kubermatic / kubermatic/flowmatic

Allow passing arguments to NVIDIA GPU Operator

Open
#25 0 comments 0 reactions 0 assignees View on GitHub
sig/app-management
Dominant language
Shell
Stars
1
Forks
3
PR merge metrics
No merged PRs in 30d

Description

For the NVIDIA GPU Operator, Addon users should be able to specify some important arguments such as:

- container runtime (`docker` / `crio` / `containerd`)
- default GPU mode (`passthrough` / `baremetal` / `vgpu`)

For this, the best approach may be to separate the NVIDIA GPU operator / AMD GPU Device Plugin into separate addons.

Contributor guide

Open the contributing guide

Research direction

Start by locating the NVIDIA GPU Operator and AMD GPU Device Plugin addon definitions, then trace how addon users currently configure operator settings. Define how container runtime and GPU mode values should be exposed and whether the addons must be separated. Done means users can provide both requested arguments with supported values.

Written by the indexing model from the issue text.

Assessment

Tech stack
kubernetes
Domain
infrastructure
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.