EC2 Auto Scaling

• Automatically add/remove EC2 instances based on certain conditions you set
• Ensures sufficient capacity at all times, without over-provisioning
• Scaling - “adding more instances”

Creating an Auto Scaling Group

• EC2 console > Auto Scaling Groups
• You create a group of rules that will ensure that a certain amount of EC2 instances are running based on certain conditions
  • Choose a launch template
    • a feature that tells AWS how your EC2 instances should be configured
    • since we won’t be creating them manually
    • Creating a launch template → same with EC2 launch wizard
    • You need to choose VPCs and subnets that can be used for launching
  • Choose a load balancer (Elastic Load Balancer (ELB)) + target group
    • make sure u choose the same VPCs and subnets that you chose in the previous step
    • Set up the rule for which kind of requests you want to listen
      • ex. HTTPs requests
      • Create a new target group - a collection of instances (EC2, containers, IP addresses, etc.) that receive traffic from a load balancer
        • group of instances that will be used for forwarding requests to
      • If a load balancer is connected to an auto scaling group, then it will automatically add all instances that are created by that auto scaling group to this target group
    • Health check
      • is an instance healthy - capable of receiving requests
  • Choose Auto scaling conditions
    • set the capacities you want to have for your instances
      • desired capacity - the default capacity you want to have
      • minimum/maximum capacity
    • scaling policy
      • define the metric where you want to scale up
      • CPU utilization, network, etc
  • Choose notifs
    • maybe an email so you know when an instance is added
  • Tags
    • identify resources