2.5. Reflection Checkpoint

Phase 2 covers the highest-weight domain in the exam. Before continuing to Phase 3, verify you can answer these:

1. What EC2 metrics are not published by default, and how do you get them?
2. You need an alarm to fire only when CPU > 90% for 3 consecutive minutes on a 60-second metric. What are the Period, Evaluation Periods, and Datapoints to Alarm values?
3. What is the difference between CloudWatch Logs metric filters and CloudWatch Logs Insights?
4. A CloudTrail trail is configured. API calls appear in the Event History console within seconds, but in the S3 bucket after 15 minutes. Is this expected behavior?
5. What are X-Ray annotations vs. metadata — and which is searchable?
6. An EventBridge rule needs to fire when a Config rule reports an S3 bucket as non-compliant. What source and detail-type does the event pattern use?
7. When would you choose Systems Manager Automation over a Lambda function for remediation?
8. Why is gp3 generally preferred over gp2 for new workloads?
9. An HPC workload on Linux needs shared storage from S3 at >50 GB/s throughput. What storage service do you choose?
10. What problem does RDS Proxy solve for Lambda + RDS architectures?

💾 CHECKPOINT — Session 1 Complete Files saved: complete-guide.md (Phase 1 + Phase 2), keywords.json, relationships.json Resume Point: Begin Phase 3 (Reliability and Business Continuity) in the next session.