For the last few years, Backblaze has released quarterly reports on its hard drive reliability and replacement rates. Backblaze is a provider of cloud backup services and uses its own custom storage solutions (dubbed Drive Pods) to hold hundreds of gigabytes of data in each pod. This month, the company detailed a significant shift in its own deployment practices. Over the last three months, Backblaze has swapped out 3,500 2TB HGST and Western Digital hard drives for 2,400 Seagate 8GB drives. This resulted in a net drop in drive deployments, though it also delivered a significant increase in total storage capacity.
We’ve collected Backblaze’s reliability information in the slide show below. All slides can be clicked on to enlarge them and open them in a new window.
The trends here match what we’ve observed in previous quarters, and there aren’t any significant changes to report to relative drive rankings. When you check drive failure rates, however, make certain to reference the number of hours the drives have been in operation as well. Drives with fewer than 100,000 recorded hours is a sign that Backblaze never deployed very many of those drives to start with. Some of the high failure rates could be explained by small sample sizes — the smaller the sample size, the greater the chance that bad luck will skew the data. Even if we only look at the drives with 500,000+ hours of use, however, we still see significant failure variation in between families and manufacturers. Western Digital’s 3TB drives fail far more often than HGST’s, while Seagate’s 4TB drives have a much higher failure rate than its 6TB drives.
Backblaze has explained before that it can tolerate a relatively high failure rate before it starts avoiding drives altogether, but the company has been known to take that step (it stopped using a specific type of Seagate drive at one point due to unacceptably high failure rates). Current Seagate drives have been much better and the company’s 8TB drives are showing an excellent annualized failure rate, though they’ve also only been in operation for a few weeks.
As always, Backblaze’s data sets should be taken as a representative sample of how drives perform in this specific workload. Backblaze’s buying practices prioritize low cost drives over any other type, and they don’t buy the enterprise drives that WD, Seagate, and other manufacturers position specifically for these kinds of deployments. Whether or not this has any impact on consumer drive failure rates isn’t known — HDD manufacturers advertise their enterprise hardware as having gone through additional validation and being designed specifically for high-vibration environments, but there are few studies on whether or not these claims result in meaningfully better performance or reliability.
Backblaze’s operating environment has very little in common with a consumer desktop or laptop and may not cleanly match the failure rates we would see in these products. The company readily acknowledges these limitations, but continues to provide its data on the grounds that having some information about real-world failure rates and how long hard drives live for is better than having none at all. We agree. Readers often ask which hard drive brands are the most reliable, but this information is extremely difficult to come by. Most studies of real-world failure rates don’t name brands or manufacturers, which limits their real-world applicability.