Jiaxin Shan

@ying2025 sorry for late response. I am busy with internal support work recently and I will ask our team members to share some performance results they did recently on the...

> Otherwise, could you share some benchmark results and about routing strategies ? We can share some results done by us tomorrow. Need some polishment. @ying2025 I check these results,...

Seems it can not fetch `http://10.163.193.9:8080/metrics:`. Did you get a chance to check this endpoint inside the pod? BTW, are you using vLLM? by default the target port is 8000,...

this is expected scenarios, we just need to remove these pods counting in autoscaling calculation. This has been done.

Hi @ying2025 Thanks for your feedback. We actually have an issue tracking the helm development https://github.com/vllm-project/aibrix/issues/66. If people like to install using helm, we will try to promote this story

@ying2025 Could you elaborate with more details? which mirror repository are you talking about?

@ericmeadows this is on going and we will provide a chart pretty soon.

Option 1: leverage `helm-v1-alpha` https://book.kubebuilder.io/plugins/available/helm-v1-alpha. I need to bump the kubebuilder version. here's my current version for backup purpose. `kubebuilder version Version: main.version{KubeBuilderVersion:"3.15.0", KubernetesVendor:"1.27.1", GitCommit:"c01af8fb2cf7c8e11b06b6b491f7974fc1232d1a", BuildDate:"2024-05-15T10:16:58Z", GoOs:"darwin", GoArch:"arm64"}` I downloaded...

#1323 has been added to support helm deployment. Please check https://github.com/vllm-project/aibrix/tree/main/dist/chart for details. I will close this issue now. One remaining issue is to provide more documentation

@happyandslow If you’d still like to move forward with this PR, could you please address the reviewer’s comments?