Throughout your journey with Microsoft Azure, you will hear and use terms such as regions, availability zones, resources, subscriptions, and many more.

This module focuses on the fundamental architectural components of Azure.
These components can be divided into two main groups:

• Physical infrastructure
• Management infrastructure

Physical Infrastructure

Azure’s physical infrastructure starts with data centers.
Conceptually, these data centers are similar to those of large enterprises: they are facilities with resources organized in racks, equipped with dedicated power supply, cooling systems, and network infrastructure.

As a global cloud provider, Azure has data centers worldwide.
However, these individual data centers are not directly accessible.
They are grouped into Azure regions or Azure availability zones, designed to help you achieve optimal resilience and reliability for your critical workloads.

The Global Infrastructure site allows you to interactively explore Azure’s underlying infrastructure.

Regions

A region is a geographic area on the planet that contains at least one, but potentially several, data centers located close to each other and interconnected by a low-latency network.
Azure intelligently assigns and controls resources in each region to ensure proper workload balancing.

When you deploy a resource in Azure, you often need to choose the region where you want that resource to be deployed.

Availability Zones

Availability zones are physically separate data centers within the same Azure region.
Each availability zone consists of one or more data centers with independent power, cooling, and network connectivity.

An availability zone is designed as an isolation boundary: if one zone fails, the others continue to operate.
Availability zones are connected to each other through high-speed, private fiber-optic networks.

Using Availability Zones in Your Applications

You want your services and data to be redundant to protect your information in case of a failure.
When hosting your own infrastructure, implementing redundancy requires duplicating the hardware environment.
Azure can help make your application highly available through availability zones.

You can use availability zones to run critical applications and build high availability into your application architecture by co-locating your compute, storage, network, and data resources in one availability zone, then replicating them across other zones.

Keep in mind that there may be a cost associated with duplicating your services and transferring data between availability zones.

Availability zones are primarily used for:

• Virtual Machines (VMs)
• Managed disks
• Load balancers
• SQL databases

Azure services that support availability zones fall into three categories:

• Zonal services: You assign the resource to a specific zone (e.g., VMs, managed disks, IP addresses).
• Zone-redundant services: The platform automatically replicates resources across zones (e.g., zone-redundant storage, SQL Database).
• Non-regional services: These services are always available across Azure geographies and are resilient to zone or region failures.

Region Pairs

Even with the additional resilience provided by availability zones, a major event could impact multiple zones within the same region.
For even greater resilience, Azure uses the concept of region pairs.

Most Azure regions are paired with another region in the same geographic area (such as the United States, Europe, or Asia), at least 300 miles (about 480 km) apart.
This approach enables resource replication across a geography, reducing the risk of outages caused by events such as:

• Natural disasters
• Civil unrest
• Power outages
• Network failures affecting an entire region

For example, if one region in a pair is affected by a natural disaster, services automatically fail over to the other region in the pair.

Examples of Azure region pairs:

• West US paired with East US
• Southeast Asia paired with East Asia

These region pairs are directly connected and far enough apart to be isolated from regional disasters, providing reliable services and data redundancy.

Additional Benefits of Region Pairs:

• In the event of a widespread Azure outage, one region in each pair is prioritized to ensure that at least one is restored as quickly as possible for applications hosted in that pair.
• Planned Azure updates are deployed to one region at a time within the pair to minimize downtime and reduce the risk of application failure.
• Data remains within the same geographic area as its paired region (except for Brazil South) for tax and legal jurisdiction reasons.

In a one-way pairing, the primary region does not provide failover for its secondary region.
For example:

• The secondary region for West India is South India, but South India does not depend on West India.
• The secondary region for South India is Central India.
  Brazil South is unique because it is paired with a region outside its geography: its secondary region is South Central US.
  However, South Central US does not have Brazil South as its secondary region.

Sovereign Regions

In addition to standard regions, Azure also has sovereign regions.
Sovereign regions are instances of Azure that are isolated from the main instance.
You might need to use a sovereign region for compliance or legal reasons.

Azure sovereign regions include:

• US DoD Central, US Gov Virginia, US Gov Iowa, etc.:
  These regions are physically and logically isolated instances of Azure, intended for U.S. government agencies and their partners.
  Data centers are operated by screened U.S. personnel and have additional compliance certifications.
• China East, China North, etc.:
  These regions are available through a unique partnership between Microsoft and 21Vianet, where Microsoft does not directly operate the data centers.

Next Unit: Describe Azure’s Management Infrastructure.