ECS disaster recovery continuously replicates instance data to a disaster recovery site, delivering seconds-level Recovery Point Objective (RPO) and minutes-level Recovery Time Objective (RTO) for critical applications.
How it works
ECS disaster recovery monitors disk-level data changes on your operating systems and ECS instances, captures changes in real time, and synchronizes them to a disaster recovery site. If the production site fails, you fail over to the disaster recovery site and resume operations within minutes.
Scenarios
ECS disaster recovery supports two deployment scenarios:
| Scenario | Protects against | RPO | RTO |
|---|---|---|---|
| Cross-region disaster recovery | Regional disasters such as tsunamis and earthquakes | 1 minute | 15 minutes |
| Cross-zone disaster recovery | Single-zone faults caused by force majeure events (fire, power outage) or equipment failures (software, hardware) | Per your business goals | Per your business goals |
Cross-region disaster recovery
Deploy the disaster recovery system across two Alibaba Cloud regions. If a regional disaster disrupts the production site, fail over to the other region. Geographic separation ensures high reliability.
This scenario achieves an RPO of 1 minute and an RTO of 15 minutes.
To configure this scenario, follow Cross-region disaster recovery.
Cross-zone disaster recovery
Protects applications against force majeure events (fire, power outage) and equipment failures (software, hardware) within a single zone. Replicate across zones to meet your RPO and RTO goals.
To configure this scenario, follow Cross-zone disaster recovery.
Capabilities
| Capability | Description |
|---|---|
| Real-time data replication | Monitors disk-level changes on operating systems and ECS instances, captures data, and synchronizes it to disaster recovery sites in real time. Achieves second-level or minute-level RPO. |
| Quick application recovery | Runs applications in another region or zone within minutes after a failover. |
| Non-disruptive disaster recovery drills | Run DR drills on cloud-deployed servers at any time without affecting the production environment or interrupting replication. |
Benefits
| Benefit | Description |
|---|---|
| Cost-effective | Uses only disk and minimal compute resources. Requires only software-level authorization during disaster recovery. |
| Easy to use | Start replication, run DR drills, and restore data in one click. No dedicated disaster recovery center required. |
| Highly reliable | Maintains continuous replication and resumable transfers despite source server overloads, restarts, gateway power-offs, or network jitter. Alibaba Cloud guarantees cloud data reliability. |
| Highly secure | Encrypts data with AES-256 and HTTPS for end-to-end security. |
| Verifiable | Run DR drills on cloud systems at any time without affecting production. |
Billing
ECS disaster recovery incurs the following fees:
| Fee component | Billed to | Details |
|---|---|---|
| ECS disaster recovery software | Cloud Backup bill | Pricing of Cloud Backup |
| Cross-region replication traffic | Cloud Backup bill | Cross-region replication traffic generated by disaster recovery. Pricing of Cloud Backup |
| ECS instances and disks at the disaster recovery site | ECS bill | Pay-as-you-go ECS instances and disks at the disaster recovery site. Pay-as-you-go |
Other disaster recovery options
Cloud Backup also offers disk-based async replication for ECS disaster recovery. Overview of ECS disaster recovery (async replication)