Every IT architect strives to deliver an optimized and cost effective solution to their customer. Therefore, they must be able to explain and understand the different options that a customer has to then assist them in choosing the best option possible taking the trade-offs into account. There is a mutual partnership relationship between the architect and the customer that is ongoing in order to produce a quality output deliverable.
Amazon AWS infrastructure is broken up by regions and availability zones. They continue to expand their infrastructure constantly as their business grows. So what is a region, a region is a named set of AWS resources in the same separate geographic area. AWS provides you with a choice of different regions around the world in order to help customers meet their requirements. Each region is completely isolated from the other regions.
There are only a set number of available regions to choose from when but as the customers grow, then AWS will continue to provide the infrastructure that meets the global requirements. In North America AWS has 3 regions to choose from and a GovCloud region:
Inside of these regions, there are Availability Zones. These are basically AWS data centers within these regions that are connected to each AZ in the region via low latency links.
Availability zones allows you to architect your applications to be as resilient as possible by separating them out as failure domains so that there is not a single point of failure. As with all architects, we must design architectures that assumes that things will fail.
With the implementation of Auto-Scaling, ELBs and multiple Availability Zones then you can build a reliable architecture that takes only minutes to setup instead of days and weeks. I’ll go over Auto-Scaling and ELBs on a separate post.
- Regions and Availability Zones: http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-regions-availability-zones.html
- AWS Global Infrastructure: http://aws.amazon.com/about-aws/global-infrastructure/