Make Autoscaling scale-out more quickly for customMetric

0

I'm setting up an Autoscaling group and it's using TargetTrackingScaling that is tracking a custom metric that is being updated every minute (via a scheduled lambda that queries the inservice instances).

It's working fine, but I'd like the scale-out action to happen more quickly. The source of this looks to be due to the ASG Alarm that gets auto created. It's looking for 3 datapoints before triggering, this means it takes 3 minutes for the alarm to fire.

Is there a way I can configure the ASG/Scaling policy such that the alarm only needs 1 datapoint to fire? Or if not can I create a custom alarm and use that instead?

The metric is {Unit: 'Percent', Statistic: 'Average'}

profile picture
ede
asked 2 years ago650 views
2 Answers
1

Few cents --> Are you looking for this feature on 'TargetTrackingPolicy'? To my knowledge, the CloudWatch alarms for this policy are managed by AWS and is recommended not to edit. https://docs.aws.amazon.com/autoscaling/ec2/userguide/as-scaling-target-tracking.html (refer Cosideration #1).

You could control the number of datapoints for 'Step' and 'Simple' scaling policies as you manage the CW alarm.

profile pictureAWS
answered 2 years ago
0

Yes I'm looking at TargetTrackingPolicy - will see if I can achieve what I want with Step or Simple policies.

profile picture
ede
answered 2 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions