Disc drive failure in datacentres contradicts the reliability rates promised by suppliers.
Researchers from Carnegie Mellon University in the US examined disc drive component failure covering 100,000 discs in large-scale IT installations, and discovered failure was becoming an increasing problem ahead of the disc’s expected lifecycle.
The researchers analysed disc replacement data from a number of large production systems, including high-performance computing sites and internet services sites.
The data included SCSI, Fibre Channel, and SATA-based discs. The mean time to failure (MTTF) of those drives, as specified in their supplier datasheets, ranged from one million to 1.5 million hours, suggesting a nominal annual failure rate of, at most, 0.88%.
But the researchers found that in the field, annual disc replacement rates typically exceeded 1%, with 2%-4% common, and up to 13% observed on some systems.
It was also found that there was little difference in replacement rates between SCSI, FC and SATA drives, suggesting disc-independent factors, such as operating conditions, affect replacement rates more than component specific factors.
Related article: High-end storage considerations
Comment on this article: firstname.lastname@example.org