Load Balancing & Auto Scaling

Elastic Load Balancing (ELB) distributes incoming traffic across multiple targets (EC2 instances, Lambda, containers). Auto Scaling automatically adjusts the number of instances based on demand.

Analogy: A load balancer is like a smart receptionist at a busy office — they direct visitors to the right person. Auto scaling is like hiring or letting go of staff based on how busy the office is.

Load Balancer + Auto Scaling Flow

sequenceDiagram
    participant User as Internet Users
    participant DNS as Route53
    participant ELB as Load Balancer
    participant ASG as Auto Scaling Group
    participant EC2_1 as EC2 Instance
    participant EC2_2 as EC2 Instance
    participant EC2_3 as EC2 Instance (new)

    User->>DNS: www.example.com
    DNS-->>User: ELB DNS name

    User->>ELB: HTTP request
    ELB->>EC2_1: Forward request
    EC2_1-->>ELB: Response
    ELB-->>User: Response

    Note over ELB,ASG: Traffic spikes...

    ASG->>ASG: CPU > 70% threshold
    ASG->>ASG: Launch new instance
    ASG->>EC2_3: Start + configure
    EC2_3->>ELB: Health check passes ✅
    ELB->>EC2_3: Now routing traffic here too

    Note over ELB,ASG: Traffic drops...

    ASG->>ASG: CPU < 30% threshold
    ASG->>EC2_3: Terminate instance
    ELB->>EC2_3: Stop routing

ELB — Elastic Load Balancer

Three types of load balancers:

Type	Protocol	Best For
ALB (Application LB)	HTTP/HTTPS, WebSocket	Web apps, microservices
NLB (Network LB)	TCP/UDP	Ultra-high performance, static IP
CLB (Classic LB)	HTTP/HTTPS/TCP	Legacy apps (avoid for new projects)

ALB features:

Path-based routing (/api/* → one target group, /web/* → another)
Host-based routing (api.example.com → one group, app.example.com → another)
WebSocket and HTTP/2 support
Integrated with AWS WAF (web application firewall)

# Create an ALB
aws elbv2 create-load-balancer \
    --name my-alb \
    --subnets subnet-123 subnet-456 \
    --security-groups sg-789

# Create target group
aws elbv2 create-target-group \
    --name my-targets \
    --protocol HTTP \
    --port 80 \
    --vpc-id vpc-123

Auto Scaling Group

Auto Scaling ensures:

Min size — always run at least N instances
Max size — never exceed M instances (cost control)
Desired capacity — start with N instances

# Create launch template (AMI, instance type, etc.)
aws ec2 create-launch-template \
    --launch-template-name my-template \
    --launch-template-data '{"ImageId":"ami-123","InstanceType":"t3.micro"}'

# Create auto scaling group
aws autoscaling create-auto-scaling-group \
    --auto-scaling-group-name my-asg \
    --launch-template LaunchTemplateName=my-template \
    --min-size 2 \
    --max-size 10 \
    --desired-capacity 3 \
    --vpc-zone-identifier subnet-123,subnet-456

# Add scaling policy (target tracking)
aws autoscaling put-scaling-policy \
    --auto-scaling-group-name my-asg \
    --policy-name cpu-target \
    --policy-type TargetTrackingScaling \
    --target-tracking-configuration '{"TargetValue":70.0,"PredefinedMetricSpecification":{"PredefinedMetricType":"ASGAverageCPUUtilization"}}'

ELB + Auto Scaling Architecture

flowchart TB
    Internet --> Route53
    Route53 --> ALB[Application Load Balancer]

    subgraph ASG["Auto Scaling Group"]
        direction LR
        EC2_A[EC2 Instance<br/>AZ A] --> ALB
        EC2_B[EC2 Instance<br/>AZ B] --> ALB
        EC2_C[EC2 Instance<br/>AZ C] --> ALB
    end

    ASG --> ASG_Logic[Scaling Policy<br/>CPU > 70% → add<br/>CPU < 30% → remove]

    style ALB fill:#7c3aed,color:#fff
    style ASG fill:#059669,color:#fff
    style Route53 fill:#3b82f6,color:#fff

Health Checks

ELB health checks ping instances every N seconds
If an instance fails 3 consecutive checks → it’s marked unhealthy
Auto Scaling replaces unhealthy instances automatically
Always implement a /health endpoint in your app

# Simple health check endpoint (Node.js Express example)
app.get('/health', (req, res) => {
  const dbConnected = checkDatabaseConnection();
  if (dbConnected) {
    res.status(200).json({ status: 'healthy' });
  } else {
    res.status(503).json({ status: 'unhealthy' });
  }
});

In Simple Words

ELB distributes traffic across instances — improves reliability and performance
ALB is the modern standard for HTTP/HTTPS apps (path/host-based routing)
Auto Scaling adds/removes instances automatically based on CPU, memory, or custom metrics
Together they provide high availability and cost efficiency
Always have at least 2 instances across 2 AZs for production
Health checks ensure traffic only goes to healthy instances