What approach does SnapMirror use to maintain data consistency across asynchronous replication when there is a WAN delay?

What approach does SnapMirror use to maintain data consistency across asynchronous replication when there is a WAN delay?

• A. Scheduled transfers and applies consistent snaps on the target; application quiescence improves consistency and reduces RPO variability.
• B. Real-time synchronous replication over WAN.
• C. Backups to tape.
• D. Manual snapshot recreation.

Answer: (A)

This question focuses on how SnapMirror keeps data consistent when replication is asynchronous and there’s WAN delay. The approach relies on scheduled transfers of point-in-time images that are made consistent by quiescing the applications before the snapshot. By taking a quiesced snapshot on the source, you capture a coherent state of the data (a transaction- or crash-consistent image). This consistent image is then transferred to the target and applied, so the replica reflects a known good state at the moment the snapshot was created. WAN delay makes real-time, continuous syncing impractical for strict consistency, so using scheduled transfers of consistent snaps provides predictable, repeatable recovery points. Application quiescence improves data integrity by ensuring that the snapshot captures a stable state across applications, which in turn reduces RPO variability—the difference in potential data loss between the source and the latest replica becomes more tightly bounded. Other approaches don’t fit this scenario: attempting real-time synchronous replication over a WAN isn’t typically feasible due to latency and bandwidth limits, and backups to tape or manual snapshot recreation aren’t designed to maintain ongoing, consistent replication between sites.

This question focuses on how SnapMirror keeps data consistent when replication is asynchronous and there’s WAN delay. The approach relies on scheduled transfers of point-in-time images that are made consistent by quiescing the applications before the snapshot. By taking a quiesced snapshot on the source, you capture a coherent state of the data (a transaction- or crash-consistent image). This consistent image is then transferred to the target and applied, so the replica reflects a known good state at the moment the snapshot was created.

WAN delay makes real-time, continuous syncing impractical for strict consistency, so using scheduled transfers of consistent snaps provides predictable, repeatable recovery points. Application quiescence improves data integrity by ensuring that the snapshot captures a stable state across applications, which in turn reduces RPO variability—the difference in potential data loss between the source and the latest replica becomes more tightly bounded.

Other approaches don’t fit this scenario: attempting real-time synchronous replication over a WAN isn’t typically feasible due to latency and bandwidth limits, and backups to tape or manual snapshot recreation aren’t designed to maintain ongoing, consistent replication between sites.