Skip to content

Tutorial: Add a end to end for using IB on AKS #52

@surajssd

Description

@surajssd
  • 1: I want to use AKS to deploy a distributed inference workload using vLLM backed by IB using the sriov plugin. I can use the gpu operator + network operator combination.
  • 2: I want to use AKS to deploy a distributed inference workload using vLLM backed by IB using the rdma-shared-plugin. I don't wanna deal with the GPU operator so I will use the AKS managed GPU drivers.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions