AWS Global Infrastructure: Regions, Availability Zones, and Edge Locations

An AWS Region is a specific geographic area — for example, one Region located in Africa, another in Europe, another in North America. Each Region operates independently of the others, with its own separate infrastructure.

Choosing a Region isn't a minor detail. It affects how far network traffic has to travel to reach your users (latency), and it can affect legal or regulatory obligations about where certain data is allowed to be stored — organisations should evaluate their own applicable regulatory and data-residency requirements before treating a Region choice as final.

Within a single Region, AWS operates multiple Availability Zones (AZs) — physically separate data centre clusters with their own independent power, cooling, and networking, but connected to each other by fast, low-latency links.

This is the distinction worth remembering: a Region contains multiple Availability Zones — a Region is not an Availability Zone, and an Availability Zone is not a Region. Spreading a system's resources across more than one AZ inside the same Region means that if one AZ has a problem — a power failure, for instance — the others keep running and your application can keep working.

Edge locations are smaller AWS sites, far more numerous than full Regions, used mainly to cache and deliver content closer to end users — reducing the distance data has to travel for things like images, videos, and static website assets. You'll see this in practice with Amazon CloudFront in Module 4. For now, the key idea is: an edge location isn't a place your application "runs" the way a Region is — it's a place your content gets cached closer to whoever's requesting it.

Three examples make this concrete:

1. Latency — a South African company with almost all its customers in Southern Africa should pick a Region physically close to them, not one on another continent, to keep response times low.
2. Resiliency — a company that wants its application to survive a single data centre failure should spread its resources across multiple Availability Zones within one Region — picking a "better" Region alone doesn't achieve this on its own.
3. Compliance and data residency — a company handling data with legal residency requirements needs to evaluate which Region(s) satisfy those requirements; this is a Region-level decision, not an Availability Zone-level one.

Two more factors worth knowing about: not every AWS service is available in every Region, and prices can differ by Region — both are reasons to treat "which Region?" as a real decision rather than a default.