Latest Posts

Azure Global Infrastructure

Azure Global Infrastructure

You need to understand a few terms in order to know Azure Global Infrastructure in a better way.

Globally, Azure infrastructure has 2 main components.

  1. Physical Component: Hundreds of datacenters, grouped into different regions, across the globe (and Microsoft keeps adding more)
  2. Connection Component: The datacenters are connected through low-latency, fiber connection providing high-speed data transfer.

What is Azure Datacenter?

In simple terms…

  • Many interconnected computers put together, make Server.
  • Many interconnected servers put together in a physical building/premise, make a Datacenter

What is Region?

Definition: An Azure region is a set of data centers, deployed within a latency-defined perimeter and connected through a dedicated regional low-latency network.

Three terms in the above definition:

  • Set of Datacenters: Each region has more than one datacenter
  • latency-defined perimeter: Latency is the time it takes to travel data from one data center to another data center in this context.
  • Regional low-latency network: In the same region, data centers are connected with the fiber connection for high-speed data transfer.

Simple definition: More than one datacenter connected with each other using a fiber connection is called Region.

How to Choose a Region?

  • Location: Try to select the region which is closest to your geography of application and data transaction. E.g., if your application users are from India, better for you to select the India region while creating Azure resources.
  • Price: The price of the Azure products and services vary from region to region.
  • Features: Not all the features are available in all the regions. If you need some specific feature, you may have to compromise with region selection.

What is Azure Geography?

Definition: Azure Geography is the part of the world, where there is 1 or more than 1 region.

e.g., USA, UK, India

Why Azure Geography is important?

  • Reason#1 - Latency: If you are working in the USA but your application user base is in India, so it is advisable that you choose India region datacenters. Your data is in India, your application is also in India and your user base is also India. This way you can minimize the data travel time (latency) between user and data storage.
  • Reason#2 – Compliance: Every country (Geography) has or might have some regulations about data. E.g., financial, health, or government information-related data must not leave the country and reside in that country only. So, for that matter, it becomes obvious to choose the region, only after checking the compliance information before starting the development.
  • Reason#3 - Some features not available: There are certain products/services which are available only in specific regions of the world. Now, in this case, you will have to prioritize between Reason# 1 and Reason# 3.

What is Paired Region?

Definition: Two regions can be paired within the same azure geography only.

  • Azure serializes platform updates or planned maintenance across paired regions.
  • At a time only one region of a pair is updated with changes.
  • If there is an outage, then at least 1 region will be prioritized in each pair for recovery.
  • Some Azure services take advantage of the paired region and automatically replicate the data in the paired region in cases of outage of 1 region. E.g., Geo-redundant storage
  • Not all azure services provide this data replication functionality, in those cases, customer needs to manage the recovery and replication.

What are the Azure Availability Zones?

Definition: In each Azure region, there is a unique physical location having 1 or more than 1 data center, equipped with independent power, cooling facilities, and networking.

The entire idea of Availability Zones ensures that your data stays protected and synchronized even in times of disaster.

  • The availability zones are tolerant to local failures like earth quack, flood, fires, large power failures.
  • Azure Availability Zones can survive local failures because of redundancy and logical isolation of Azure services.
  • Minimum 3 availability zones are kept, in whichever region Availability Zones are enabled. So even if 1 out of 3 availability zones is facing problems, there 2 more available for your service.
  • Availability zones are connected with each other with a high-performance fiber network, giving less than 2ms of round-trip latency.

What is the Azure Edge Zones?

  • Azure edge zones are just extensions of the azure. It lets you run the throughput intensive and low latency apps run closer to your user-base area.
  • For now, Azure Edge Zones support Virtual Machines (VMs), Containers, and a selected set of services.
  • Azure Edge zones are part of Microsoft Global Architecture and provide secure, reliable, and high-bandwidth connectivity for the apps running at the Azure Edge Zones (Near to user base).

What is the Microsoft global wide-area network (WAN)

  • The Microsoft global wide-area network (WAN) provides high speed and high availability connectivity between all the data centers in all the regions across the world.
  • This connectivity ultimately provides a seamless experience for azure services users.

What is an Azure point of presence?

  • Azure Point of Presence - PoP, is nothing but the physical location or access point of the data centers which acts like data traffic entry and exit point.

We value your Feedback:

Page URL:







Azure Global Infrastructure
© 2024 Code SharePoint