All Products
Search

This Product

    E-MapReduce:Disaster recovery management

    Document Center

    E-MapReduce:Disaster recovery management

    Last Updated:Jul 17, 2026

    Multi-zone deployment for instances with a storage-compute separation architecture keeps data accessible even if an availability zone (AZ) becomes unavailable. The underlying data uses zone-redundant storage for higher reliability.

    Comparison of availability zone types

    Metric

    Single-zone

    Two-zone

    Three-zone

    Disaster recovery capability

    Does not support cross-datacenter disaster recovery.

    Withstands a single availability zone failure.

    Withstands two concurrent availability zone failures.

    Data storage

    Data is stored redundantly across multiple devices and facilities within the same availability zone.

    Data is stored redundantly across multiple availability zones within the same region.

    Fault recovery capability

    BE/CN node failures are automatically recovered. A zone-level failure interrupts all tasks.

    If the primary availability zone fails, the FE and compute nodes can be switched to a standby zone to maintain service continuity.

    If the primary availability zone fails, you can select a healthy standby zone and perform a switchover. The system migrates the FE and compute nodes to that zone to maintain service continuity.

    Deployment cost

    No additional cost.

    • Cost of two sets of FE nodes.

    • Cost of one set of compute nodes.

    • Additional cost for data storage.

    • Cost of three sets of FE nodes.

    • Cost of one set of compute nodes.

    • Additional cost for data storage.

    Limitations

    • This feature is available only for instances that use a storage and compute separation architecture.

    • You cannot change the primary availability zone after creating the instance.

    • This feature does not support cross-region deployments.

    • After a switchover, the system migrates the frontend and compute nodes to the standby availability zone. The instance's network configuration and domain name remain unchanged. Ensure that the vSwitch in the standby availability zone has sufficient available IP addresses.

    Usage notes

    • After you enable multi-zone disaster recovery, you cannot disable it.

    • If the production instance fails before some data is fully synchronized to the disaster recovery nodes, that unsynchronized data is lost after a switchover.

    • When you adjust FE resources, for example, by scaling out nodes, upgrading configurations, or expanding disks, the system synchronizes the Observer resources in the standby availability zone with the FE resources.

    Billing

    This feature incurs the following additional costs:

    • This feature requires additional FE nodes, which incur extra CU cost. This charge appears on your bill as OBSERVER Computing Resources. For more information, see CU cost.

    • The underlying storage uses zone-redundant storage, which incurs additional storage costs. For more information, see Data storage (multi-zone) fees.

    Enable multi-zone disaster recovery

    New instance

    1. Go to the E-MapReduce Serverless StarRocks instance list page.

      1. Log on to the E-MapReduce console.

      2. In the navigation pane on the left, choose EMR Serverless > StarRocks.

      3. In the top menu bar, select the required region.

    2. On the Instances page, click Create Instance.

    3. On the EMR Serverless StarRocks page, for the Multi-zone Disaster Recovery parameter, select Two Zones or Three Zones and specify the backup VSwitches for the standby zones.

      For more information about how to create an instance, see Create an instance.

    Existing instance

    1. Go to the disaster recovery management page.

      1. Log on to the E-MapReduce console.

      2. In the left navigation pane, select EMR Serverless > StarRocks.

      3. Click the name of the target instance.

      4. Click the Disaster Recovery tab.

    2. On the Disaster Recovery tab, click Disaster Recovery Settings.

    3. In the panel that appears, for the Multi-zone Disaster Recovery parameter, select Two Zones or Three Zones and specify the backup VSwitches for the standby zones.

    4. Select the service agreement and click OK.

    Switch availability zones

    Important

    • A zone switchover makes the instance unavailable for one to two hours. To minimize business impact, perform this operation only when the instance is completely down and cannot be recovered.

    • The switchover may fail if the destination availability zone has insufficient server inventory. We recommend submitting a ticket in advance to reserve resources in the destination zone.

    1. Go to the disaster recovery management page.

      1. Log on to the E-MapReduce console.

      2. In the left navigation pane, select EMR Serverless > StarRocks.

      3. Click the name of the target instance.

      4. Click the Disaster Recovery tab.

    2. In the row of the target standby availability zone, click Switch Zone.

    3. In the dialog box that appears, enter the name of the current instance to confirm, and then click OK.