A company replicates data between two ONTAP clusters for disaster recovery. Each cluster has multiple SVMs that serve data to different departments. During testing, the administrator needs to fail over a specific department's data to the secondary cluster while keeping other departments online. Which configuration allows this level of granular recovery?

A company replicates data between two ONTAP clusters for disaster recovery. Each cluster has multiple SVMs that serve data to different departments. During testing, the administrator needs to fail over a specific department's data to the secondary cluster while keeping other departments online. Which configuration allows this level of granular recovery?

• A. Node-based MetroCluster mirroring
• B. Volume-level SnapVault replication only
• C. Aggregate-level replication between clusters
• D. SVM-level SnapMirror replication

Answer: (D)

Granular disaster recovery is achieved by mirroring and failover at the level of the storage virtual machine (SVM). When you set up SnapMirror at the SVM level, you replicate the entire SVM namespace and its data to a partner cluster. This lets you fail over just that SVM to the secondary cluster, so only that department’s data becomes active on DR, while all other SVMs and their workloads remain online on the primary site. This approach provides isolation between departments because each SVM represents a separate namespace and set of volumes. You can promote the secondary cluster’s SVM for that department when needed, without impacting other departments’ SVMs. Other options are broader in scope. Node-based MetroCluster mirroring handles failover at the cluster/node level and typically affects larger portions of the environment rather than a single department’s namespace. Volume-level SnapMirror would require managing and failing over individual volumes, which is more operationally complex and doesn’t provide the clean, single-namespace failover that an SVM-level solution offers. Aggregate-level replication would include all data within an aggregate, impacting more than the targeted department. So, SVM-level SnapMirror replication is the right fit for per-department DR with minimal disruption to the rest of the users.

Granular disaster recovery is achieved by mirroring and failover at the level of the storage virtual machine (SVM). When you set up SnapMirror at the SVM level, you replicate the entire SVM namespace and its data to a partner cluster. This lets you fail over just that SVM to the secondary cluster, so only that department’s data becomes active on DR, while all other SVMs and their workloads remain online on the primary site.

This approach provides isolation between departments because each SVM represents a separate namespace and set of volumes. You can promote the secondary cluster’s SVM for that department when needed, without impacting other departments’ SVMs.

Other options are broader in scope. Node-based MetroCluster mirroring handles failover at the cluster/node level and typically affects larger portions of the environment rather than a single department’s namespace. Volume-level SnapMirror would require managing and failing over individual volumes, which is more operationally complex and doesn’t provide the clean, single-namespace failover that an SVM-level solution offers. Aggregate-level replication would include all data within an aggregate, impacting more than the targeted department.

So, SVM-level SnapMirror replication is the right fit for per-department DR with minimal disruption to the rest of the users.