Quiz: Module 19: Advanced Topics

A company serves a global user base from a single AWS Region (us-east-1). Users in Asia and Europe report slow page load times for the company's static website hosted on S3. Which service should the company add to reduce latency for global users, and how does it work?

A development team builds a Lambda function that reads a product catalog from DynamoDB. The catalog changes once per day, but the function is invoked 10,000 times per hour. Each invocation reads the same catalog data. The team wants to reduce DynamoDB read costs and improve response latency. Which caching strategy should the team implement?

Which AWS service automates the setup of a multi-account landing zone with organizational units, centralized logging, guardrails, and a compliance dashboard?

A solutions architect is designing a workflow for processing insurance claims. The workflow involves: (1) validating the claim, (2) checking for fraud (which takes 30 seconds), (3) if fraud is detected, routing to a human reviewer who may take hours to respond, (4) if no fraud, automatically approving the claim and notifying the customer. Which AWS service is best suited for orchestrating this workflow, and why?

A data engineering team stores 500 GB of application logs in S3 as CSV files. They use Amazon Athena to run daily queries that filter by date and aggregate error counts. Each query scans the entire 500 GB dataset and costs approximately $2.50 per query. Which TWO optimizations would reduce the query cost most significantly? (Select TWO.)Select multiple

A company currently runs all workloads in a single AWS account. The security team is concerned that a compromised IAM credential in the development environment could affect production resources. The finance team cannot attribute costs to specific projects. Which multi-account change addresses both concerns?

A solutions architect is evaluating whether to use AWS Step Functions (Standard Workflow) or an SQS queue with Lambda for processing 100,000 short-lived events per hour. Each event requires a single Lambda function invocation with no branching, no parallel execution, and no human approval. Which approach is more appropriate, and why?