Exam AWS Certified Solutions Architect - Associate SAA-C02 topic 1 question 395 discussion

'Application interruptions' - So application is missing inventory updates as it seems. If the distrubution center sends updates to SQS where it will be buffered if the application is facing interruptions. Application can then resume processing from SQS once it's back up and running. I'll go with D.

D. - Outages happen to the inventory API hosted on EC2 instances, NOT the on-prem shipping applications A. Modify the shipping application to write to a local database. - Wrong. The inventory API must be called to update the inventory, why would store it locally? B. Modify the application APIs to run serverless using AWS Lambda - Wrong. If the EC2 instance goes down after invoking the Lambda function, missed transactions will still happen C. Configure Amazon API Gateway to call the EC2 inventory application APIs. - Wrong. Doesn't matter who calls the API - the EC2 instances can still go down while processing the request, resulting in missed transactions D. Modify the application to send inventory updates using Amazon Simple Queue Service (Amazon SQS). - Correct. If the EC2 instance goes down while processing the request, the visibility timeout on the consumed message will expire, and become visible again for processing by another healthy instance

D is the answer

Instruction is clear for architect role

DDDDDDDDDDDDDD

D is correct

DDDDDDDDDDDDDDDDDD

How do you 'send' using SQS? It works on polling, so based on the wording cannot be D,must be B

propose to increase the resilience of an application? B make sense for this.

The answer is D, this question is about loosely couple architecture. Why SQS will store the message until it is delete.

Considering that the on-premise shipping application need to continue with Business of shipping label and can't rely/retry on API, it's best to write it in local database which can withstand any failure including any network issue (in case of SQS, it still has to make a API call, if network issue, this can't succeed and transaction would be lost)

Should be D. https://aws.amazon.com/sqs/features/ Resiliency: When part of your system fails, it doesn’t need to take the entire system down. Message queues decouple components of your system, so if a process that is reading messages from the queue fails, messages can still be added to the queue to be processed when the system recovers.

D - SQS allows messages to stay in queue until there's an available resource to process the message.

SQS can address the On-Prem Shipping App resiliency using App decoupling to solve the missed transaction issue. The API-based Inventory Reporting App then can poll the SQS queue to process & update to DynamoDB.

I'd go with D

Due to lack of information about what app is being interrupted, SQS seems to be always a good solution to decouple applications and to not lose transactions/messages

organization has seen application outages, resulting in missed transactions. "Application outage causing the missed transaction....." point to be noted