System and method for configuring a device array upon detecting addition of a storage device

摘要

A system, method and computer program product for detecting an additional storage device within an “n” device array. The “n” device array is configured to store “n” device array formatted data. The “n” device array is reconfigured into an “n+m” device array. The “n” device array formatted data is written to the “n+m” device array in an “n+m” device array format. Reconfiguring the “n” device array may include converting the “n” device array formatted data into “n+m” device array formatted data. The “n+m” device array may comprise two storage devices including a mirrored storage device, or may include at least three storage devices including a coded target storage device such as a parity storage device. The coded target storage device may be a distributed coded target as in a RAID array.

主权项

1. A method of configuring a device array comprising:
detecting an additional storage device within an “n” device array including a flash solid-state storage device, wherein the “n” device array is configured to store “n” device array formatted data including a plurality of data segments having “x” data elements; reconfiguring the “n” device array into an “n+m” device array; writing the “n” device array formatted data to the “n+m device array in an “n+m” device array format; wherein reconfiguring the “n” device array into an “n+m” device array includes converting the “n” device array formatted data into the “n+m” device array format by forming data segments having “x+m” data elements from the data segments having “x” data elements, wherein each of the “x+m” data elements are equal in size to each of the “x” data elements; wherein writing the “n” device array formatted data to the “n+m” device array includes writing each data segment having “x+m” data elements to the “n+m” device array after each data segment having “x+m” data elements is formed and a coded element for each data segment is calculated, wherein the coded element for each data segment is calculated and written once, such that a number of write operations executed on the flash solid-state storage device is reduced; after writing the “n” device array formatted data to the “n+m” device array in the “n+m” device array format, if an insufficient number of data elements exist to form a complete data segment, utilizing filler data elements to complete the data segment; and determining a preferred high-availability format for the “n+m” device array based upon, at least in part, a number of flash solid-state storage devices in the “n+m” device array.