AWS HealthOmics adds 14 real-time run metrics in CloudWatch

AWS announced on September 11, 2026, that HealthOmics now publishes 14 new real-time run metrics to Amazon CloudWatch. The update exposes resource utilization during execution of the service’s fully managed bioinformatics workflows, according to the AWS announcement.

Resource signals during execution

The metrics cover CPU and GPU usage, memory usage, file system usage and I/O, network throughput, and ephemeral storage usage. AWS says these signals can help teams identify compute bottlenecks, detect memory or storage exhaustion before a task fails, and track file system throughput without opening a support case.

AWS also says the metrics use the “Amazon CloudWatch OpenTelemetry standard,” supporting integration with third-party observability tools alongside native CloudWatch dashboards and alarms. These capabilities are described in the launch announcement.

Engineering implications: diagnose before resizing

The practical opportunity is to compare allocated resources with observed demand rather than resize workflows solely from completion times or failures. For example, teams could examine compute utilization alongside I/O activity before deciding whether a slow task needs more CPU capacity or closer investigation of its data access pattern.

Treat those comparisons as diagnostic clues, not proof of a root cause. A sensible rollout would establish baselines for representative workflows, then use repeated observations to guide configuration changes. Before designing time-sensitive alerts, verify the metric reporting cadence rather than treating “real-time” as a guaranteed refresh interval.

Availability and ingestion charges

As of the September 11 announcement, AWS listed availability in nine Regions:

• US East: N. Virginia and Ohio
• US West: Oregon
• Europe: Frankfurt, Ireland, and London
• Asia Pacific: Singapore, Seoul, and Tokyo

AWS says CloudWatch charges for these run metrics based on the volume of metric data ingested. Teams evaluating potential savings from right-sizing should therefore include telemetry ingestion in their cost calculations. Both the regional availability and charging basis are stated in the AWS announcement.