AWS RDS dashboard

The Amazon Relational Database Service (RDS) dashboard in Cloud Provider Observability shows CloudWatch metrics for DB instances and Aurora clusters in the following tabs:

Overview
Aurora
Rightsizing

To open the dashboard, go to Observability > Cloud provider > AWS > AWS RDS.

View DB instances at a high level

The Overview tab provides a high-level look at DB instances across your accounts and regions within a selected time frame.

At the top of the page, tables highlight instances that may need attention:

CPU >85% instances lists instances with high CPU utilization.
ReadIOPS >200 instances lists instances with elevated read I/O.

Summary stats show DB instances and Clusters counts, along with Average Burst Balance across instances.

The DB fleet overview table lists instances with sortable metrics that include CPU utilization, read and write latency, read and write IOPS, replica lag, burst balance, swap usage, and EBS balance percentages. Default columns also include Account ID, Instance ID, Cluster, Region, Job, and tag fields.

When an instance belongs to an Aurora cluster, the Cluster column shows the cluster name as a link. Click the cluster name to open the Aurora cluster drilldown for that cluster.

Use the search box to filter by job, account, region, or instance. Refine the view with Datasource, Job, Account, Region, and DB Instance filters, and adjust the time range as needed.

Monitor Aurora clusters

The Aurora tab monitors Aurora clusters across your accounts and regions. If panels show no data, confirm that CloudWatch metrics are configured for AWS/RDS and that tags are enabled. Click Necessary metrics in the banner to view the metrics required for the dashboard.

Use Datasource, Job, Account, Region, and DB Cluster filters to scope the view. Search by job, account, region, or cluster to find specific clusters quickly.

Review the cluster overview table

The Cluster overview table summarizes current storage, capacity, and I/O metrics for Aurora clusters. Columns include:

Account ID
Cluster (links to the cluster drilldown)
Region
Job
Capacity (avg ACUs) and Capacity (max ACUs)
Volume Used (avg)
Read IOPs (avg) and Write IOPs (avg)

Explore cluster metrics over time

Below the table, time series panels show Aurora cluster metrics for the selected filters and time range. Each panel includes an Explore button to open the underlying query.

Panels on the Aurora tab include:

Serverless Database Capacity: Number of Aurora Capacity Units allocated to each cluster (average and maximum).
ACU Utilization: Percentage of allocated capacity units actively in use.
Database Connections: Total active connections across all instances in each cluster.
Deadlocks: Average number of deadlocks per second.
Buffer Cache Hit Ratio: Percentage of read requests served from the buffer cache.
Commit Latency: Average time to complete a commit operation.
Temp Storage IOPs: I/O operations on local temporary storage. Aurora Serverless v2 only.
Temp Storage Throughput: Bytes per second transferred to or from local temporary storage. Aurora Serverless v2 only.

Drill into cluster detail

From the Cluster overview table on the Aurora tab, click a cluster name to open a drilldown page scoped to that cluster. You can also reach the drilldown from the Cluster column on the Overview tab.

The cluster drilldown shows time series panels for a single cluster, including:

Serverless Database Capacity: Capacity units allocated to the cluster over time (average and maximum).
ACU Utilization: Percentage of allocated capacity units in use.
Database Connections: Total active connections to the cluster.
Deadlocks: Average number of deadlocks per second.
Volume Bytes Used: Total storage consumed by the shared Aurora volume for the cluster.
Volume IOPs: Read and write I/O operations on the shared Aurora storage volume.
Buffer Cache Hit Ratio: Percentage of read requests served from the buffer cache.
Commit Latency: Average time to complete a commit operation.
Temp Storage IOPs and Temp Storage Throughput: Local temporary storage metrics. Aurora Serverless v2 only.

Refine the drilldown with Datasource, Job, Account, and Region filters.

Drill into instance detail

From the DB fleet overview table on the Overview tab, click an Instance ID or Job link to open an instance drilldown page for that DB instance.

The instance drilldown shows CloudWatch metrics for the selected instance within your time range. To customize which metrics appear on instance drilldown pages, refer to Customize drilldown metrics.

For preconfigured dashboards that include drilldown information for RDS instances, Cloud Provider Observability integrates with Database Observability. Refer to Add cloud provider integrations for setup configuration.

View resources for optimization

On the Rightsizing tab, review DB instance utilization to identify underused instances or instances with high utilization.

Troubleshoot missing data

RDS pages and panels require specific CloudWatch metrics to display data correctly. If columns or panels show no data, try the following steps.

On the Aurora tab, click Necessary metrics in the informational banner to view the metrics required for the dashboard.
Navigate to the Configuration Details tab for your CloudWatch metrics scrape job.
Click Edit next to the scrape job you want to change.
Expand the metrics details for AWS/RDS.
Click Reset to defaults to restore the metrics required for RDS dashboards.
Click Save service settings.

Aurora-specific panels on the Aurora tab and cluster drilldown rely on cluster-level metrics such as ServerlessDatabaseCapacity, ACUUtilization, VolumeBytesUsed, VolumeReadIOPs, VolumeWriteIOPs, BufferCacheHitRatio, CommitLatency, Deadlocks, TempStorageIOPs, and TempStorageThroughput. Temp storage panels apply to Aurora Serverless v2 only.