I just read, from a Slashdot posting, Google have just released an analysis of their experience with disk failures titled “Failure Trends in a Large Disk Drive Population” and it offers some surprising conclusions such as (the emphasis is mine):
… we did not find much correlation between failure rate and either elevated temperature or utilization … Our results confirm the findings of previous smaller population studies that suggest that some of the SMART parameters are well-correlated with higher failure probabilities. We find, for example, that after their first scan error, drives are 39 times more likely to fail within 60 days than drives with no such errors. … Despite those strong correlations, we find that failure prediction models based on SMART parameters alone are likely to be severely limited in their prediction accuracy, given that a large fraction of our failed drives have shown no SMART error signals whatsoever. This result suggests that SMART models are more useful in predicting trends for large aggregate populations than for individual components. It also suggests that powerful predictive models need to make use of signals beyond those provided by SMART.
What does this all mean? It means that predicting drive failures is at present more-or-less impossible with any kind of accuracy.
