Outshining Mirrors: MTTDL of Fixed-Order SSPiRAL Layouts

Complementary trends in hardware and applications are driving an increase in demand for data volume and bandwidth, resulting in an increased risk of data loss and a growing need for improved storage reliability. There is a growing need to survive the failure of multiple storage devices in larger storage arrays, as well as the need to survive the loss of multiple nodes in clustered storage. Redundant storage schemes are the obvious solution, and such applications commonly employ one of two strategies: a combination of replication and parity applied efficiently across an array of devices, or a failure-recovery scheme based on erasure coding.