Lab: Designing for Performance with Auto Scaling and ElastiCache

This lab focuses on the "Democratize Advanced Technologies" and "Architecting for Performance" principles from the AWS Well-Architected Framework. You will build a performance-optimized environment using Amazon EC2 Auto Scaling and Amazon ElastiCache (Redis) to handle variable workloads and reduce data retrieval latency.

Prerequisites

Before starting this lab, ensure you have:

An AWS Account with administrative permissions.
AWS CLI installed and configured (aws configure).
Basic knowledge of VPC Networking (Subnets, Security Groups).
Target Region: us-east-1 (Virginia).

Learning Objectives

By the end of this lab, you will be able to:

Deploy a purpose-built caching layer using Amazon ElastiCache.
Configure an Elastic architecture using Launch Templates and Auto Scaling Groups (ASG).
Implement performance monitoring using Amazon CloudWatch alarms.
Evaluate performance design principles through hands-on implementation.

Architecture Overview

The following diagram illustrates the performant architecture you will deploy:

Loading Diagram...

Step-by-Step Instructions

Step 1: Create Security Groups

We need to allow traffic between the application instances and the cache cluster.

bash

# Create Web Security Group
aws ec2 create-security-group \
    --group-name lab-web-sg \
    --description "SG for Web Servers"

# Create Cache Security Group
aws ec2 create-security-group \
    --group-name lab-cache-sg \
    --description "SG for Redis Cache"

# Allow Web SG to talk to Cache SG on port 6379
aws ec2 authorize-security-group-ingress \
    --group-name lab-cache-sg \
    --protocol tcp \
    --port 6379 \
    --source-group lab-web-sg

▶Console Alternative

Navigate to

EC2 > Security Groups

. Create 'lab-web-sg'. Create 'lab-cache-sg' and add an Inbound Rule for Type: Redis (6379) with Source: 'lab-web-sg'.

Step 2: Deploy Amazon ElastiCache (Redis)

Following the principle of "Democratizing Advanced Technologies," we use a managed service instead of manual installation.

bash

aws elasticache create-cache-cluster \
    --cache-cluster-id lab-redis-cluster \
    --engine redis \
    --cache-node-type cache.t3.micro \
    --num-cache-nodes 1 \
    --security-group-ids <YOUR_CACHE_SG_ID>

[!IMPORTANT] Replace <YOUR_CACHE_SG_ID> with the ID returned from the previous step.

Step 3: Create Launch Template and Auto Scaling Group

This setup ensures the application can scale horizontally based on performance metrics.

bash

# Create Launch Template
aws ec2 create-launch-template \
    --launch-template-name lab-app-template \
    --launch-template-data '{"ImageId":"ami-0cff7528ff583bf9a","InstanceType":"t2.micro","SecurityGroupIds":["lab-web-sg"]}'

# Create ASG
aws autoscaling create-auto-scaling-group \
    --auto-scaling-group-name lab-asg \
    --launch-template LaunchTemplateName=lab-app-template \
    --min-size 1 \
    --max-size 3 \
    --desired-capacity 1 \
    --availability-zones us-east-1a us-east-1b

Step 4: Configure Target Tracking Scaling Policy

To meet performance objectives dynamically, we will scale based on average CPU utilization.

bash

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

Checkpoints

ElastiCache Status: Run aws elasticache describe-cache-clusters --cache-cluster-id lab-redis-cluster. The CacheClusterStatus should be available.
ASG Instances: Run aws autoscaling describe-auto-scaling-groups --auto-scaling-group-names lab-asg. Ensure at least one instance is InService.
CloudWatch Metrics: Check the CloudWatch console to see if the TargetTracking alarm has been automatically created by the scaling policy.

Concept Review

Design Principle	Implementation in Lab
Democratize Tech	Used Managed ElastiCache instead of managing our own Redis server on EC2.
Go Global in Minutes	Used Multi-AZ deployment for the Auto Scaling Group.
Elasticity	Implemented Target Tracking Scaling to respond to performance demand.
Monitoring	Utilized CloudWatch Metrics for automated scaling decisions.

Performance Optimization Cycle

\begin{tikzpicture}[node distance=2.5cm, auto, >=stealth] \node [circle, draw, text width=2cm, align=center] (monitor) {Monitor Performance}; \node [circle, draw, text width=2cm, align=center, right of=monitor, xshift=2cm] (analyze) {Analyze Metrics}; \node [circle, draw, text width=2cm, align=center, below of=analyze] (optimize) {Optimize Design}; \node [circle, draw, text width=2cm, align=center, left of=optimize, xshift=-2cm] (redeploy) {Redeploy Solution};

code

\draw [->, thick] (monitor) -- (analyze);
\draw [->, thick] (analyze) -- (optimize);
\draw [->, thick] (optimize) -- (redeploy);
\draw [->, thick] (redeploy) -- (monitor);

\node at (2.25,-1.25) [font=\small] {CloudWatch};
\node at (2.25,-3.75) [font=\small] {Right-Sizing};

\end{tikzpicture}

Troubleshooting

Issue	Possible Cause	Fix
Redis Connection Timeout	Security Group mismatch	Ensure `lab-cache-sg` allows port 6379 from `lab-web-sg`.
ASG Not Scaling	Insufficient load	The Target Tracking policy requires sustained CPU usage > 50% to trigger.
Cluster Creation Stuck	Subnet Group issues	Ensure your VPC has an ElastiCache Subnet Group if not using the default VPC.

Clean-Up / Teardown

[!WARNING] Failure to delete these resources will result in ongoing charges to your AWS account.

bash

# 1. Delete Auto Scaling Group
aws autoscaling delete-auto-scaling-group --auto-scaling-group-name lab-asg --force-delete

# 2. Delete Launch Template
aws ec2 delete-launch-template --launch-template-name lab-app-template

# 3. Delete ElastiCache Cluster
aws elasticache delete-cache-cluster --cache-cluster-id lab-redis-cluster

# 4. Delete Security Groups (wait for instances to terminate first)
aws ec2 delete-security-group --group-name lab-cache-sg
aws ec2 delete-security-group --group-name lab-web-sg

Stretch Challenge

Implement Read Replicas: Modify the ElastiCache setup to use a Replication Group instead of a single cluster. This improves performance for read-heavy workloads by offloading queries from the primary node.

Cost Estimate

EC2 (t2.micro): ~$0.0116/hour (Free Tier eligible).
ElastiCache (cache.t3.micro): ~$0.017/hour.
CloudWatch: Free for standard metrics used here.
Total Estimated Spend: < $0.50 for a 1-hour lab session.