MySQL Query Optimization Tool for Your Workload

Releem helps developers and DBAs improve slow and frequently executed queries using data from a connected MySQL or MariaDB database. See which queries need attention and get recommendations for query logic, indexes, and schema.

Automatic weekly analysis reviews the top 100 most frequent queries and the top 100 slowest queries. For a specific issue, select a query in Query Analytics and request recommendations on demand.

Each recommendation explains the issue and the suggested change. Your team reviews and applies changes, then uses production query metrics to check the result.

Review recommendations in the Releem web dashboard using your browser. Query data collection requires the Releem Agent with SQL Query Optimization enabled and the required database permissions. The agent runs on Linux and Windows Server. See the setup instructions for supported MySQL, MariaDB and Percona Server deployments, including AWS RDS and Google Cloud SQL.

Releem automatically analyzes production SQL queries once a week to detect performance issues without requiring manual action from engineers.

During each analysis cycle, Releem reviews:

• The top 100 most frequent queries, which often have the highest cumulative impact on database load
• The top 100 slowest queries, which directly affect application latency

Based on real execution statistics, Releem identifies queries that are inefficient for the current data volume and workload. When an issue is detected, Releem suggests recommendations and marks the query as requiring attention.

This workflow is designed to catch performance regressions early, including queries that became slow over time and new queries that were deployed without being tested under real production conditions.

Releem sent notifications via email and engineers can review recommendations in the dashboard. All changes remain fully under the engineer's control.

In addition to automatic analysis, Releem allows engineers to request optimization recommendations for any specific query.

From the Query Analytics dashboard, a developer or DBA can select a query and trigger optimization manually. Releem analyzes the selected query using real execution statistics, access patterns, and existing indexes, then generates optimization recommendations tailored to that query.

This workflow is useful when investigating a known performance issue, reviewing a newly deployed query, validating changes before further optimization, debugging slow query log entries.

It provides targeted analysis without waiting for the next automatic cycle, while using the same production data and safety model.

In the manual workflow described here, all recommendations are presented for review. Engineers decide which changes to apply and execute them manually when appropriate.

Releem analyzes SQL queries at the execution-plan level and classifies issues based on how data is accessed, filtered, joined, and processed in production.

Optimization recommendations focus on improving index usage, reducing unnecessary data processing, optimizing joins, sorting and grouping operations, and correcting schema or index inefficiencies that negatively affect execution plans. In cases where query performance is influenced by server settings, Releem also accounts for configuration-related constraints when suggesting improvements.

Each recommendation is directly tied to observed execution behavior and clearly indicates what can be changed and at which level, whether in query logic, schema, or server configuration.

See the query, its execution metrics, and an index recommendation together in Query Inspection. Recommendations include an explanation of the issue and concrete SQL statements or guidance to review against your application and schema.

Your team decides which changes to apply and when. In the manual workflow described here, you review and apply the recommended changes using your existing deployment process.

If a recommendation is not relevant, you can delete it. Deleted recommendations will not be suggested again unless the query or workload changes and a new recommendation is generated.

After an optimization recommendation is applied, Releem continues to observe the query using production execution data. This makes it possible to verify whether the change had the intended effect under real workload conditions rather than synthetic benchmarks.

Releem compares query behavior before and after the change, focusing on execution time, frequency, and overall performance impact. This allows engineers to confirm improvements, detect regressions early, and understand the actual effect of each optimization.

By tying recommendations to measurable results, Releem helps teams make incremental, validated improvements to query performance while maintaining confidence in production stability.

Releem is built around the idea that managing MySQL databases should be straightforward. SQL query optimization is one part of a broader advisory system:

1. Configuration tuning - Releem analyzes database configuration, workload and system resources, detects inefficient or risky settings, and suggests safe configuration changes to improve performance and stability without manual trial and error.
2. Health checks - Releem continuously evaluates key database and system metrics to highlight issues that need attention.
3. Schema optimization - Releem checks schema and index structures to detect issues that slow down queries, such as missing or inefficient indexes, and suggests safe improvements without requiring application code changes.
4. MySQL monitoring - By tracking workload metrics, Releem identifies performance bottlenecks early, helping you stay ahead of problems without manual investigation.
5. Broad compatibility - Releem works with MySQL, MariaDB, Percona Server, and cluster configurations including Percona XtraDB Cluster, Galera Cluster, and InnoDB Cluster. It supports both self-hosted and cloud-managed databases on AWS RDS, Google Cloud SQL, and Azure Database.