F#325 Add NVIDIA NIM service appliance for OpenNebula GPU deployments#335
Open
JavierPalomaresTorrecilla wants to merge 6 commits intomasterfrom
Open
F#325 Add NVIDIA NIM service appliance for OpenNebula GPU deployments#335JavierPalomaresTorrecilla wants to merge 6 commits intomasterfrom
JavierPalomaresTorrecilla wants to merge 6 commits intomasterfrom
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR adds the
service_Nimappliance to enable OpenNebula to provision virtual machines that run the NVIDIA NIM container on GPU-enabled infrastructure.The appliance expects the following input parameters:
ONEAPP_NIM_NVIDIA_REGISTRY(mandatory).ONEAPP_NIM_NVIDIA_REGISTRY_KEY(mandatory).ONEAPP_NIM_NVIDIA_IMAGE_REF(mandatory).ONEAPP_NIM_NVIDIA_REGISTRY_USER(optional fornvcr.io, otherwise used for registry authentication).When
ONEAPP_NIM_NVIDIA_REGISTRYis set tonvcr.io, the appliance follows the documented NVIDIA Container Registry authentication flow and uses the special username$oauthtokentogether with the provided API key, as described in the NVIDIA NGC User Guide. In that case,ONEAPP_NIM_NVIDIA_REGISTRY_USERdoes not need to be provided through OpenNebula context.The appliance configures the runtime needed to launch the NVIDIA NIM container, including propagation of
NGC_API_KEYinto the container, cache mounting, shared memory allocation, port exposure, NVIDIA runtime integration, and readiness-gated bootstrap through/v1/health/ready. As a result, the service is only markedREADYafter container startup completes successfully and the readiness check passes.The original target image used during validation, nvcr.io/nim/openai/gpt-oss-120b:latest, exceeded the available single-GPU memory during model initialization, and the appliance correctly remained in bootstrap without reporting
READY. Validation was then completed successfully with nvcr.io/nim/openai/gpt-oss-20b:latest, which reached readiness and was verified through/v1/health/live,/v1/health/ready,/v1/models,/v1/version,/v1/metadata, and streaming inference via/v1/chat/completions.