Cloud recovery platform

AWS Elastic Disaster Recovery

Recover your servers into AWS.

The overview

Continuous replication with recovery instances launched in AWS, plus drills and failback workflows for supported source infrastructure.

Scope of this profile

AWS DRS provides server replication and recovery into EC2. Operating the application, redirecting users, and arranging support remain design and ownership decisions.

Workload support

Provider-documented

Source infrastructure includes physical servers, VMware vSphere, Hyper-V, other clouds, and EC2. The FAQ covers applications and databases on supported Windows/Linux systems, with examples including Oracle, MySQL, SQL Server and SAP. [1]

Windows sources require x86_64; the matrix includes Windows Server 2016 through 2025. Linux support is version/kernel-specific and excludes 32-bit systems. Arm64 support is limited to Graviton EC2 sources recovering within AWS; on-premises and other-cloud Linux sources require x86_64. [2] [3]

Recovery locations

Provider-documented

The DRS endpoint list includes US East (N. Virginia and Ohio), US West (Oregon and N. California), Canada Central, London, Frankfurt, Singapore, Sydney, and both US GovCloud regions, among others. Recovery instances run in the selected supported AWS region. [4] [5]

Testing scope

Provider-documented

Drills launch recovery copies without changing source servers or stopping replication. Recovery plans can order dependent groups. DRS assists with failback, while redirecting production traffic is performed outside DRS, such as through DNS or another traffic-management system. [5]

Cost details

Provider-documented

The published DRS service rate is US$0.028 per replicating source-server hour. At an illustrative 730 hours, that is US$20.44 per server per month for the DRS fee alone; it is not the full recovery bill. AWS lists no upfront or minimum service fee. [6]

Additional charges include staging EBS volumes and snapshots, replication EC2 instances, drill/recovery EC2 and disks, conversion instances, and applicable data transfer. The DRS fee includes test and recovery launches; the AWS resources those launches consume are billed separately. [6]

Sources & review

Official pages reviewed October 16, 2026. Source links support the details above. Prices and supported configurations vary by service.

  1. AWS DRS infrastructure and application FAQ ↗
  2. Supported Windows versions and architecture ↗
  3. Linux versions, kernels, and Graviton restrictions ↗
  4. AWS DRS region endpoints and quotas ↗
  5. Recovery, drills, traffic failover, and failback ↗
  6. AWS DRS pricing and additional charges ↗

Bring this into the conversation

How will we redirect traffic, validate applications, and manage failback? Traffic failover is handled outside DRS.

Ask for a proposal covering the same workloads, a measured application recovery exercise, and separate costs for normal operation, a drill, and several weeks in recovery.

Use the buyer’s checklist ↗