Hello team, Are there any plans to support multi clickhouse cluster from the query-service? A separate clickhouse cluster for logs, metrics and traces?

What's the use case?

For our use case we have TBs of metrics and logs flowing in, we want to store logs and metrics in different Clickhouse clusters and then query it using query-service

Why do you want to store them separately?

Won't a single Clickhouse cluster be a single point of failure so much of logs and metrics data?

I am not sure how the per-signal cluster solves that. You will need to use replication if the concern is the single point of failure.

Let's say if there are a lot metrics being written to the Clickhouse cluster and the cluster gets overloaded. This can bottleneck the logs pipeline from sending logs to the clickhouse cluster

You would solve that by capacity planning, and putting query restrictions so that some rouge querying/ingestion won't cause problems to the entire cluster etc..

Understood, thank you! yes the log and metrics data is in TBs. we keep it for close to 2 months