Skip to content

Comments

Add background cache updates for PodMapper and MIG cache in NVML prov…#600

Open
jaeeyoungkim wants to merge 1 commit intoNVIDIA:mainfrom
whatap:feature/4.4.2-4.7.1
Open

Add background cache updates for PodMapper and MIG cache in NVML prov…#600
jaeeyoungkim wants to merge 1 commit intoNVIDIA:mainfrom
whatap:feature/4.4.2-4.7.1

Conversation

@jaeeyoungkim
Copy link

Add PodMapper functionality with caching and lifecycle management
Enhanced PodMapper to include asynchronous cache updates, improved device-to-pod mapping, support for DRA (Dynamic Resource Allocation), and proper lifecycle handling with Run and Stop methods. Updated MIG device caching to handle failures gracefully.

Fixes #599

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

High CPU usage and Hang on H100, H200 MIG nodes with DCGM_EXPORTER_KUBERNETES enabled

1 participant