Database Monitoring (DBM) Recommendations draw attention to potential optimizations and problematic areas across your database fleet.
How it works
Datadog analyzes metrics and sample data from DBM to identify your systems’ highest-priority issues. A severity indicator is calculated for each recommendation, highlighting the most impactful areas to focus on. High-severity recommendations may indicate immediate or impending problems, while lower-severity recommendations can be addressed asynchronously to proactively maintain database health.
Supported recommendation types
| Recommendation Type | Description | MongoDB | MySQL | Oracle | PostgreSQL | SQL Server |
|---|
| Function in Filter | The query calls a function on columns being filtered, leading to expensive sequential scans that can’t take advantage of typical column-based indexes. | | | | | |
| High Impact Blocker | The query is causing a significant amount of waiting time for blocked queries. | | | | | |
| High Row Count | The query returns a large number of rows in its result set. | | | | | |
| Long Running Query | The query has durations that have exceeded a threshold of 30 seconds. | | | | | |
| Low Disk Space | The database instance is running low on disk space.
Note: Only available on Amazon RDS. | | | | | |
| Missing Index | The query’s execution plan performs expensive sequential scans. When detected, Datadog recommends using an index to expedite the query. | | | | | |
| Unused Index | The index has not been used in any execution plans recently. | | | | | |
Further reading
Additional helpful documentation, links, and articles:
