Question: Does HAMi support in-place GPU memory adjustment for running Pods (without restart)?

[maeve675]

Hi HAMi maintainers, thanks for the great project.

I’m currently evaluating HAMi as an alternative to Tencent GPU Manager in a Kubernetes GPU sharing environment, and I’d like to clarify whether HAMi supports runtime GPU resource adjustment for already running Pods.

[archlitchi]

Currently, we haven't implemented that feature yet