Quiz: Module 16: Reliability and Disaster Recovery

A company runs an e-commerce application that generates $10,000 in revenue per hour. The business requires that the application be restored within 15 minutes of a failure and can tolerate losing at most 5 minutes of transaction data. Which disaster recovery strategy meets these requirements?

A solutions architect is reviewing a web application architecture. The application runs on a single EC2 instance in one Availability Zone, connects to a Single-AZ RDS database, and uses a single NAT gateway. The architect needs to identify single points of failure. Which THREE components are single points of failure in this architecture? (Select THREE.)Select multiple

Which resilience pattern prevents a failing downstream service from consuming all resources in the calling service, potentially causing the caller to fail as well?

A company stores critical financial records in an S3 bucket and an RDS PostgreSQL database. The compliance team requires that backups be taken daily, retained for 90 days, and stored in a different Region from the primary data. Which AWS service provides centralized management of these backup requirements across both S3 and RDS?

A development team implements retry logic for calls to a downstream API. During a downstream outage, all instances of the calling service retry simultaneously every second, overwhelming the downstream service as soon as it recovers. Which improvement should the team make to their retry logic?

A company runs a global SaaS platform that serves customers in North America, Europe, and Asia. The platform must remain available even if an entire AWS Region becomes unavailable. Users expect sub-100ms latency regardless of their location. Which architecture and DR strategy should the company implement?

A team wants to validate that their Auto Scaling group correctly replaces failed EC2 instances and that the ALB routes traffic to healthy instances during an instance failure. They want to test this in a controlled manner without waiting for a real failure. Which AWS service should the team use to simulate the instance failure?