Huamin Chen
Huamin Chen
@huoqifeng I don't know much about nvidia on s390, the nvml pkg can be [turned on and off](https://github.com/sustainable-computing-io/kepler/blob/main/cmd/exporter.go#L39) at runtime. You can make a conditional build for s390 to turn...
@williamcaban nice code reading! PR welcome
@jichenjc `pod_cpu_instructions` is used for VPA dashboard [here](https://github.com/sustainable-computing-io/clever/blob/main/dashboards/clever-dashboard-w-IPS.json#L329)
@sustainable-computing-io/maintainer
@marceloamaral I think current model works for time-sharing case (e.g. on GKE), in the memory-sharing (e.g. AliCloud) case, more robust model is needed.
@huoqifeng here you go 
@jichenjc would #152 help?
here is my containerd config ```console # uname -a Linux hchen-KVM 5.15.0-46-generic #49~20.04.1-Ubuntu SMP Thu Aug 4 19:15:44 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux # crictl info |jq '.config.containerd.runtimes' {...
@jichenjc it you want to give it a try? would you also like to write a unit test too? That'll make sure the regex works for all.