Predictive scaling overview in AWS - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

How does AWS Predictive Scaling forecast traffic?

AWS Predictive Scaling uses historical data to forecast future traffic patterns. Which method best describes how it predicts scaling needs?

A It uses machine learning models analyzing past usage trends to predict future demand.

B It randomly scales instances based on current CPU usage without historical data.

C It duplicates the current number of instances every hour regardless of traffic.

D It scales only when manual intervention triggers alarms.

💡 Hint

Think about how a system can prepare ahead by learning from past patterns.

Predictive Scaling uses machine learning to analyze historical traffic and forecast future needs, allowing proactive scaling.

❓ Architecture

intermediate

Which AWS service combination enables Predictive Scaling for EC2 Auto Scaling groups?

To implement Predictive Scaling for an EC2 Auto Scaling group, which combination of AWS services is required?

A Amazon EC2 with AWS Elastic Beanstalk and Amazon SNS.

B Amazon S3 with AWS Lambda and Amazon RDS.

C Amazon EC2 Auto Scaling with Amazon CloudWatch and AWS Auto Scaling predictive policies.

D AWS Glue with Amazon Athena and Amazon QuickSight.

💡 Hint

Focus on services related to scaling and monitoring EC2 instances.

Predictive Scaling uses EC2 Auto Scaling groups with CloudWatch metrics and predictive scaling policies to forecast and adjust capacity.

❓ service_behavior

advanced

What happens if Predictive Scaling forecast is inaccurate?

If AWS Predictive Scaling forecasts demand incorrectly, what is the expected behavior of the Auto Scaling group?

A The Auto Scaling group deletes all instances and restarts with default capacity.

B The Auto Scaling group stops scaling until manual intervention occurs.

C The Auto Scaling group permanently disables scaling policies.

D The Auto Scaling group falls back to reactive scaling based on real-time CloudWatch alarms.

💡 Hint

Consider how AWS ensures availability even if predictions are off.

If predictions are wrong, Auto Scaling uses reactive scaling triggered by CloudWatch alarms to adjust capacity dynamically.

❓ security

advanced

Which IAM permission is essential for enabling Predictive Scaling on an Auto Scaling group?

To allow AWS Predictive Scaling to function, which IAM permission must be granted to the Auto Scaling service role?

A "s3:ListBucket" to access S3 buckets for logs.

B "autoscaling:PutScheduledUpdateGroupAction" to create scheduled scaling actions.

C "ec2:TerminateInstances" to delete instances manually.

D "lambda:InvokeFunction" to trigger Lambda functions on scaling events.

💡 Hint

Think about permissions related to scheduling scaling actions.

Predictive Scaling creates scheduled scaling actions, so the service role needs permission to put these actions.

✅ Best Practice

expert

How should you configure Predictive Scaling to handle sudden unexpected traffic spikes?

Given that Predictive Scaling forecasts based on historical data, what is the best practice to ensure your application handles sudden unexpected traffic spikes?

A Combine Predictive Scaling with reactive scaling policies triggered by CloudWatch alarms for immediate response.

B Rely solely on Predictive Scaling and disable reactive scaling to avoid conflicts.

C Set Predictive Scaling to scale down aggressively to save costs during spikes.

D Manually increase instance count before expected spikes and disable all automatic scaling.

💡 Hint

Consider how to cover both forecasted and unforecasted demand.

Combining predictive and reactive scaling ensures the system can handle both expected and unexpected traffic changes effectively.