Skip to Content

An overview of AWS global infrastructure

Posted on October 27, 2022 by

Categories: AWS


The most significant cloud infrastructure in the world, AWS (Amazon Web Services), offers over 200 services to more than a million clients across 240 countries and territories. Across several geographies, AWS is continuously growing its infrastructure. 81 availability zones and 25 geographical areas are currently covered by AWS globally. It will soon span 24 availability zones across several nations and 8 more geographic areas. The worldwide infrastructure of AWS will be explained using the following terms in this blog.

  • Regions
  • Zones of Availability
  • Regional Zones
  • Edge Placements

Regions and Direct Connect Locations

The actual place where various clusters of data centers are installed to offer various cloud services is known as an AWS Region. The availability zone is a logical collection of data centers, and a region must have at least 3 availability zones. Within the area, these availability zones are physically separated from one another. Each availability zone within the area is equipped with redundant, ultra-low latency networking and its own power, cooling, and physical security system. Create your application to operate in various availability zones if you want to run highly available apps on AWS. Your application will continue to operate in another availability zone if one goes down.

AWS has the most significant global infrastructure compared to other well-known cloud providers and offers cloud services to clients worldwide. Currently, AWS is maintaining about 21 regions in the following geographies.

Zones of Availability

A region’s ultra-low latency and redundant networking data centers are gathered in an availability zone. Within each AWS region, there are a minimum of three availability zones. An application can be set up over several availability zones in a given area for high availability. The application will still be available in the second availability zone if one availability zone goes down.

A region’s availability zones are all connected via a high bandwidth, external latency network, which speeds up communication between them. Each availability zone has a backup network, connection, and power supply, and all communication between them is secured. Synchronous replication between availability zones is possible because of the rapid network speed.

Your application will be protected against natural calamities like floods, earthquakes, etc., if it is distributed across different availability zones. All availability zones are located within a 100 km radius, and the average distance between two availability zones is substantial.

Regional Zones

Different AWS services like computation, storage, and databases are positioned closer to the end users using AWS local zones. Real-time gaming, media, and entertainment applications that need a latency of under 10 milliseconds may be run closer to the end consumers using AWS local zones.

Utilizing AWS services like Amazon EC2, Amazon VPC, Amazon EBS, Amazon EFS, and Amazon ELB, low-latency applications may be operated in AWS local zones. Local zones provide high bandwidth and secure connectivity between AWS and local zone workloads, enabling end users to connect to the services without interruption. Below is a list of the principal applications for local zones.

  • running edge apps with low latency
  • keeping in compliance with the strict data residency criteria
  • Migrations to hybrid clouds made simple

Edge AWS locations

The AWS edge locations are the close-by data centers created to serve content from various AWS services with the least amount of delay. Amazon has several edge locations spread out around the world. AWS initially determines if the requested data is present at the edge location when a user submits a query. If the data is present at the edge location, the end user receives it without first requesting it from the AWS servers. Otherwise, data is sent and stored at the edge location after being retrieved from the AWS servers.

CloudFront: CloudFront stores a copy of the data closer to the user by caching it at edge sites. The end consumers can then receive this stuff more quickly.

Web application firewall: At the edge locations, a web application firewall filters undesired data before entering the AWS region.
Route53: A DNS server, Route53 handles DNS requests from edge sites. The DNS query is made, cached at the edge location, and returned from the edge location the following time.

Direct link sites for AWS

AWS offers direct connect locations to link your local data center to the AWS so that your local services and AWS services may communicate discreetly. Your private network can be connected through fiber optic cables to AWS direct connect sites using AWS direct connect locations. A virtual interface to AWS public services like Amazon VPC and Amazon S3 is formed using AWS direct connect, directly avoiding ISPs in the network route. Access to public services is provided via AWS direct connect in the area where it is implemented.

The AWS global infrastructure has more than 100 direct connect points for a better hybrid environment. The list of geographical areas where the AWS direct link is available is shown below.