A dual-controller, active/active storage array is servicing I/O for 10 mission-critical systems. Controller A loses power and fails and all traffic is immediately routed to controller B and all pendin

Battery backed cache provides power to the cache modules during an unexpected power outage, allowing data to be flushed to non-volatile storage, but it does not directly facilitate the transfer of uncommitted data to a peer controller upon a controller failure.

Data mirroring (e.g., RAID 1 or RAID 10) provides redundancy for data stored on physical disks, protecting against disk failures, but it does not specifically address the integrity of data residing in a controller's volatile cache during a controller failure.

Cache mirroring, also known as cache coherency or controller-to-controller cache synchronization, ensures that data written to one controller's cache is simultaneously mirrored to the peer controller's cache. This allows the surviving controller to commit any unwritten data from the failed controller's cache to disk, preventing data loss during a controller failure.

Disk mirroring is a form of data redundancy (like RAID 1) at the disk level, which protects against disk failures by writing identical data to two or more disks, but it is distinct from protecting uncommitted data in the controller cache.