Regarding aws auto scaling group

#1

Aws provide a feature known as auto scaling group which automatically scale in or out the number of instances if the load on the application occurs or if one of the instances of the application is down.

Does this feature can be listed in orchestrator of instances?

#2

Hi @tushar4530

I suppose we could call an Autoscaling Group an “orchestrator of instances”, since that is pretty much what it does.

It has three important settings.

  • Minimum Capacity - It will try to not let fewer instances than this be in service
  • Desired Capacity - It will try to maintain this number of instances in service under normal load.
  • Max Capacity - It will not provision more than this number of instances when it scales out under load. Without this, if a bad actor did a denial of service attack on you, then you could end up paying for a potentially unlimited number of instances

If an instance fails the AWS health checks, then it will take that instance out of service and replace with a fresh one.

It also plays a part in AWS Elastic Beanstalk (a managed application deployment system), and in managing the number of nodes in a Kubernetes cluster when coupled with Kubernetes cluster autoscaler product.

All in all, a useful service!

#3

Thank you for taking time and replying for this silly question. Your answer helped a lot.