AWS Certified Solutions Architect Professional SAP-C02 Practice Question
A company runs a multi-AZ web application on Amazon EC2 instances behind an Application Load Balancer. The instances are in an Auto Scaling group (ASG). A CloudWatch metric alarm currently notifies the on-call channel whenever the ASG's average CPU utilization exceeds 85 percent, after which an engineer manually increases the desired capacity. Management wants this alert to remediate itself automatically while reducing unnecessary scale-out events caused by short CPU spikes. A notification must still be sent whenever the remediation occurs.
Which solution meets these requirements and follows AWS operational-excellence best practices?
Replace the metric alarm with two scheduled scaling actions that raise desired capacity every weekday at 09:00 and lower it at 18:00 while sending an SNS notification when each action runs.
Create a CloudWatch metric alarm on the ASG's average CPU with a 60-second period and three evaluation periods. Attach a step-scaling policy that adds one instance when the alarm is in ALARM state, and configure the alarm to publish to an SNS topic used by the on-call channel.
Enable predictive scaling for the ASG, delete the CPU alarm, and rely on predictive scaling to add capacity ahead of demand. Configure predictive scaling to send an SNS notification whenever it forecasts a capacity change.
Create an EventBridge rule that runs an AWS Lambda function every minute. The function calls GetMetricData for CPU utilization and, if it is above 85 percent, invokes SetDesiredCapacity to add one instance and then sends a notification to the on-call channel.
Configuring a CloudWatch metric alarm to invoke a step-scaling policy satisfies all requirements. The alarm's three 60-second evaluation periods smooth out brief spikes, lowering the chance of false scale-outs. When the alarm enters ALARM state, the attached step-scaling policy automatically adds an instance to the ASG, providing the required remediation. Publishing the same alarm to an SNS topic continues to notify the operations chat channel.
Scheduled actions (option B) ignore real-time load and cannot react to unexpected traffic. Predictive scaling (option C) may forecast demand but, by removing the alarm, no longer guarantees a response to sudden spikes or sends an alert when remediation happens. A custom Lambda loop (option D) duplicates native Auto Scaling capabilities, adds operational overhead, and introduces a potential single point of failure.
Ask Bash
Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.
What is a CloudWatch metric alarm and how does it work with Auto Scaling?
Open an interactive chat with Bash
What is the difference between step-scaling and predictive scaling?
Open an interactive chat with Bash
How does using multiple evaluation periods in a CloudWatch alarm reduce unnecessary scaling?
Open an interactive chat with Bash
AWS Certified Solutions Architect Professional SAP-C02
Continuous Improvement for Existing Solutions
Your Score:
Report Issue
Bash, the Crucial Exams Chat Bot
AI Bot
Loading...
Loading...
Loading...
IT & Cybersecurity Package Join Premium for Full Access