Parity Declustering for Fault-Tolerant Storage Systems via $t$-designs

TL;DR

This paper introduces a novel method for parity declustering in fault-tolerant storage systems using $t$-designs, enabling efficient and uniform disk reconstruction for multiple disk failures.

Contribution

It extends existing parity declustering techniques from one-failure to multiple failures by leveraging $t$-designs, broadening applicability.

Findings

Provides a theoretical framework for $t$-design based declustering

Ensures uniform workload distribution during disk recovery

Supports arbitrary failure tolerance levels

Abstract

Parity declustering allows faster reconstruction of a disk array when some disk fails. Moreover, it guarantees uniform reconstruction workload on all surviving disks. It has been shown that parity declustering for one-failure tolerant array codes can be obtained via Balanced Incomplete Block Designs. We extend this technique for array codes that can tolerate an arbitrary number of disk failures via $t$-designs.