fighterhit
fighterhit
details里面还有好多没展开啊，能否全部展开呢想打印一下，语雀还没更新。
> Thank you @sozgat , I will try it later.
> @jodh-intel Really appreciate for your kindly help, we are going to try this way to use GPU pass-through mode with Kata2.0. What am I now concerned is that how...
> @fighterhit I think we can use Kubevirt device plugin with kata2.0, but I still need to test it. Thank you for your advice @han2ni3bal , if you have any...
> @fighterhit I think we can use Kubevirt device plugin with kata2.0, but I still need to test it. Hi @han2ni3bal , have you tested the kubevirt gpu device plugin?...
> > The official NVIDIA k8s-device-plugin supports GPU health monitoring, so that GPU having xid error will become usable and won't get assigned to pod. > > So is gpu-manager...
Today I try to reproduce the problem. First I create 7 NVIDIA GPU Pods, each occupying 1 GPU. - The NVIDIA GPU Pod description ```yaml apiVersion: apps/v1 kind: Deployment metadata:...
Hi @mYmNeo , thanks for your quick answer. Actually, in order not to affect the current k8s environment, we created a new scheduler and started admission as its extended scheduler....
Hi @qifengz, @zwpaper has pointed out the reason. If you read the code, you may find that `gpu-manager` actually sorts GPUs according to **topology** and **the number of processes** running...
> > Hi @qifengz, @zwpaper has pointed out the reason. If you read the code, you may find that `gpu-manager` actually sorts GPUs according to **topology** and **the number of...