Skip to content
5 min read·Lesson 2 of 10

Azure Global Infrastructure

Learn how Azure organises its infrastructure into geographies, regions, region pairs, and availability zones — and why these matter for resilience and compliance.

Azure's global infrastructure is one of the largest in the world, with data centres in over 60 regions across 140 countries. Understanding how it's organised is essential for designing resilient architectures and answering AZ-900 exam questions on availability and compliance.

Geographies

An Azure geography is a market (typically a country or group of countries) that contains one or more Azure regions. Geographies ensure that data residency and compliance requirements are respected — data stored in a geography stays within that geography by default.

Examples: United States, United Kingdom, European Union, Australia, Japan.

Regions

A region is a set of data centres in a defined geographic area, connected by a low-latency network. Azure has 60+ regions. Examples: East US, West Europe, Southeast Asia.

Why regions matter:

  • Latency: Deploy close to your users.
  • Data residency: Keep data within a country or regulatory boundary.
  • Service availability: Not all Azure services are available in every region.
  • Cost: Prices vary by region.

Region Pairs

Most Azure regions are paired with another region in the same geography. Region pairs:

  • Are at least 300 miles apart to reduce the chance of both being affected by the same disaster
  • Enable geo-redundant storage replication and geo-redundant database backups
  • Receive Azure platform updates in a staggered rollout (reducing simultaneous outage risk)
  • Example pairs: East USWest US; North EuropeWest Europe; UK SouthUK West

Availability Zones

Availability Zones (AZs) are physically separate data centre facilities within a single region. Each zone has independent power, cooling, and networking. Zones are connected by high-bandwidth, ultra-low-latency fibre.

Using multiple AZs protects against:

  • Power failures at a single facility
  • Network outages within a zone
  • Physical disasters affecting one zone

Not all Azure regions have Availability Zones. Check the Azure documentation for zone-enabled regions.

Infrastructure Comparison

ConceptAzureAWSGCP
Geographic boundaryGeography
Geographic areaRegionRegionRegion
Isolated data centreAvailability ZoneAvailability ZoneZone
Cross-region pairingRegion PairDual-region

Azure Edge Locations

In addition to regions and zones, Azure has:

  • Azure CDN PoPs: 100+ edge nodes globally for content caching and acceleration
  • Azure Front Door: Global load balancer and WAF with anycast edge presence
  • Azure Edge Zones: Extensions of Azure regions into metropolitan areas for ultra-low-latency workloads (e.g., 5G applications)
AZ-900 exam tip: Know the hierarchy — Geography → Region → Availability Zone — and the purpose of region pairs. The exam often asks about the difference between redundancy options and which scenario benefits from AZs vs region pairs.

Key Takeaways

  • Azure has 60+ regions globally, organised into geographies that respect data residency boundaries.
  • Availability Zones are physically separate data centres within a region, providing zone-redundant protection.
  • Region pairs provide cross-region replication for disaster recovery within the same geography.
  • Some Azure services are regional, some are zone-redundant, and some are global.
  • Choosing the right region affects latency, compliance, service availability, and cost.

Test your knowledge

Try exam-style practice questions to reinforce what you've learned.

Practice Questions →