Google uses millions of hard drives and started a study to that sampled one hundred thousand drives for five years to determine what caused their failure. The findings was published in a paper, here are the highlights:

• SMART (Self Monitoring And Reporting Technology) is a good indicator of impending mechanical failure, but a high amount of drives gave no warning even though SMART monitored items were the cause of the failure. If you have a drive that reports a SMART error you should perform a full backup of the drive immediately and make plans to replace it.
• The study found that increased temperature or activity had little or no correlation to the failure rate. Though, it was found that drives spinning up and spinning down most often had the highest failure rates. You might want to deactivate the Energy Saving functions that put a hard drive to sleep because of this.
• About 3% of drives failed in the first three months, 1.8% in the first 6 months, 1.7% in the first year. Failure rates jump to approximately 8% in the second year, 9% in the third year, fall to 6% in the fourth year, and jump back to 7% in the fifth year.

http://labs.google.com/papers/disk_failures.pdf

Tags: hard drive, hard drives, hard drive failure, smart, energy save, google, google labs