Resilience in Amazon EFS
The Amazon global infrastructure is built around Amazon Web Services Regions and Availability Zones (AZs). Amazon Web Services Regions provide multiple physically separated and isolated AZs, which are connected with low-latency, high-throughput, and highly redundant networking. With AZs, you can design and operate applications and databases that automatically fail over between zones without interruption. AZs are more highly available, fault tolerant, and scalable than traditional single or multiple data center infrastructures.
Amazon EFS file systems are resilient to one or more Availability Zone failures within an Amazon Web Services Region. Mount targets themselves are designed to be highly available. As you design for high availability and failover to other AZs, keep in mind that while the IP addresses and DNS for your mount targets in each AZ are static, they are redundant components backed by multiple resources. For more information, see How Amazon EFS works with Amazon EC2 .
For more information about Amazon Web Services Regions and Availability Zones, see Amazon Global Infrastructure