On Fault Tolerance of Data Storage Systems: A Holistic Perspective

Abstract: Data storage systems serve as the foundation of digital society. The enormous data generated by people on a daily basis make the fault tolerance of data storage systems increasingly important. Unfortunately, modern storage systems consist of complicated hardware and software layers interacting with each other, which may contain latent bugs that elude extensive testing and lead to data corruption, system downtime, or even unrecoverable data loss in practice. In this chapter, we take a holistic view to introduce the typical architecture and major components of modern data storage systems (e.g., solid state drives, persistent memories, local file systems, and distributed storage management at scale). Next, we discuss a few representative bug detection and fault tolerance techniques across layers with a focus on issues that affect system recovery and data integrity. Finally, we conclude with open challenges and future work.