Evaluating approximately balanced parity-declustered data layouts for disk arrays

Optimizing large data transfers in parity-declustered data layouts

“…Their depth is k(v − 1). • Randomized Permutation Layouts (RPL): These data layouts modify a randomized construction due to Schwabe, Sutherland, and Holmer [19]. They can be constructed for any v and k, and satisfy Conditions 1, 2, 4, and 6.…”

“…This tradeoff of additional redundant space for reduced failure recovery time can be achieved using a technique called parity declustering, in which the stripe size k is chosen to be smaller than the array size v. Parity-declustered data layouts have been considered by, among others, Muntz and Lui [12], Holland and Gibson [10], Stockmeyer [20], Schwabe and Sutherland [16,18], Schwabe, Sutherland, and Holmer [19], Alvarez, Burkhard, and Cristian [1], and Alvarez, Burkhard, Stockmeyer, and Cristian [2].…”

“…Schwabe, Sutherland, and Holmer [19] introduced randomized constructions that applied to even wider classes of arrays, but in doing so weakened their satisfaction of Condition 3 by allowing the reconstruction workload to be only approximately balanced.…”