A Group Repair Codes with Low Recovery-overhead in Distributed Storage System

作     者：Deng, Wenjie Li, Cong Hong, Tieyuan Tang, Dan 

作者机构：School of Software Engineering Chengdu University of Information Technology Sichuan Province Engineering Technology Research Center of Support Software of Informatization Application Chengdu610225 China 

出 版 物：《International Journal of Network Security》 (Int. J. Netw. Secur.)

年 卷 期：2024年第26卷第4期

页      面：610-621页

核心收录：

基　　金：This work was supported by the Science and Technology Support Project of Sichuan Province (2022YFG0037) and (2022YFG0033) 

主　　题：Bandwidth 

摘      要：In distributed storage systems, erasure codes are widely used to ensure data reliability and provide higher storage efficiency than replication techniques. However, the cost of using erasure codes is the increased network bandwidth required during the repair process, which can lead to performance bottlenecks in the system. This paper proposes a CGRC (Crisscross Group Repair Codes), which constructs an encoding array to generate multiple types of parity blocks in a cross-interleaved manner and adjusts the distribution of data blocks within the parity block groups. This approach aims to reduce the bandwidth overhead for recovery while maintaining fault tolerance performance. Additionally, the paper presents an efficient decoding algorithm that achieves optimal repair efficiency for CGRC. Simulation results indicate that CGRC, compared with other block codes, the recovery efficiency of single-node can be improved by 48.56% at most, 77.5% compared with RS, and the recovery efficiency of multinodes can be improved by 16.9% at most compared with other block code. Moreover, CGRC ensures higher fault tolerance performance and increases reliability under the same storage efficiency. © (2024). All Rights Reserved.