AWS Certified Solutions Architect Associate SAA-C03 Practice Question
An e-commerce web tier runs on an Auto Scaling group of Amazon EC2 instances behind an Application Load Balancer (ALB). The current scaling policy tracks only ASGAverageCPUUtilization at 50 percent. During flash-sale events, customers occasionally experience increased HTTP 5xx errors even though CPU utilization across the group stays under 40 percent.
Which change to the scaling configuration will BEST help maintain high availability during these peak events?
Replace CPU tracking with GroupDesiredCapacity so that scaling occurs whenever the desired instance count is low.
Track CPUCreditBalance and scale out when the credit balance drops below a threshold.
Add ALBRequestCountPerTarget as a target-tracking metric alongside Average CPU Utilization.
Continue to scale only on Average CPU Utilization because it directly reflects overall load.
CPU utilization does not always rise in direct proportion to request volume-especially for I/O-bound or partially idle workloads. Adding the ALBRequestCountPerTarget metric (or another throughput-oriented metric such as ASGAverageNetworkIn) supplies additional context about incoming traffic so that the Auto Scaling group can add capacity before back-end instances become saturated and start returning errors. Relying only on CPU or on inventory-style metrics such as GroupDesiredCapacity does not capture real-time demand. CPUCreditBalance applies only to burstable instance families and is not a reliable trigger for web-tier scaling on its own.
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 are some other metrics that can be used for auto-scaling besides CPU utilization?
Open an interactive chat with Bash
What is an auto-scaling group and how does it work in AWS?
Open an interactive chat with Bash
Why is high availability important in an e-commerce application?