Regarding re:post article "How do I use CloudWatch metrics to calculate the average throughput and average number of IOPS that my EBS volume provides?"

You're absolutely right to point out this potential confusion. When calculating average throughput and IOPS using CloudWatch metric math expressions, the period selection is important for accurate results.

The default monitoring frequency for EBS volumes is 5 minutes, which means metrics are collected and aggregated at 5-minute intervals by default. When you select a 1-minute period in your metric math calculation, you may see unexpected results because the underlying data points are still being collected at 5-minute intervals - the system is essentially interpolating or displaying the same 5-minute data point across multiple 1-minute intervals.

For the most accurate calculations when using the standard monitoring configuration, you should align your period selection with the actual data collection frequency, which is 5 minutes by default. Your recommendation to explicitly state "Default value is 5 minutes" would help customers understand this important detail and avoid confusion.

It's worth noting that newer metrics like VolumeAvgIOPS and VolumeAvgThroughput are available at 1-minute frequency by default for all EBS volumes attached to Nitro instances, but for the traditional metric math approach using metrics like EBSReadOps, EBSWriteOps, VolumeReadBytes, and VolumeWriteBytes, the 5-minute default period remains the standard unless detailed monitoring is specifically enabled.

Your suggested clarification would definitely help customers avoid this common pitfall and get accurate results from their metric math expressions.
Sources
Understand discrepancies in Amazon EBS volume metrics | AWS re:Post
Amazon EBS introduces additional performance monitoring metrics for EBS volumes - AWS
Troubleshoot Amazon EBS performance issues on EC2 | AWS re:Post
Amazon CloudWatch metrics for Amazon EBS - Amazon EBS