If you are starting with AWS, the first thing you should learn is cloud is divided into Availability-zones, regions, and endpoints.
Regions, AZs, and endpoints are important and fundamental concepts that engineers should know.
What are regions?
The cloud is divided into Regions. Regions are separate geographic areas. When you are building a solution for a particular geographic you will select that region to build the solution.
For example, if you are building a solution for the USA, you will select the USA and use services available in the USA region.
Remember, a few AWS services might not be available in some regions. So, when you design a solution first thing is to select the region and make sure the services you are going to use are available in that region.
All the regions in AWS are highly secure. You can replicate the date between the regions.
How to select a region?
1 . Pricing
2 . Latency
3 . Service availability
4 . User/customer location
5 . Data compliance and security
Select the region from AWS console you can select from the top-right corner.
What are Availability zones
Availability zones are physical locations within regions. A region has 2 or more availability zone. AZs in the same region are connected by low-latency networks.
Availability zones help in making applications and services highly available, fault-tolerant, and scalable.
For a few AWS’s services you can select AZs and in a few AWS will select for you.
Remember, AZs job is to make the application highly available. HA (highly available) means even if one server goes down the other server will move to the healthy one.
Example of AZs in EC2:
What are Edge Locations
Edge locations are AWS data centres designed to deliver services with the lowest latency. These DCs are closer to the users than the regions. DCs are spread worldwide and serve the data to users with low latency.
In AWS, there are a few services which use the Edge Locations:
1 . Route 53
2 . CloudFront
3 . WAF (Web Application Firewall) and AWS Shield
4 . AWS Global Accelerator
AWS has more edge locations.
You can’t run your workloads directly to Edge locations. Only AWS services can utilize the Edge locations.
CloudFront has two tiers:
a . Point of Presences (PoPs)
PoP are numerous but have smaller cache
b . Regional Edge Locations
Regional Edge locations are fewer in number but have a large cache
Local Zones let us run the workloads directly in Local zones. It is not available across the world as of now. These are different from Edge locations.
These are important for a few reasons:
1 . In AWS, services are based on Regions and Global. Eg: S3, IAM are global. Buckets in S3 are region-based. A bucket created in one region cannot be accessed in another region by default.
2 . The goal of every solution we are designing is to serve the content with low latency. So, understanding regions, AZs, Edge-points, etc. will guide you to understanding how latency, regions, and end points are related.
3 . High-Availability is one of the pillars of the well-architected design. For the high-availability design, we should design a solution to be available in multiple AZs.
Follow me on Twitter