Slide 3 of 5

Key takeaways

What you learned

  1. Two approaches to database monitoring, each for a different need:

    ApproachBest forDepth
    Database ObservabilityQuery performance analysis (MySQL, PostgreSQL)Deepest: individual queries, explain plans, and wait events
    IntegrationsInfrastructure health monitoring (any supported database)Broad: metrics, dashboards, and alerts

  2. How to choose:

    • Database supported by DB Observability? → Use it for the deepest visibility
    • Not supported? → Use an integration for infrastructure monitoring

  3. Key details:

    • Database Observability supports managed databases (RDS, Aurora, CloudSQL, Azure) for MySQL and PostgreSQL
    • Integrations use Grafana Alloy with built in exporters. Most set up in 15–20 minutes.
    • These approaches are complementary. Use Database Observability for MySQL and PostgreSQL, and integrations for other databases in the same environment.

The one thing to remember

Start with Database Observability for MySQL and PostgreSQL. It gives you the deepest visibility. For other databases, use integrations.

Script

Here’s the decision framework in short. If your database is MySQL or PostgreSQL, start with Database Observability. It gives the deepest visibility, down to individual queries, explain plans, and wait events. It also supports managed databases like RDS, Aurora, Cloud SQL, and Azure Database.

If your database isn’t supported by Database Observability, use an integration. Integrations cover MongoDB, Redis, Memcached, Cassandra, MySQL, PostgreSQL, and Oracle Database with pre-built dashboards and alerts. Most set up in 15 to 20 minutes using Grafana Alloy.

The most important thing to remember is that these approaches are complementary. You can use Database Observability for your MySQL or PostgreSQL database and integrations for another database technology in the same environment.

Previous Next