liveBook · Manning

Availability zone

As explained in Amazon Web Services in Action, Third Edition: An in-depth guide to AWS

An Availability Zone is an isolated location within an AWS Region containing one or more independent data centers.

An availability zone (AZ) is an isolated location within an AWS Region containing at least one data center. A Region contains multiple AZs, connected by low-latency links. An AZ name such as us-east-1a identifies a zone within a Region, but the letter can map to a different physical zone in another AWS account. section 13.2.1 section 13.2.5

Figure 13.3 A region consists of multiple availability zones connected through low-latency links.

The diagram places multiple isolated AZs inside one Region and shows the links between them. section 13.2.1

A single EC2 instance can fail, and recovery in its original AZ cannot solve an outage of that entire AZ. Distributing a workload across AZs lets it recover from that larger failure. The design must account for resources tied to one zone: a subnet belongs to one AZ, an EC2 instance launches in one subnet, and an EBS volume is unavailable during its AZ’s outage. section 13.2 section 13.2.4 section 16.0

Find zones and plan a web application's recovery

Run `aws ec2 describe-availability-zones --region $Region` to see the AZs available in a chosen Region. For a web application on EC2, use subnets in different AZs with an Auto Scaling group rather than relying on one instance. A replacement launched in another AZ uses another subnet, so it cannot retain the failed instance’s private IP address. If clients need a stable endpoint, options include associating an Elastic IP during instance bootstrap or using an Elastic Load Balancer. section 13.2.5 section 13.2.4 section 16.0

Check how each stateful component handles an AZ outage:

  • An EBS volume remains tied to its AZ. EFS instead shares a filesystem across EC2 instances in different AZs and replicates its data across them. section 9.0 section 13.2
  • RDS Multi-AZ places primary and standby databases in different AZs. Set `MultiAZ: true` on an `AWS::RDS::DBInstance` in CloudFormation to enable that deployment; expect additional charges and a short failover downtime. section 10.5.1 section 13.3.2

Cross-AZ traffic incurs charges, and latency within one AZ is lower than latency between AZs. Choose a deployment by checking each service’s resilience characteristics and availability commitments—not merely by counting AZs. section 13.2.1 section 13.3.2

How Amazon Web Services in Action, Third Edition: An in-depth guide to AWS covers it

How deep it goes
Defines regional placement and naming, then works through cross-zone storage, networking, and failover constraints without exposing physical data-center mappings.
Examples
Uses an EC2 recovery scenario, a zone-discovery CLI command, and a CloudFormation `MultiAZ: true` setting.
What it assumes
Introduces Regions and AZs from their physical layout while assuming basic familiarity with EC2 instances and web applications.
What you can do after
Build it from scratch by placing redundant EC2 capacity in different AZs and choosing storage and endpoints that support failover.
How it connects
Connects availability zones to VPC subnets, Auto Scaling groups, EBS, EFS, and RDS Multi-AZ deployments.

Also explained in

Compare how 4 books explain availability zone →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Amazon Web Services in Action, Third Edition: An in-depth guide to AWS. The section links point to the parts of the book it draws on.