Performance question for bigger clusters #140
Description
Our cluster has ~160 nodes.
We are scraping summary exporter with prometheus and started getting timeouts.
When testing /nodes endpoint from within container, it takes 11s to get back node metrics, so we will increase scrape timeout. But I am wondering as cluster growth, is there some possibility to improve this performance?
I have no idea, but I am guessing not everyone has somewhat big clusters, so wanted to bring this up and see if you have any thoughts on this?
Runnings version v0.4.6 on v1.32.9-eks-113cf36 cluster on eks with bottlerocket nodes.