RAID-0e: A Resilient Striping Array Architecture for Balanced Performance and Availability
Yanzhao Jia, Zhaobo Wu, Zheyi Cao, Shihao Ji, Xu Tianhao, Zihui Song
TL;DR
RAID-0e addresses the problem of achieving higher data availability without sacrificing RAID 0 read performance. It introduces a separate Parity Domain to provide recovery from common, non-catastrophic media faults without sacrificing read throughput while incurring a scalable overhead of $N+1$ or $N+M$ models. The paper details the architectural principles, the parity computation and on-the-fly recovery workflows, redundancy and performance analyses, failure modes, and practical use cases, along with limitations and avenues for future work. The approach offers a scalable, cost-efficient alternative for workloads where read performance is critical and full drive failures are less frequent than media degradation, effectively balancing performance, cost, and resilience.
Abstract
This paper introduces a novel disk array architecture, designated RAID-0e (Resilient Striping Array), designed to superimpose a low-overhead fault tolerance layer upon traditional RAID 0 (striping). By employing a logically and physically separate parity domain to protect a primary data domain, RAID-0e mitigates the risk of array-wide data loss from common, non-catastrophic media failures, such as isolated bad blocks, transient read errors, or sector-level corruption. The architecture is engineered to preserve the intrinsic read performance advantages of RAID 0 while significantly enhancing data availability and operational resilience. This document provides a comprehensive exposition of the architectural principles, operational workflows, performance characteristics, failure mode analysis, and security considerations of RAID-0e. It is presented as an experimental yet pragmatic solution for environments seeking a new equilibrium between I/O performance, storage cost, and data resilience, particularly where full drive failure is a secondary concern to media degradation.