Auto Scaling Groups

Auto Scaling is one of EC2's most powerful features. An Auto Scaling Group (ASG) automatically adjusts the number of EC2 instances based on demand, ensuring you have the right amount of capacity at all times — no more, no less. This lesson covers ASG concepts, scaling policies, health checks, and best practices.

---

What Is an Auto Scaling Group?

An Auto Scaling Group is a logical grouping of EC2 instances that share the same configuration and scaling rules. The ASG:

1. Launches new instances when demand increases
2. Terminates instances when demand decreases
3. Replaces unhealthy instances automatically
4. Distributes instances across Availability Zones for high availability

Core Parameters

Parameter	Description
Minimum size	The fewest instances the ASG will maintain (floor)
Maximum size	The most instances the ASG can scale to (ceiling)
Desired capacity	The number of instances the ASG currently targets
Launch template	The configuration used to launch new instances (AMI, instance type, key pair, security groups, user data)
Availability Zones	The AZs across which instances are distributed
Health check type	EC2 (instance status) or ELB (load balancer health check)
Health check grace period	Time to wait before checking a new instance's health (allows time for startup)

How the ASG Maintains Capacity

The ASG constantly monitors its instances and takes action:

If current capacity < desired capacity → Launch new instances
If current capacity > desired capacity → Terminate excess instances
If an instance is unhealthy           → Terminate and replace it

Instances are distributed as evenly as possible across the configured Availability Zones. If one AZ becomes unavailable, the ASG launches replacement instances in the remaining AZs.

---

Creating an Auto Scaling Group

Step 1: Create a Launch Template

A Launch Template defines the instance configuration:

aws ec2 create-launch-template \
  --launch-template-name my-web-app \
  --version-description "v1.0" \
  --launch-template-data '{
    "ImageId": "ami-0abcdef1234567890",
    "InstanceType": "t3.medium",
    "KeyName": "my-key",
    "SecurityGroupIds": ["sg-0123456789abcdef0"],
    "UserData": "'$(base64 -w0 <<'USERDATA'
#!/bin/bash
yum update -y
yum install -y httpd
systemctl start httpd
systemctl enable httpd
USERDATA
)'"
  }'

Step 2: Create the ASG

aws autoscaling create-auto-scaling-group \
  --auto-scaling-group-name my-web-asg \
  --launch-template LaunchTemplateName=my-web-app,Version='$Latest' \
  --min-size 2 \
  --max-size 10 \
  --desired-capacity 2 \
  --vpc-zone-identifier "subnet-0abc123,subnet-0def456" \
  --health-check-type ELB \
  --health-check-grace-period 300 \
  --target-group-arns "arn:aws:elasticloadbalancing:us-east-1:123456789012:targetgroup/my-tg/abc123"

---

Scaling Policies

Scaling policies determine when and how the ASG adjusts capacity.

1. Target Tracking Scaling

The simplest and most commonly used policy. You define a target metric value, and the ASG automatically adjusts capacity to maintain it.

aws autoscaling put-scaling-policy \
  --auto-scaling-group-name my-web-asg \
  --policy-name cpu-target-tracking \
  --policy-type TargetTrackingScaling \
  --target-tracking-configuration '{
    "PredefinedMetricSpecification": {
      "PredefinedMetricType": "ASGAverageCPUUtilization"
    },
    "TargetValue": 50.0
  }'

Common target metrics:

Metric	Description
ASGAverageCPUUtilization	Average CPU across all instances
ASGAverageNetworkIn	Average inbound network bytes
ASGAverageNetworkOut	Average outbound network bytes
ALBRequestCountPerTarget	Average requests per target in a target group
Custom metric	Any CloudWatch metric you define

2. Step Scaling

Define different scaling actions for different alarm thresholds:

CPU Range	Action
70-80%	Add 1 instance
80-90%	Add 2 instances
> 90%	Add 3 instances
< 30%	Remove 1 instance

3. Simple Scaling