signed-off-by: Adam Rajfer <arajfer@nvidia.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR #901 introduced
export_partition = cpu_partition or cfg.execution.partition. When cpu_partition is null (existing configs), export falls back to execution.partition — typically a GPU partition like 'batch' — with no --gpus-per-node in the sbatch header, and schedulers that enforce GPU specification on non-CPU partitions reject the submission:sbatch: error: Cannot find GPU specification, you may not submit a job
not requesting GPUs in a non-CPU partition, partition: batch
The existing
--gpus 0on the inner srun only applies to the step, not to the allocation. Declare--gpus-per-node=0at the sbatch level when falling back so the scheduler accepts an allocation with no GPUs.