Hacker News Reader

Comment by BugsJustFindMe | Hacker News Reader

The missing piece here is to cross-reference with a failure/reliability reporting service to produce Price per TB Hour.

It seems to me that in modern drives, there is no longer that much variance in average longevity. At least close enough that when I have tried to compare drives, the cheaper drive almost always ends up being the cheaper drive even when adjusted for longevity.

I can't prove it but it sure seems like once BackBlaze started posting reliability figures from their fleet it forced the hand of any manufacturer that was cutting corners to keep up their reliability or get hammered in the market.

I still have a bad taste in my mouth with regard to the Seagate HDD failure plague from about a decade ago.

Kudos to the BackBlaze team

I still reference the latest BackBlaze stats when I'm buying a drive, and only purchase models which have fared well.

Better to pay a premium up front rather than risk total loss when it dies prematurely. Seagate has screwed me over one too many times..

I go the exact opposite way these days, I buy 3 drives from 3 different brands and run them in a ZFS raidz. This reduces the risk of all the drives failing at the same time (which can happen with any batch of disks) but also makes it easier to determine which physical drive needs replacing when you have a failed disk in a NAS (since each disk is unique).

Reply on news.ycombinator.com