Skip to content

Databases

Add a database from + Service → Databases. It lands on the canvas, joins the environment private network, and can inject connection strings through variables.

Every managed database can run as a single Standard instance or as a replicated, high-availability deployment. Pick the topology when you create it.

You choose Standard or HA when you create the instance (you can raise replica count later).

Standard Single instance Your app Database primary 1 volume (PVC) One pod, one volume. Simplest, lowest cost. High availability Primary + replicas, automatic failover Your app Service routes traffic Primary node 1 · reads + writes Replica node 2 · reads Replica node 3 · reads streaming replication Replicas across nodes, survives a node or pod failure.

Standard runs one instance and one volume. HA runs a primary plus replicas across nodes, with automatic failover.

A single instance with its own disk. Simplest and cheapest. Fine for development, internal tools, and workloads that can take a short restart. If the instance moves, the database is briefly unavailable until it comes back.

A primary plus replicas on different nodes. Writes go to the primary and stream to replicas. If the primary fails, a replica is promoted. You can also read from replicas. Use this for production.

Standard High availability
Instances Single Primary + 1 or more replicas
Node spread One node Replicas on separate nodes
Failover None (restart) Automatic primary promotion
Read scaling Primary only Reads can fan out to replicas
Best for Dev, internal, non-critical Production, uptime SLAs

Each database is a managed service on your canvas: backups, private networking, and connection strings without extra config. The table lists engines and topologies.

Database Engine Topologies
PostgreSQL CloudNativePG Standard + HA
MySQL MariaDB (MySQL-compatible) Standard + HA
CockroachDB Distributed SQL Standard + HA
Database Engine Topologies
Redis Valkey (Redis-compatible) Standard + HA
Memcached Memcached Standard only
Database Engine Topologies
MongoDB DocumentDB (MongoDB-compatible) Standard + HA
CouchDB Apache CouchDB Standard + HA
Database Engine Topologies
OpenSearch OpenSearch Standard + HA
Milvus Milvus (vector DB) Standard + HA
Database Engine Topologies
ClickHouse ClickHouse Standard + HA
Database Engine Topologies
Cassandra Apache Cassandra Standard + HA
ScyllaDB ScyllaDB (Cassandra-compatible) Standard + HA

See Backups & recovery for snapshot behavior across every database, plus point-in-time recovery for PostgreSQL.

Looking for messaging? Kafka and RabbitMQ are documented under Messaging.