Designing Highly Available and/or Fault-Tolerant Architectures

Architectures that are highly available mean that in the event of an Availability Zone (AZ) going down, your applications can remain operational. Werner Vogels is famous for the phrase, “Everything fails, all the time,” and they are good words to live by, especially when operating mission-critical applications and infrastructure. To explore the concepts in this chapter, you are going to see how you might transform a standard three-tier architecture to be highly available.

First, let us recap the AWS global infrastructure. Remember that there are Regions, which are geographically distinct locations around the world within which you can deploy resources. Within Regions, you have AZs. AZs may be made up of several data centers, and each zone is separated so that they are on different flood plains and using different energy suppliers. This means that a localized issue should never take out all of the AZs in...