While studying for the AWS Solutions Architect Professional exam, I wanted to memorize the Auto Scaling Metrics. It was suggested that knowing these metrics cold would be helpful on the exam.
During a review, it became obvious that for Auto Scaling metrics, there were three broad categories of Size, Capacity, and Instances. Each Category could be further explained by some Detail.
| Metric | Category | Detail |
| GroupMinSize | Size | Min |
| GroupMaxSize | Size | Max |
| GroupDesiredCapacity | Capacity | Desired |
| GroupInServiceInstances | Instances | Desired |
| GroupPendingInstances | Instances | Desired |
| GroupStandbyInstances | Instances | Desired |
| GroupTerminatingInstances | Instances | Desired |
| GroupTotalInstances | Instances | Desired |
For the Instances Category, there are some interesting distinctions to be made among the set. InService discludes pending and terminating. Total includes InService, Pending and Terminating. Standby includes instances running, but not InService.
Recall from Lifecycle Hooks the general states for an EC2 instance:
Pending –> InService –>Terminating.
If we then order the metrics: GroupPendingInstances, GroupInServiceInstance and GroupTerminatingInstances, we can then easily add the remaining Total and Standby metrics for overall clarity.
Photo by Chris Liverani on Unsplash
g-blog 