This lesson's interactive features are locked, please to keep using them
Auto Scaling Groups
The process we've followed works fine for one server, but what if we needed hundreds of servers?
I was working with a school district that needed hundreds of servers, but during the weekend they only needed two. Or during the summer it was one. We need a way to handle this.
Auto Scaling Groups (ASGs) help with this. We can give AWS an AMI and tell it how many instances we want.
This gets even more powerful as we instrument our servers. If a server starts to get busy, AWS can automatically provision more. If servers are sitting idle, we can reduce the server count.
Another thing I use these for is to rebuild a server if it crashes (an ASG of 1). My cat photo server is not known for its reliability, and I'm not known for patience. So if my single server crashes, the auto scaling group throws it away and starts a fresh one from the AMI.
Cost check: The ASG itself doesn't cost anything, but the servers do. Each instance costs about $0.0104/hr (~$7.60/mo for t3.micro). For cost reasons, we won't create an ASG in this course, but now you know where to look when you need it.