SigNoz Cloud - This page is relevant for SigNoz Cloud editions.
Self-Host - This page is relevant for self-hosted SigNoz editions.

AWS Monitoring - FAQs

Why doesn't my AWS metric have an average statistic form?

Some AWS metrics collected through CloudWatch may not always export an average statistic directly from the collector. This is because the OpenTelemetry Collector exports metrics using the statistic forms provided by AWS CloudWatch Metric Streams.

Available metric forms:

AWS CloudWatch provides the following statistic forms for most metrics:

  • min - Minimum value
  • max - Maximum value
  • sum - Sum of all values
  • count - Number of data points

To calculate the average:

You can calculate the average manually using the formula:

average = sum / count

This will give you the same result as the average statistic you would see in AWS CloudWatch console.

Example: EC2 CPU Utilization

For the aws_EC2_CPUUtilization metric:

  1. Use the formula: aws_EC2_CPUUtilization_sum / aws_EC2_CPUUtilization_count
  2. This calculates the average CPU utilization percentage
  3. The result matches the average metric shown in CloudWatch

Why this approach?

The available metric forms (min/max/sum/count) follow the AWS CloudWatch Metric Streams standard. For detailed information about how these metrics are calculated and exported, refer to:

This approach allows you to:

  • Migrate alarms that previously used the average statistic in CloudWatch
  • Create custom calculations and aggregations
  • Maintain the same monitoring functionality across platforms

Last updated: February 6, 2026

Edit on GitHub

Was this page helpful?