SAA-C03 Study Guide: Active-Active Architecture
In the AWS Certified Solutions Architect – Associate (SAA-C03) exam, Active-Active is the gold standard for High Availability (HA) and Disaster Recovery (DR). Unlike other strategies where some resources sit idle, an Active-Active configuration ensures that all deployed resources across multiple locations (Availability Zones or Regions) are actively serving traffic simultaneously.
Core Concepts of Active-Active
In an Active-Active setup, the workload is distributed across two or more nodes or regions. This provides the lowest possible Recovery Time Objective (RTO) and Recovery Point Objective (RPO), often approaching zero.
1. Global Traffic Management
To achieve Active-Active at a global scale, Amazon Route 53 is essential. It uses routing policies like Latency-based routing, Weighted routing, or Multi-value answer routing to send users to the “best” healthy endpoint. Health checks ensure that if one region fails, DNS immediately stops directing traffic there.
2. Data Synchronization
The biggest challenge in Active-Active is keeping data consistent. AWS offers several managed solutions:
- Amazon DynamoDB Global Tables: Provides multi-region, multi-active replication. You can write to any region, and changes are propagated globally.
- Amazon Aurora Global Database: While primarily used for low-latency reads globally, it supports “Write Forwarding” in some configurations, allowing secondary regions to send writes to the primary.
Comparison: DR Strategies
| Strategy | RTO / RPO | Cost | Complexity |
|---|---|---|---|
| Pilot Light | Hours | Low | Medium |
| Warm Standby | Minutes | Medium | Medium |
| Active-Active | Near Zero | High | High |
Decision Matrix / If–Then Guide
| If the requirement is… | Then choose… |
|---|---|
| Zero downtime for global users | Route 53 Latency Routing + Active-Active Multi-Region |
| Multi-region write capabilities | DynamoDB Global Tables |
| Minimize costs while maintaining HA | Multi-AZ Active-Active (Single Region) |
| Regional compliance/data residency | Active-Active with Geolocation Routing |
Exam Tips and Gotchas
- RTO/RPO focus: If an exam question mentions “Zero RTO” or “Immediate recovery,” Active-Active is almost always the answer.
- Cost vs. Availability: Active-Active is the most expensive DR strategy because you are paying for full capacity in two or more locations 24/7.
- Session Stickiness: In Active-Active web tiers, ensure you use a distributed session store (like ElastiCache) so a user isn’t logged out if their request hits a different region.
- DynamoDB Conflict Resolution: Remember that DynamoDB Global Tables use “Last Writer Wins” for conflict resolution.
Topics covered:
Summary of key subtopics covered in this guide:
- Definition of Active-Active vs. Active-Passive.
- Route 53 Routing Policies (Latency, Weighted, Multi-value).
- Data consistency tools (DynamoDB Global Tables, Aurora Global).
- RTO and RPO metrics for high-end Disaster Recovery.
- Cost-benefit trade-offs of redundant infrastructure.
Infographic: Active-Active Multi-Region Architecture
Route 53: Health checks and Latency/Failover routing.
DynamoDB: Global Tables for multi-master writes.
ACM: SSL certificates must be present in BOTH regions.
Users are routed to the lowest latency region, improving global performance. Auto Scaling groups in each region handle local spikes independently.
Warning: Costs are essentially doubled. Use for mission-critical apps where downtime cost > infrastructure cost.