Resilient Cloud Solutions

What this domain actually tests

Multi-region and multi-AZ architecture, and it is the domain where the exam gives you numbers and expects you to pick the pattern.

The four DR patterns, in ascending order of cost and descending order of recovery time:

1. Backup and restore — cheapest, slowest. Hours to days.
2. Pilot light — core data replicated, minimal infrastructure running, scale up on failover.
3. Warm standby — a scaled-down but fully functional copy running continuously.
4. Multi-site active/active — full capacity in both regions, near-zero recovery time, most expensive.

The question shape is consistent: a scenario states an RTO and RPO explicitly ("we can tolerate 30 minutes of downtime and 5 minutes of data loss") and asks for the solution.

The traps

The exam wants the cheapest option that meets the requirement, not the best one. This is the trap that catches engineers. Multi-site active/active meets almost every RTO you will be given — and it is wrong whenever something cheaper also meets it. Read the RTO, find the cheapest pattern that clears it, and stop.

RTO and RPO are different things and both matter. RTO is how long you can be down. RPO is how much data you can lose. A solution can meet one and fail the other. Backup and restore might satisfy a generous RTO while blowing a tight RPO, because the last backup was six hours ago.

Know the replication options and what they buy:

• S3 Cross-Region Replication — object-level, asynchronous.
• Aurora Global Database — sub-second cross-region replication, fast promotion of a secondary. This is the answer when the requirement is a very low RPO on a relational database.
• DynamoDB global tables — multi-region, multi-active (you can write in both regions).

That "multi-active" property is unique and questions exploit it: if the requirement is to write in two regions simultaneously, global tables are the answer and Aurora Global Database is not.

Route 53 health checks and failover routing are how the switch actually happens. A DR design with no automated failover mechanism is incomplete, and the exam notices.

How to study it

Make a table with the four patterns down the side and rough RTO/RPO ranges across the top. Then practise mapping requirements to patterns until it takes seconds — this is a lookup, not a reasoning exercise, and treating it as a lookup is what buys you time on exam day.

Then learn the three replication technologies by their distinguishing property rather than their description: Aurora Global for the lowest RPO, DynamoDB global tables for multi-active writes, S3 CRR for objects.

Finally, remember tie-breaker four from the exam as a whole: automated failover beats a documented manual procedure, every time. If one option has a human promoting a replica and another has Route 53 doing it on a health check, the human loses.