Failure trends in a large disk drive population

TLDR

Hard disk drives store over 90 % of new information, yet little is known about their failure patterns and the factors influencing lifespan, especially because most studies rely on limited data or accelerated aging and lack real‑time health signal collection. The study aims to collect health signals from a large disk drive population to enable detailed failure analysis. Data were collected through detailed observations of a large disk drive population deployed in a production Internet services environment. The study found that SMART parameters strongly correlate with failures, yet models relying solely on them are unlikely to predict individual drive failures, and temperature and activity levels are less predictive than previously thought.

Abstract

It is estimated that over 90% of all new information produced in the world is being stored on magnetic media, most of it on hard disk drives. Despite their importance, there is relatively little published work on the failure patterns of disk drives, and the key factors that affect their lifetime. Most available data are either based on extrapolation from accelerated aging experiments or from relatively modest sized field studies. Moreover, larger population studies rarely have the infrastructure in place to collect health signals from components in operation, which is critical information for detailed failure analysis. We present data collected from detailed observations of a large disk drive population in a production Internet services deployment. The population observed is many times larger than that of previous studies. In addition to presenting failure statistics, we analyze the correlation between failures and several parameters generally believed to impact longevity. Our analysis identifies several parameters from the drive's self monitoring facility (SMART) that correlate highly with failures. Despite this high correlation, we conclude that models based on SMART parameters alone are unlikely to be useful for predicting individual drive failures. Surprisingly, we found that temperature and activity levels were much less correlated with drive failures than previously reported.