The AWS global infrastructure

AWS is built from Regions, each containing multiple isolated Availability Zones, plus a large network of edge locations. Design for high availability with multiple AZs and use multiple Regions for disaster recovery, latency, or data residency.

Building blocks

• Region: A separate geographic area (for example, us-east-1). Each Region is isolated from the others and contains multiple Availability Zones. You choose where your data lives.
• Availability Zone (AZ): One or more discrete data centers with redundant power, networking, and connectivity. AZs in a Region are physically separated and connected by low-latency links.
• Edge location: A site used by Amazon CloudFront, Route 53, and AWS Global Accelerator to serve users with lower latency. There are far more edge locations than Regions.

High availability

• AZs do not share single points of failure, so running across at least two AZs keeps an application available if one AZ has a problem.
• Typical pattern: a load balancer spreading traffic to instances in multiple AZs, with Auto Scaling replacing unhealthy instances and a Multi-AZ database.

When to use multiple Regions

• Disaster recovery and business continuity: Survive the loss of an entire Region.
• Low latency for end users: Run close to users in different parts of the world.
• Data sovereignty: Keep data inside a country or jurisdiction to meet legal requirements. Data doesn't leave a Region unless you move it.

Choosing a Region

• Consider compliance and data residency, proximity to users, which services and features are available in that Region, and price (prices differ by Region).