What is the difference between Continuous Data Replication (CDR) disaster recovery and async replication disaster recovery?
Async replication is a feature that provides cross-region or cross-zone data protection using the data replication capabilities of Elastic Block Storage. For more information, see How async replication works.
This topic describes the differences between async replication for cloud disks and Continuous Data Replication (CDR).
Comparison Criteria | CDR disaster recovery | Async replication disaster recovery |
Main scenarios | Suitable for single-instance disaster recovery. Suitable for customers who are not concerned about system intrusion. | Suitable for instance group consistency disaster recovery. Suitable for customers who want to avoid system intrusion. |
System intrusion | Yes | No |
Replication method | An agent is installed in the OS of the protected instance. It replicates data writes in real time to a gateway. The gateway sends the data to OSS. The data is then written to a disk at the disaster recovery site. | Data is replicated using Elastic Block Storage features, such as async replication and snapshots. |
Implementing recovery | Supports multiple recovery points. The backup site creates shadow ECS and gateway instances for the protected instance. It then pulls data from a data transit station, uses the shadow ECS instances to write the data, and creates recovery points through a snapshot mechanism. | Supports only a single recovery point. A recovery point is generated by copying a snapshot to the secondary site. |
Consistency group | Not supported | Supported |