Why auto scaling matters in AWS - See It in Action

Why Auto Scaling Matters

📖 Scenario: You are managing a website that gets different amounts of visitors at different times. Sometimes many people visit at once, and other times only a few. You want your website to work well all the time without spending too much money.

🎯 Goal: Build a simple AWS Auto Scaling setup that adjusts the number of servers automatically based on the website traffic.

📋 What You'll Learn

Create a launch configuration for the server

Set a minimum and maximum number of servers

Create an Auto Scaling group using the launch configuration

Set a scaling policy to add or remove servers based on CPU usage

💡 Why This Matters

🌍 Real World

Websites and applications often have changing traffic. Auto Scaling helps keep them fast and available without wasting money.

💼 Career

Cloud engineers and DevOps professionals use Auto Scaling to build reliable and cost-effective cloud systems.

1

Create a launch configuration

Write the AWS CLI command to create a launch configuration named MyLaunchConfig using the Amazon Linux 2 AMI with ID ami-0abcdef1234567890 and instance type t2.micro.

AWS

# Your AWS CLI command here to create launch configuration

Hint

Use the aws autoscaling create-launch-configuration command with the exact names and IDs given.

2

Set minimum and maximum server counts

Create variables named min_size and max_size and set them to 1 and 3 respectively to define the minimum and maximum number of servers in the Auto Scaling group.

AWS

aws autoscaling create-launch-configuration --launch-configuration-name MyLaunchConfig --image-id ami-0abcdef1234567890 --instance-type t2.micro

# Define min_size and max_size variables
# Your code here

Hint

Just assign the numbers 1 and 3 to variables named exactly min_size and max_size.

3

Create the Auto Scaling group

Write the AWS CLI command to create an Auto Scaling group named MyAutoScalingGroup using the launch configuration MyLaunchConfig, with minimum size 1, maximum size 3, and availability zone us-east-1a.

AWS

aws autoscaling create-launch-configuration --launch-configuration-name MyLaunchConfig --image-id ami-0abcdef1234567890 --instance-type t2.micro
min_size=1
max_size=3

# Create Auto Scaling group command
# Your code here

Hint

Use the aws autoscaling create-auto-scaling-group command with the exact names and sizes.

4

Add a scaling policy based on CPU usage

Write the AWS CLI command to create a scaling policy named ScaleUpPolicy for the Auto Scaling group MyAutoScalingGroup that adds 1 instance when triggered.

AWS

aws autoscaling create-launch-configuration --launch-configuration-name MyLaunchConfig --image-id ami-0abcdef1234567890 --instance-type t2.micro
min_size=1
max_size=3
aws autoscaling create-auto-scaling-group --auto-scaling-group-name MyAutoScalingGroup --launch-configuration-name MyLaunchConfig --min-size 1 --max-size 3 --availability-zones us-east-1a

# Create scaling policy command
# Your code here

Hint

Use the aws autoscaling put-scaling-policy command with the exact group and policy names and scaling adjustment.

Practice

1. Why is auto scaling important in cloud infrastructure?

easy

A. It automatically adjusts the number of servers to match demand.

B. It manually requires admins to add servers during traffic spikes.

C. It reduces the need for backups and data recovery.

D. It guarantees zero downtime without any configuration.

Solution

1. Step 1: Understand auto scaling purpose

   Auto scaling automatically changes server count based on demand to keep apps running smoothly.

2. Step 2: Compare options

   Only It automatically adjusts the number of servers to match demand. describes automatic adjustment. Others describe manual work, unrelated features, or guarantees that are not true.

3. Final Answer:

   It automatically adjusts the number of servers to match demand. -> Option A

4. Quick Check:

   Auto scaling = automatic server adjustment [OK]

Hint: Auto scaling means automatic server count changes [OK]

Common Mistakes:

• Thinking auto scaling is manual
• Confusing auto scaling with backups
• Assuming zero downtime without setup

2. Which AWS service is used to set up auto scaling for EC2 instances?

easy

A. AWS Lambda

B. AWS CloudTrail

C. Amazon EC2 Auto Scaling

D. Amazon S3

Solution

1. Step 1: Identify service for auto scaling EC2

   AWS provides a specific service called Amazon EC2 Auto Scaling to manage scaling of EC2 instances.

2. Step 2: Eliminate unrelated services

   AWS Lambda runs code, S3 stores files, CloudTrail logs activity. None handle auto scaling EC2.

3. Final Answer:

   Amazon EC2 Auto Scaling -> Option C

4. Quick Check:

   EC2 auto scaling service = Amazon EC2 Auto Scaling [OK]

Hint: EC2 auto scaling uses Amazon EC2 Auto Scaling service [OK]

Common Mistakes:

• Confusing Lambda with auto scaling
• Choosing storage or logging services
• Not knowing specific auto scaling service

3. Given this auto scaling policy: scale out when CPU > 70%, scale in when CPU < 30%. If CPU usage is 75%, what happens?

medium

A. The number of servers increases.

B. The number of servers decreases.

C. No change in server count.

D. The servers are replaced.

Solution

1. Step 1: Analyze the scale out condition

   The policy says to add servers when CPU usage is above 70%. CPU is 75%, so condition is met.

2. Step 2: Determine action

   Since CPU is high, auto scaling will increase server count to handle load.

3. Final Answer:

   The number of servers increases. -> Option A

4. Quick Check:

   CPU > 70% triggers scale out = increase servers [OK]

Hint: High CPU triggers scale out (add servers) [OK]

Common Mistakes:

• Confusing scale out with scale in
• Thinking servers are replaced automatically
• Assuming no change on high CPU

4. You set an auto scaling policy but notice your app still crashes during traffic spikes. What is a likely cause?

medium

A. The app does not need more servers during spikes.

B. Auto scaling automatically fixes all crashes without setup.

C. Auto scaling only works for storage, not servers.

D. The scaling policy thresholds are set too high, delaying scaling.

Solution

1. Step 1: Understand scaling delay impact

   If thresholds are too high, scaling triggers late, causing crashes during spikes.

2. Step 2: Eliminate incorrect options

   Auto scaling needs proper setup, does not fix all crashes automatically. It works for servers, not just storage. App likely needs more servers during spikes.

3. Final Answer:

   The scaling policy thresholds are set too high, delaying scaling. -> Option D

4. Quick Check:

   High thresholds delay scaling = crashes [OK]

Hint: Check if scaling thresholds are too high [OK]

Common Mistakes:

• Assuming auto scaling fixes crashes automatically
• Thinking auto scaling only manages storage
• Ignoring app server needs during spikes

5. You want to optimize costs by scaling down servers during low traffic but avoid frequent scaling actions. Which approach is best?

hard

A. Set aggressive scale-in thresholds with short cooldown periods.

B. Use moderate scale-in thresholds with longer cooldown periods.

C. Disable scale-in to keep servers always running.

D. Manually add and remove servers daily.

Solution

1. Step 1: Understand cost optimization and stability

   To save costs, scale down servers when traffic is low but avoid too frequent scaling to prevent instability.

2. Step 2: Evaluate options

   Aggressive thresholds and short cooldown cause frequent scaling (not good). Disabling scale-in wastes money. Manual scaling is inefficient. Moderate thresholds with longer cooldown balance cost and stability.

3. Final Answer:

   Use moderate scale-in thresholds with longer cooldown periods. -> Option B

4. Quick Check:

   Moderate thresholds + longer cooldown = cost and stability [OK]

Hint: Balance thresholds and cooldown for cost and stability [OK]

Common Mistakes:

• Setting too aggressive scale-in causing flapping
• Disabling scale-in wastes money
• Relying on manual scaling