Adversarial Error Resilience in Distributed Storage Using MRD Codes and MDS Array Codes

This paper presents a novel coding scheme for distributed storage systems containing nodes with adversarial errors. The key challenge in such systems is the propagation of the erroneous data from a single corrupted node to the rest of the system during node repair process. We present a concatenated coding scheme which is based on two types of codes: maximum rank distance (MRD) code as an outer code and optimal repair maximal distance separable (MDS) array code as an inner code. We prove that this coding scheme attains the upper bound on the resilience capacity, i.e., amount of data stored reliably in a system with a limited number of corrupted nodes.