A Study of SSD Reliability in Large Scale Enterprise Storage Deployments

We present the first large-scale field study of NAND-based SSDs deployed in enterprise storage systems. Our study is based on field data, collected over 2.5 years, for a sample of almost 1.4 million drives from the total SSD population of a major enterprise storage vendor (NetApp). The data allows us to study a large number of factors that were not included in prior work, such as the effect of firmware versions, the reliability of TLC NAND, and correlations between drives within a RAID group. Our analysis provides insight into flash reliability, along with a number of practical implications.