Support for Configuring GPU Access for Both Containers and VMs on the Same Node
Open
Nobody has claimed this yet.
feature
lifecycle/frozen
- Dominant language
- Go
- Stars
- 2.9k
- Forks
- 552
- Avg merge
- 2d 4h
- Merged PRs (30d)
- 90
Description
Issue or feature description
We have a node in our Kubernetes cluster with an 8-way A100 GPU. Our goal is to configure GPU access for both containers and VMs on this node. Specifically, we would like to mix KubeVirt VM passthrough with normal GPU pods.
Questions:
- Is it currently possible to configure GPU access for both containers and VMs on the node?
- If this configuration is not supported, are there any workarounds or plans on the roadmap to support such a setup in the future?
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 reviewing the GPU Operator's Kubernetes configuration guidance and the issue's KubeVirt VM passthrough and GPU pod requirements. Determine whether mixed access is currently supported for the same node, and document any confirmed workaround or roadmap status as the completed outcome.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- kubernetes
- Domain
- devops, infrastructure
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100