How Much Are Your Read Replicas Actually Costing You?

The default response to a slow database has always been to add more infrastructure. Scale up the instance. Bolt on Redis and hope the invalidation holds. Pull in a DBA to untangle queries out of context. Or paste the EXPLAIN into ChatGPT and hope it tells you something useful.

But most of the time, it’s not a database problem. It’s the queries.

In this live demo, we’ll run rdst against a real RDS fleet. You’ll see how it auto-discovers instances, breaks down replica load, and surfaces the queries responsible for the majority of cost and latency.

Then we fix it.

rdst doesn’t just surface inefficiencies. It shows you how to rewrite poorly written queries and lets you deploy caching directly from your terminal. No guesswork, no extra infrastructure.

If your database bill keeps growing and you can’t explain why, this session will give you a concrete way to diagnose, optimize, and cache your workload. All with a few simple commands. Live Q&A at the end.