I’ve written a edition of this for every Backblaze post we do, so here’s the newest. Backblaze’s data set is not perfect. The agency uses client drives in a decidedly enterprise context, which doesn’t remotely reflect the usage styles these drives would accept in the event that they were being used by a standard purchaser. It constructs and maintains its own storage pods and has iterated on these designs through the years, often with advancements meant to minimize vibration or noise. This can also theoretically influence drive reliability.
As the video below illustrates, hard drives don’t adore it if you happen to scream at them. Always make certain to speak along with your spinning media in a considerate, calm voice. The reason we continue to talk about and feature Backblaze’s data set is a similar reason we consult with the Steam Hardware Survey. We know the info sets are imperfect — the SHS’s list of video cards doesn’t encompass diverse AMD GPUs like Vega 56 or 64 and only added the RX 570 last month. The total variety of “Other” GPUs listed is 10.
92 percent — good enough to meaningfully impact the AMD versus Nvidia split listed at the end level of the page. But while these data sets are imperfect, they’re also one of the best window we have into an important question. HDD brands do not free up the sort of reliability advice that HDD clients often want and the handful of third party stories on the topic commonly don’t determine drive owners or models.