feat(processor): enable EFA for whole-host reservations (single-node, not just multinode) - #221
Merged
Merged
Conversation
… not just multinode) EFA was only attached to multinode full-node pods. Enable it for any whole-host reservation (gpu_count == max_gpus) on an EFA-capable SKU, including single-node ones — a whole-host pod owns the node, so the existing hostNetwork + sshd:2222 model stays conflict-free. - Centralize the gate in _pod_uses_efa (efa_count > 0 and gpu_count == max_gpus); get_pod_resource_limits/requests now call it instead of duplicating the condition. Drops the is_multinode requirement. - Subhost (partial-node) pods still get NO EFA: they share a node and would collide on hostNetwork port 2222 — deferred. - Fix b200 efa_count 32 -> 8 to match the node's advertised efa_network_cards (p6-b200.48xlarge caps at 8), else whole-host b200 pods can't schedule. Needs a reservation_processor lambda deploy to take effect.
Collaborator
|
cc @wdvr I just realize that I don't have owner permission to the repo T_T |
wdvr
approved these changes
Jul 16, 2026
Collaborator
|
@wdvr Ty for your help! Are u planning to do a release for this change? I don't think I have the permission to run https://github.com/wdvr/osdc/actions/workflows/publish.yml |
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
EFA was only attached to multinode full-node pods. Enable it for any whole-host reservation (gpu_count == max_gpus) on an EFA-capable SKU, including single-node ones — a whole-host pod owns the node, so the existing hostNetwork + sshd:2222 model stays conflict-free.
Needs a reservation_processor lambda deploy to take effect.