- Home
- Disaster Recovery Orchestration Software
- AWS Elastic Disaster Recovery
AWS Elastic Disaster Recovery
What it is
AWS Elastic Disaster Recovery helps restore applications hosted on physical infrastructure, VMware vSphere, Microsoft Hyper-V, or other cloud providers to AWS. It is designed to limit downtime and data loss by launching recovery instances within minutes from either the latest server state or an earlier point in time. Recovery can target EC2 instances in another Availability Zone or AWS Region. After recovery, replicated data can be sent back to the primary site for failback. Teams can run non-disruptive recovery and failback drills while replication continues. Supported workloads include applications and databases on supported Windows and Linux versions, including Oracle, MySQL, SQL Server, and SAP. Point-in-time recovery can launch an operational copy from before a security incident. Data is encrypted in transit with TLS 1.2 and can use EBS encryption at rest; replication can also use VPN, Direct Connect, or VPC peering. The listed plan is $0.03 per month, billed at $0.028 per source server per hour. Additional AWS resource charges apply, and AWS recommends Business or Enterprise Support for production workloads.
For and against
For
- Recovers from physical, VMware, Hyper-V, or cloud sources.
- Supports recovery to another Availability Zone or Region.
- Recovery and failback drills can run during replication.
- Point-in-time recovery supports pre-incident restoration.
Against
- No free plan is listed.
- Additional AWS resource charges apply.
- AWS recommends Business or Enterprise Support for production workloads.
- One staging account supports up to 300 replicating servers.
The verdict
AWS Elastic Disaster Recovery combines recovery launches, point-in-time options, and failback support for workloads moving to AWS. Account-scale limits, support needs, and additional AWS resource charges are important planning considerations.
Who it suits
AWS Elastic Disaster Recovery suits organizations that need to recover supported Windows or Linux workloads from on-premises infrastructure or other cloud providers onto AWS. It may also fit teams that need to exercise recovery and failback while replication continues.
Plans and prices
AWS Elastic Disaster Recovery
Pricing includes continuous data replication, test launches, recovery launches, and point-in-time recovery; additional AWS resource charges apply
Read 4 Oct 2026 · aws.amazon.com
On the record
- Agent limits
- The installation guide states a maximum of 64 TiB per volume and 63 volumes per source server in its current requirements page.docs.aws.amazon.com · read 21 Oct 2026
- Agent requirements
- The replication agent requires 64-bit x86_64 systems, or arm64 Linux only on AWS Graviton EC2, plus IAM credentials and outbound TCP 443 access.docs.aws.amazon.com · read 21 Oct 2026
- API and SDK
- The public DRS API exposes operations for replication templates, launch templates, recovery plans, source networks and recovery instances; Boto3 provides a drs client.docs.aws.amazon.com · read 21 Oct 2026
- Automated actions
- Post-launch actions can run AWS Systems Manager commands or automations, with predefined options such as installing a CloudWatch agent.aws.amazon.com · read 4 Oct 2026
- AWS destinations
- AWS DRS can recover EC2 instances in another Availability Zone or AWS Region.aws.amazon.com · read 4 Oct 2026
- Compliance scope
- AWS lists AWS Elastic Disaster Recovery as in scope for SOC 1, SOC 2 and SOC 3 assurance programs.aws.amazon.com · read 21 Oct 2026
- Documentation depth
- AWS publishes a user guide, API reference, installation procedures, operating-system matrices, networking requirements, FAQs and release notes.docs.aws.amazon.com · read 21 Oct 2026
- Encryption
- Replication traffic uses TLS 1.2 or later, while staging and recovery EBS volumes use AWS KMS encryption by default or a customer-managed key.docs.aws.amazon.com · read 21 Oct 2026
- Existing-instance recovery
- Recovery can target an original or predefined existing EC2 instance instead of creating a new one.aws.amazon.com · read 21 Oct 2026
- Failback
- After a disruption, recovered workloads can replicate back to the original environment and fail back when ready.aws.amazon.com · read 21 Oct 2026
- Fast recovery mode
- A July 2026 release added Fast recovery mode, which skips conversion and launches directly from replicated snapshots when the agent requirement is met.docs.aws.amazon.com · read 21 Oct 2026
- IPv6
- An April 2026 release added IPv6 support for replication and control-plane connections in Regions where DRS and EC2 support IPv6.aws.amazon.com · read 21 Oct 2026
- Launch management
- Launch management settings can be applied as defaults to new servers or changed in bulk for existing servers.aws.amazon.com · read 21 Oct 2026
- Limit: G2 reviewers
- G2 reviewers report pricing complexity, AWS ecosystem dependence, slow initial replication, setup difficulty for nontechnical users, support delays and RTO/RPO limitations.g2.com · read 21 Oct 2026
- Limit: PeerSpot reviewers
- PeerSpot reviewers request better monitoring and reporting, clearer cost visibility, simpler setup, improved interface and faster support; one reviewer reported failback failure.peerspot.com · read 21 Oct 2026
- Limit: source coverage
- A PeerSpot reviewer said the service did not support all of their backup types in a MongoDB-oriented use case.peerspot.com · read 21 Oct 2026
- Linux installation prerequisites
- Linux installation requires root or sudo, at least 4 GB free on /, 500 MB on /tmp, matching kernel headers, gcc and make, and an executable /tmp.docs.aws.amazon.com · read 21 Oct 2026
- Maker facts
- Amazon Web Services launched in 2006, is headquartered in Seattle, Washington, and lists more than 10,001 employees on LinkedIn.linkedin.com · read 21 Oct 2026
