Kubernetes: request all suitable GPUs #3259
Merged
Conversation
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
Previously, KubernetesCompute only used GPU from the first offer to set node affinity, and if that type of GPU was not available (e.g., another job or even some non-dstack pod had already taken it), the job eventually failed with FAILED_TO_START_DUE_TO_NO_CAPACITY, even if there were other GPUs matching the run spec requirements. Now, we inspect all nodes to request all suitable GPUs (any of). In addition, we now use upper bounds of Ranges (CPU, memory, disk) as limits except for GPU, which cannot have request =/= limit (as it cannot be overcommited). Part-of: #3126
r4victor
approved these changes
Nov 6, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Previously, KubernetesCompute only used GPU from the first offer to set node affinity, and if that type of GPU was not available (e.g., another job or even some non-dstack pod had already taken it), the job eventually failed with FAILED_TO_START_DUE_TO_NO_CAPACITY, even if there were other GPUs matching the run spec requirements.
Now, we inspect all nodes to request all suitable GPUs (any of).
In addition, we now use upper bounds of Ranges (CPU, memory, disk) as limits except for GPU, which cannot have request =/= limit (as it cannot be overcommited).
Part-of: #3126