Regions, Availability Zones, Edge Locations, and High Availability
AWS runs its cloud across a globally distributed infrastructure made of three nested units. Understanding their relationships is critical for the exam because AWS asks you to choose the right configuration for latency, compliance, and resilience.
| Concern | Use This | Why |
|---|---|---|
| Protect against data center failure | Multiple AZs in one Region | AZs have independent power and network. Deploying across 2+ AZs gives high availability. |
| Protect against regional disaster | Multiple Regions | Regions are fully isolated. Use for disaster recovery, business continuity, or data sovereignty. |
| Reduce latency for global users | Edge Locations + CloudFront | Content is cached at the edge, closer to users, reducing round-trip time. |
| Meet data residency laws | A specific Region | Data stored in a Region does not leave that Region unless you configure it to. |
| Lowest latency for mobile/IoT users | AWS Wavelength Zones | Embed AWS compute at the edge of 5G networks. (Know the name; exam mentions it rarely.) |
The exam loves to test whether you understand high availability versus fault tolerance in the context of AZs.
Run redundant resources across multiple AZs in the same Region. If one AZ fails, traffic fails over to the other. This is the standard AWS pattern for resilience. Example: An Auto Scaling group spanning AZ 1a and AZ 1b behind an Application Load Balancer.
A stricter form of resilience where the system continues operating without interruption during a failure. Often requires active-active multi-AZ or multi-Region setups with automated failover. The exam usually accepts multi-AZ as the answer for both HA and basic fault tolerance questions.
The exam will present scenarios and ask why you would deploy to more than one Region. Memorize these four reasons:
Both use Edge Locations, but they solve different problems. The exam tests whether you can distinguish them.
| Service | What It Does | Best For |
|---|---|---|
| Amazon CloudFront | Content Delivery Network (CDN). Caches static content (images, videos, HTML) at Edge Locations and serves it from the nearest edge. | Web sites, streaming, software distribution — anything where you want users to download content faster. |
| AWS Global Accelerator | Routes user traffic over the AWS global network to the optimal endpoint in a Region. Uses static IP addresses. | Real-time applications (gaming, VoIP, financial trading) where consistent low latency matters more than caching. |
Select one answer per question. You will receive immediate feedback.