Yu Cao
Yu Cao
Hi @cheyang, I am currently studing the code and have a question about pods that have multiple containers. I trace the code in kubelet and find that it called the...
using allocatable field for gpu healthy status tracking
Hi @cheyang, I copy the issue from device pluign repo and have a question about the case that one pod have multiple containers. I trace the code in kubelet and...
pleat take a look
Do we reduce the GPU memory from the allocatable memory pool when unhealthy card is detected? it seems that extender schedule do not have related information.