An analysis of latent sector errors in disk drives

Abstract

The reliability measures in today's disk drive-based storage systems focus predominantly on protecting against complete disk failures. Previous disk reliability studies have analyzed empirical data in an attempt to better understand and predict disk failure rates. Yet, very little is known about the incidence of latent sector errors i.e., errors that go undetected until the corresponding disk sectors are accessed.Our study analyzes data collected from production storage systems over 32 months across 1.53 million disks (both nearline and enterprise class). We analyze factors that impact latent sector errors, observe trends, and explore their implications on the design of reliability mechanisms in storage systems. To the best of our knowledge, this is the first study of such large scale -our sample size is at least an order of magnitude larger than previously published studies -and the first one to focus specifically on latent sector errors and their implications on the design and reliability of storage systems.

Improving the availability of supercomputer job input data using temporal replication

“…However, RAID is geared toward whole disk failures and does not address sector-level faults [2,11,19]. It is further impaired by controller failures and multiple disk failures within the same group.…”

Effects of Data Scrubbing on Reliability in Storage Systems

“…Each replica failure is assumed to be independent in this model, where latent sector errors in different disk drives are not correlated. According to [4], the latent sector errors develop aggressively in the initial 50 days and then stabilized with enterprise disk drives. When considering the latent sector errors in the enterprise disk drives, latent sector errors in different disk drives can be reasonably assumed to follows the exponential distribution.…”

“…Thus, the reliability model of storage systems has been based on the assumption that the data is mainly lost due to complete disk drive failures. However, recent studies [4], [5] show that factors other than complete disk drive failures influence data reliability.…”

Self-adjusting two-failure tolerant disk arrays

“…Observed data on the life expectancy of disks [1], [2] and the occurrence of latent disk sector errors [3] suggest that tolerance of at least two failures is necessary, possibly in conjunction with disk scrubbing or intra-disk redundancy [4].…”