Answer-first summary for fast verification
Answer: Use CloudWatch metric math to calculate the approximate number of messages visible per instance; create a target tracking scaling policy.
For SQS consumers, target tracking on ApproximateNumberOfMessagesVisible divided by instance count (using metric math) to scale based on backlog per instance is recommended. Age of oldest is for latency but not direct scale. ALB not applicable for SQS.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
A worker process on three EC2 instances reads from an SQS queue. Random spikes in message volume decrease performance. Which scaling solution meets this requirement?
A
Use CloudWatch metric math to calculate the age of the oldest message; create a target tracking scaling policy.
B
Use CloudWatch metric math to calculate the approximate number of messages visible per instance; create a target tracking scaling policy.
C
Create an ALB; attach it to the Auto Scaling group; use ALBRequestCountPerTarget for target tracking.
D
Create an ALB; use a scheduled scaling policy.
No comments yet.