PlanetScale Postgres adds independently configurable dedicated read replicas

PlanetScale has announced that its managed Postgres product now supports dedicated read replicas that can be configured independently of the primary cluster. The company listed the change in its product changelog on 6 October 2026.

According to PlanetScale, customers can create one or more dedicated read replicas, either in the primary region or in another region from the same cloud provider. That second option means a replica can sit away from the primary database, though the changelog does not say which regions or providers are available.

The main change is the degree of control. PlanetScale says cluster size, replica count, storage and supported Postgres parameters can each be set independently from the primary cluster. In practical terms, a replica no longer has to mirror the primary's configuration, so a team could, for example, run smaller or differently tuned replicas for read-heavy work. That example is our illustration of what the setting allows, not a use case PlanetScale describes. The entry also does not say which Postgres parameters count as supported.

PlanetScale adds that dedicated read replicas can be managed from its dashboard, API, CLI or Terraform. The company points readers to separate documentation for details, which this article did not review.

What is confirmed: only that PlanetScale itself states the capability exists and describes the options above. What is not known from the available source: pricing, whether the feature is generally available or limited to certain plans, how replication lag is handled, how applications route read traffic to the new replicas, and any limits on replica counts. We searched for independent coverage and found none; this report rests on a single company announcement and has not been verified by a second source or by hands-on testing.

Why it matters: read replicas are a common way to spread query load and keep reporting or analytics queries away from a production primary. Letting replicas differ in size and settings, and sit in another region, gives teams more room to match capacity to workload and to place data nearer to readers. Whether the feature performs as described will depend on details PlanetScale has not yet stated in this changelog entry.

Revised 16 October 2026: Guth re-read the original sources and rewrote this article. The earlier version remains available as revision 1.