AWS Global Infrastructure: Regions, AZs & Edge Locations Explained

Every AWS service runs somewhere physical. That "somewhere" is a carefully engineered global network of data centres, designed for redundancy, low latency, and compliance. Understanding this infrastructure is foundational — it determines how you design for high availability, where you store data, and how you serve global users efficiently.

AWS Regions

An AWS Region is a geographical area containing multiple, physically separated data centres. Each region is completely independent — it has its own power, cooling, networking, and is isolated from failures in other regions.

As of 2026, AWS has 33+ regions across North America, South America, Europe, Asia Pacific, the Middle East, and Africa, with more announced regularly.

Region Naming Convention

Regions have both a friendly name and an identifier:

Choosing a Region

Four factors drive region selection:

1. Compliance and data residency — Regulatory requirements may mandate that data stays within a geographic boundary (GDPR in Europe, data localisation laws in specific countries). This is the most important factor — it can be non-negotiable.
2. Latency — Choose the region closest to your users for the lowest latency.
3. Service availability — Not all AWS services launch in all regions simultaneously. us-east-1 (N. Virginia) consistently gets new services first.
4. Pricing — AWS pricing varies by region. us-east-1 is generally the cheapest. Some regions (São Paulo, Bahrain) can be 20–40% more expensive.

Availability Zones (AZs)

Within each region, AWS has Availability Zones — one or more discrete data centres with independent power, cooling, and physical security.

Each region has a minimum of 3 Availability Zones (most have 3–6). AZs within a region are:

• Physically separate — typically tens of miles apart
• Connected via high-bandwidth, low-latency private fibre
• Isolated from failures in other AZs — a fire, flood, or power failure in one AZ does not affect others

AZ Naming

AZs are named with a letter suffix: us-east-1a, us-east-1b, us-east-1c.

Note: AZ names are account-specific. Your us-east-1a and a colleague's us-east-1a may map to different physical data centres. AWS randomises the mapping to distribute load across AZs.

Why AZs Matter for Architecture

The core high-availability pattern in AWS is multi-AZ deployment:

Region: us-east-1
  ├── AZ: us-east-1a  →  Web server 1, DB primary
  ├── AZ: us-east-1b  →  Web server 2, DB standby
  └── AZ: us-east-1c  →  Web server 3

If AZ us-east-1a fails entirely, traffic is automatically routed to servers in us-east-1b and us-east-1c. This is the basis of 99.99% availability SLAs for many AWS services.

Edge Locations and CloudFront

Edge Locations are smaller AWS infrastructure points deployed in hundreds of cities globally — far more than the number of regions. They are used by Amazon CloudFront (AWS's Content Delivery Network) and Route 53 (DNS).

How CloudFront Uses Edge Locations

When a user in Sydney requests a static file hosted in us-east-1:

1. Without CloudFront: Request travels from Sydney → N. Virginia → back to Sydney. ~300ms latency.
2. With CloudFront: File is cached at the Sydney Edge Location. Request serves locally. ~5ms latency.

Edge Locations cache content (images, CSS, JS, API responses) close to users. They don't run full AWS services — you can't launch an EC2 instance at an Edge Location.

Local Zones and Wavelength Zones

Local Zones

Local Zones are extensions of AWS Regions placed in major metropolitan areas (Los Angeles, Chicago, Dallas, etc.). They bring compute, storage, and database services closer to specific population centres for latency-sensitive applications.

Use Local Zones when your users are concentrated in a city that doesn't have a nearby AWS region.

Wavelength Zones

Wavelength Zones embed AWS compute and storage within 5G carrier networks (Verizon, Vodafone, etc.). They enable ultra-low-latency applications (<10ms) for mobile users — AR/VR, connected vehicles, real-time gaming.

AWS Outposts

AWS Outposts brings AWS infrastructure physically into your own data centre or co-location facility. You get the same AWS services, APIs, and tools on-premises — fully managed by AWS.

Use Outposts when regulations require data to never leave your premises, you need extremely low latency to on-premises systems, or you want to run a hybrid workload that spans on-premises and cloud seamlessly.

Global Infrastructure Summary

AWS Global Infrastructure
│
├── Regions (33+)
│   ├── Availability Zones (3–6 per region)
│   │   └── Data centres (1–3 per AZ)
│   └── Local Zones (major metro extensions)
│
├── Edge Locations (500+)
│   └── Used by CloudFront and Route 53
│
├── Wavelength Zones
│   └── Embedded in 5G carrier networks
│
└── Outposts
    └── AWS rack in your data centre

Exam Tips: Global Infrastructure

The SAA-C03 exam frequently tests your understanding of when to use which infrastructure layer: Multi-AZ for high availability within a region (survive AZ failure), Multi-Region for disaster recovery and global user base, CloudFront + Edge Locations for low-latency content delivery globally, Local Zones for latency-sensitive apps near specific cities, and Outposts for on-premises compliance and hybrid architecture.

Previous: Lesson 1 — What Is Cloud Computing? | Next: Lesson 3 — AWS IAM

Part of the AWS Cloud Fundamentals course.

External references:

• AWS Global Infrastructure — official regions and edge locations overview