Monitoring RDS service performance
Visualize current and historical resource usage across a variety of metrics like CPU, freeable memory, swap, DB connections, and network throughput to gain insight into the operational health of the instances powering your relational databases in the cloud.
Gain a comprehensive view of the health of your Aurora DB clusters. Collect metrics that monitor data for your primary DB instance, read replicas, and cluster volumes, and track changes in usage patterns to make informed decisions on performance and scaling.
Collect native metrics for relational DB engines like MySQL, PostgreSQL, and MariaDB in addition to standard CloudWatch data. Deploy our server agent on an EC2 instance, configure the RDS DB instance endpoint, and execute our plugin to get started.
Monitor Amazon RDS database
Is your DB instance running out of disk space? Set up thresholds and alerts and track the metric free storage to make sure you get notified before the storage space is exhausted.
IOPS and throughput
Are your General Purpose (SSD) storage volumes providing the right performance for your workloads? Monitor read/write activity and throughput to make informed decisions about volume migration.
Check how frequently your storage volume is bursting above baseline performance. Visualize usage, identify trends, and increase storage capacity to meet the demands of your workload.
Latency and disk queue length
Is your queue depth consistently high or low? Compare and correlate latency with disk queue depth to identify storage bottlenecks and stay ahead of potential issues.
Metrics for your PostgreSQL DB engine
Troubleshoot PostgreSQL read replica issues
Track replication slot lag to see how far behind the read replica is lagging when it comes to received WAL data. Also, check transaction log disk usage to keep an eye on used storage.
Monitor the transaction ID health
Set up thresholds to track the number of in-flight, unvacuumed transactions and take immediate action to fix the problem before your database goes to read-only mode.
Key Features for RDS Enhanced Monitoring
Configure intelligent thresholds with multiple conditions and strategies to improve the quality of your alerts, and get notified via integrated notification channels.
Group data points of a single performance metric from multiple DB instances into a single time-series chart to quickly answer key questions.
Identify resource usage patterns and trends, and know when it's time to migrate to a higher DB instance class or change storage type.
Identify DB instances with low connection stats, and automatically execute an API call to shut them down or invoke a Lambda function to take a DB snapshot.
Use dynamic baselines to help detect outliers and anomalies in your DB instance performance with improved accuracy.