Issue Summary

We would like to achieve higher performance during read operations to Redis. however, currently, our redash setup can only connect to the primary node. How does redash work with replicas? Note we are using the AWS managed Redis aka Elasticache. There is already a reader endpoint that can be utilized for scaling Redis.

Technical details:

How do we connect?

REDASH_REDIS_URL= redis://host url:6379/0

currently we are running version 7 without docker, we are planning to upgrade it to version 8 on docker.

I don’t have an answer for you. But I’m really curious: how is Redis a bottleneck for your Redash instance? Ours runs thousands of queries every minute and Redis doesn’t blink.

It is always good to have read replication. Aslo for availability reasons. I use Elasticache nodes and I deploy different nodes on different AZs. So it make sense to utilize them as well.