News
Amazon EBS Adds New Per-Minute IOPS and Throughput Metrics for Volume Performance Monitoring
Amazon Elastic Block Store (EBS) introduced two additional Amazon CloudWatch metrics for volume-level performance visibility: VolumeAvgIOPS and VolumeAvgThroughput. These metrics provide per-minute averages of the I/O operations per second and data throughput driven on an EBS volume, giving teams clearer insight into actual workload demand so they can track trends, detect bottlenecks, and right-size provisioned performance.
The new metrics are enabled by default at 1-minute frequency and come at no additional charge. They apply to EBS volumes attached to Amazon EC2 Nitro-based instances across all commercial AWS Regions, including AWS GovCloud (US) and AWS China Regions. With CloudWatch, customers can place the metrics on dashboards for ongoing visibility and configure alarms that notify on thresholds or trigger automated actions for remediation.
VolumeAvgIOPS exposes the average read/write operations per second driven to the volume each minute, while VolumeAvgThroughput reports the average bytes per second over the same interval. Together, they help operators understand if an application is approaching or exceeding the performance they have provisioned on a given volume and whether adjustments are needed to IOPS or throughput settings. This view is useful during load testing, incident triage, and capacity planning, where minute-by-minute demand informs decisions on scaling or tuning storage tiers.
Operationally, teams can add these metrics to existing monitoring flows alongside other EBS and EC2 indicators, then set CloudWatch alarms to alert when sustained IOPS or throughput cross defined thresholds. Because the data is available without extra configuration, adoption is as simple as querying the metrics in CloudWatch and integrating them into dashboards or alerting rules used by on-call runbooks.
About the Author
David Ramel is an editor and writer at Converge 360.