Loadbalancing requests to Stores #6372
Unanswered
sidineyc
asked this question in
Questions & Answers
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi there,
I have a Thanos setup with Time partitioning shard for Stores, I was looking into optimize the performance and it has helped a little. However, it seems requests are not distributed evenly between the available replicas, mainly looking into CPU/Memory utilization.
I have the headless service enabled, and query pod has the following arguments
I found this implementation, which is still experimental and has some drawbacks as per discussion #5548
Am I right that this is a current limitation, so to say? Any plans for further improving this scalability factor?
Thanks in advance.
Beta Was this translation helpful? Give feedback.
All reactions