Query-level database observability for MySQL and PostgreSQL

Stop guessing why the database is slow. Grafana Cloud's Database Observability helps teams trace a slow request to the exact query, execution plan, and wait events behind it, and see which service is impacted. Test and ship precise fixes with built-in AI-powered recommendations.

No in-database agent | SQL redaction by default | MySQL 8+ and PostgreSQL 14+

Debug query latency with precision

Go from aggregate trends to the specific execution that caused the slowdown. See what it waited on, how it ran in a visual explain plan, and which tables it touched, with no production database access needed.

See the database in context

Connect slow requests to the queries behind them. Query samples link to their traces and back, and the Knowledge Graph shows which services depend on each database and where problems start.

Fix slow queries with AI-suggested DDL

Get CREATE INDEX statements and query rewrites based on your actual table schemas and explain plans. Review the evidence, test the fix, and ship it.

Trusted by everyone from startups to the Fortune 500

Observability solutions at every layer

Use Grafana Cloud for full-stack observability across applications, infrastructure, Kubernetes, and users:

Why use Grafana Cloud for database observability?

Debug query latency with precision

  • Isolate the queries that matter with sorting by wait time, errors, latency, or call rate;filter by your own labels, instance, or schema.

  • See what each execution waited on with breakdowns by time by wait-event type (lock, I/O, network, and more), and inspect individual query samples instead of averages.

  • Understand how it ran with visual explain plans that highlight costly operations such as full table scans, alongside table schema and index details.

  • Catch regressions by tracking each query's performance over time to see when a deploy or data growth slowed it down.

See the database in context

  • Follow traces to queries and back by opening the trace behind a query sample or jumping from a slow database span to its matching query.

  • See which services depend on each database with the Knowledge Graph, which shows the services connected to every instance and the health issues detected across your stack, so you can tell whether a problem started in the database or upstream.

  • Spot resource contention and overlay query latency with CPU, memory, disk I/O, and cloud provider metrics.

  • No separate tool needed because the database data sits next to your existing dashboards, alerts, and telemetry, with no proprietary data silos.

Fix slow queries with AI-suggested indexes and rewrites

  • Get specific fixes as Grafana Assistant suggests CREATE INDEX statements and query rewrites from every query view.

  • See the evidence behind each suggestion, which is based on your actual table schemas and explain plans; get explanations on what it found, such as a full table scan, and the expected improvement

  • Stay in control: copy the change, test it, and ship it through your normal workflow. Nothing runs against your database automatically

  • Fewer DBA tickets as developers resolve routine query problems themselves andDBAs can focus on the complex ones

Set up with open standards

  • Get open-source collection by deploying Grafana Alloy with its built-in MySQL and PostgreSQL components. Nothing is installed inside your database engine or on the database server

  • Support any environment, whether it's self-managed, Amazon RDS and Aurora, Google Cloud SQL, and Azure Database for MySQL and PostgreSQL, MySQL 8+,   PostgreSQL 14+

  • Redact SQL text by default by using Alloy inside your network, before it reaches Grafana Cloud

Real stories from real customers

“One of [Grafana Cloud’s] main differentiators is the use of AI to generate valuable insights, enabling engineering teams to quickly identify root causes... without relying directly on a DBA. This provides greater autonomy for engineering [and] speeds up the investigation process.”

Alexandre MagnoSenior Manager, Pismo

Get Grafana Cloud for database observability

free
Always free
Always
$0

No credit card required

Actually useful free tier with access to everything that Grafana Cloud has to offer.
  • Limited to 2,232 host hours (3 database hosts)

  • Telemetry usage counts toward Grafana Cloud free tier allowances

  • Community support

pro
Self-serve
Starts at
$0.07/ host hour (~$51 / host)

Automatic volume discounts. Pricing at scale. View calculator

Platform fee of $19 per month includes:
  • 2,232 database host hours, then pay as you go

  • Telemetry billed separately

  • 8×5 email support

enterprise
Full service
Prices are
Custom

Scalable unit price based on annual commit

Minimum commit of $25,000 per year.
  • Premium support

  • Deployment flexibility (Public Cloud, Federal Cloud, or Bring Your Own Cloud)

Part of your Grafana Cloud observability platform

Works seamlessly with other Grafana Cloud offerings including:

Frequently asked questions

Why choose Grafana Cloud

Open & unified platform

OpenTelemetry-native observability and no lock-in, with out-of-the-box solutions like Kubernetes Monitoring, Application Observability, Grafana SLO, and RUM delivered in one unified experience. 

Cost efficiency

Optimize costs without sacrificing insight with Adaptive Telemetry, which filters out unused data so your budget goes toward what actually drives value. Pair with cost management tools that help you monitor, control, and tune spend. 

AI & automation

Grafana Assistant powers agentic workflows, prebuilt dashboards, intelligent filters, and customized alerts—surfacing the data you need for faster, more efficient incident response. 

Secure. Compliant. Proven.
View certifications
FEDRAMP Compliant
NATSEC100 Certified
PCI DSS Compliant
A-LIGN Certified
AICPA SOC Type II Verified
GDPR Compliant

Welcome to

The open observability cloud

Built on open source, open standards, and open ecosystems

The Free Forever plan
Create free account
Plans to suit every scale
Explore Cloud plans