Exam AWS Certified Solutions Architect - Associate SAA-C03 topic 1 question 75 discussion

Agree with A>>> Lambda = serverless + autoscale (modernize), SQS= decouple (no more drops)

The catch phrase is "scale up when communication failures are detected" Scaling should not be based on communication failures, that'll be crying over spilled milk ! or rather too late. So D is wrong.

B. Analyze performance with CloudWatch and increase EC2 instance size: Vertical scaling (increasing instance size) can address temporary performance issues but is less cost-effective and scalable. It does not modernize the architecture or address dropped transactions due to overloading. C. Use SNS with EC2 instances and Auto Scaling: While SNS can facilitate messaging, it is a publish/subscribe model, which is less suited for tier-to-tier decoupling. SQS is a better fit for reliable message queuing and processing. D. Use SQS with EC2 instances and Auto Scaling: While this addresses decoupling and scaling, it retains the burden of managing and scaling EC2 instances. A serverless solution like Lambda and API Gateway provides greater operational efficiency and modernization.

Clearly A, it significantly reduces operational work. All the other options involve some maintenance and cause additional operational work to maintain the EC2instances and ASG.

Using serverless services like API Gateway and AWS Lambda and SQS helps resolve this problem with the least operational overhead.

why its not c?

Ans A - keep it simple: API Gateway + Lambda + SQS D won't work: it relies upon a failure detected - by then its too late)

The answer choice D cannot be the answer because CloudWatch is not an appropriate way to monitor transaction failures. I would have been more confused about it if CloudWatch hadn't been mentioned.

RESTful services = API Gateways https://docs.aws.amazon.com/apigateway/latest/developerguide/welcome.html

Least operational overhead = API Gateway + Lambda + SQS BC wrong applications D: Will work but more operational overhead than A with less resilience to failures

You want to "modernize the application", that a modern application is serverless, in any case a modern application does NOT use EC2 instances. Also, managing EC2 instances (with the OS etc pp) is NOT "operationally efficient". Thus not D.

For me the solution could be based on using SNS with multiple topics to organize communication between different tiers ( Using Subscriber for one and Producer for another topic to proceed with transactions over multi-tiers ). CloudWatch to monitor SNS topics queue length and scale up/down based on counts of messages (NumberOfMessagesPublished)

A is the perfect answer no need for the ASG

D is better because in answer A there is a bottleneck on a SQS - service app, D is as operationally efficient as A and solves the above issue

ASG is not as efficient as Lambda!

I feel the answer is D, Lambda would increase the complexity and overhead and it has limitation of running for 15 min.

MOST operationally efficient = Serverless = AWS Lambda functions, Amazon Simple Queue Service