Feature Overview (mandatory - Complete while in New status)

NVIDIA Inference Xfer Library (NIXL) is a tool for point-to-point communication in AI inference frameworks. We expect to see it added to vllm / vllm-d for sharing workloads in multi-node scenarios.

Goals (mandatory - Complete while in New status)

• Build nixl as a wheel

Requirements (mandatory -_ Complete while in Refinement status):
A list of specific needs, capabilities, or objectives that a Feature must deliver to satisfy the Feature. Some requirements will be flagged as MVP. If an MVP gets shifted, the Feature shifts. If a non MVP requirement slips, it does not shift the feature.

Done - Acceptance Criteria (mandatory - Complete while in Refinement status):

A wheel collection owner can add a supported version of nixl to their collection.

Use Cases - i.e. User Experience & Workflow: (Initial completion while in Refinement status):

vllm-d builds will include this package

Out of Scope _{}(Initial completion while in Refinement status):{_}

• We do not need to add the packages to any collections.

Documentation Considerations _{}(Initial completion while in Refinement status):{_}

N/A

Questions to Answer _{}(Initial completion while in Refinement status):{_}

Which version? Always the latest.

Background and Strategic Fit (Initial completion while in Refinement status):
https://github.com/ai-dynamo/nixl

Customer Considerations _{}(Initial completion while in Refinement status):{_}

N/A

Team Sign Off (Completion while in Refinement status)

• All required Epics (known at the time) are linked to the this Feature
• All required Stories, Tasks (known at the time) for the most immediate Epics have been created and estimated
• Add - Reviewers name, Team Name
• Acceptance == Feature as “Ready” - well understood and scope is clear - Acceptance Criteria (scope) is elaborated, well defined, and understood
• Note: Only set FixVersion/s: on a Feature if the delivery team agrees they have the capacity and have committed that capability for that milestone

*An engineer or tech lead from the product requesting this feature is required for the signoff below.

• …