AWS Elastic Disaster Recovery
AWS Elastic Disaster Recovery (DRS) is an Amazon Web Services service that minimizes downtime and data loss by continuously replicating on-premises or cloud-based applications into AWS for fast recovery. It is the next generation of CloudEndure Disaster Recovery, which it is designed to succeed; AWS recommends all CloudEndure Disaster Recovery customers transition to DRS.
Key points
- Successor to CloudEndure Disaster Recovery (CEDR) — AWS provides a CEDR-to-DRS Upgrade Assessment Tool and Server Upgrade Tool, and describes DRS as the recommended service for disaster recovery to AWS going forward.
- Performs continuous, block-level replication of source servers using the AWS Replication Agent, installable on physical infrastructure, hypervisor-based virtual machines (VMware, Microsoft, and others), or infrastructure on other cloud providers.
- Replicated data lands in a low-cost staging area in the customer’s own VPC, using affordable storage and minimal compute to keep ongoing replication cheap; only configuration and log data live in the DRS console’s own encrypted database.
- Recovery Point Objective (RPO) is typically in the sub-second range; Recovery Time Objective (RTO) is typically measured in minutes and depends heavily on OS boot time.
- Automatically converts replicated servers to boot and run natively on AWS when recovery or drill instances are launched, from either the most up-to-date state or a previous point in time.
- Supports non-disruptive recovery drills and failback to the original source environment once an incident is resolved.
- Because it works at the OS layer, it can protect physical servers as well as virtual ones.
- Costs scale with actual use: idle recovery-site compute is removed, and a full disaster recovery site is paid for only when needed.
- The four standard AWS disaster recovery strategies, in order of increasing speed and cost: backup and restore (data is backed up and infrastructure is provisioned from scratch during recovery — highest RTO/RPO, lowest cost), pilot light (a minimal core of critical infrastructure runs at all times and is scaled up on failover), warm standby (a scaled-down but fully functional copy of the full environment runs at all times and is scaled up on failover), and multi-site active-active (full production capacity runs simultaneously across multiple sites/Regions — lowest RTO/RPO, highest cost). DRS is generally used to implement the pilot light or warm standby strategies, replicating continuously so a full-scale environment can be launched quickly on failover.
- Distinct from AWS Application Migration Service: both services use the same underlying AWS Replication Agent and continuous, block-level replication mechanism, but they serve different purposes — Elastic Disaster Recovery keeps replicated servers continuously ready for disaster recovery and failover (an ongoing state), while Application Migration Service performs a one-time lift-and-shift migration cutover to AWS (a single event). This overlap is a common exam distractor.