Amazon EC2 Auto Scaling Can Now Terminate Up to 100 Instances in One API Call!

Hey, I'm Shii-chan! I found some news today that'll make AWS operations a little easier!

AWS What's New aws.amazon.com

What was announced?

According to AWS What's New, Amazon EC2 Auto Scaling now supports batch instance termination, letting you terminate multiple instances in a single API call! You can now pass up to 100 instance IDs at once to the TerminateInstanceInAutoScalingGroup API.

The story so far

Until now, scaling down instances in an Auto Scaling group basically meant terminating one instance per API call, so if you needed to remove a large number of instances at once, you had to make that call over and over again.

What changes

For workloads like AI/ML training jobs, container orchestrators, or event-driven architectures that spin up large fleets temporarily, you can now cut down the number of API calls needed when it's time to clean up a lot of instances at once. Scaling down just got simpler!

Dive Deep

Let's go over the details of how batch termination works.

• All instances in a batch are validated atomically before termination begins
• Existing Auto Scaling behaviors, such as lifecycle hooks and load balancer connection draining, are preserved for each instance in the batch
• Available in all AWS Regions
• No additional cost

Wrap-up

• The Auto Scaling TerminateInstanceInAutoScalingGroup API can now terminate up to 100 instances in one call
• Existing behaviors like lifecycle hooks and load balancer connection draining are preserved
• Available in every region at no extra cost
• A handy feature for teams that frequently scale down large fleets, like AI/ML training or container workloads!