Cloud storage companies don't just manage an obscene number of hard drives, they also deal with a huge number of failures. While most companies keep mum on things like failure rates, BackBlaze prefers to let the rest of us know who we should trust.
Well, as a brand-new report highlights, that situation hasn't changed much at all. In fact, Seagate's 3 terabyte failure rates are downright abysmal. The following chart speaks volumes (no pun):
BackBlaze deems a drive to be a failure when: A) It won't spin up; 2) It won't stay synced in a RAID; or 3) Its SMART values are outside of comfortable levels.
After crunching the numbers, there are a couple big takeaways. First, 4TB drives are great, with their solid blend of pricing, storage size, and durability. BackBlaze says that it's pleased with all of the 4TB drives it put into action during 2014.
Just like it is for us regular consumers, price matters a lot when it comes to cloud storage services choosing drives. That's the reason WD's 4TB drives were largely ignored by BackBlaze during 2014 - HGST and Seagate offerings were a better value. In a way, that's unfortunate, because it would have been nice to have a fuller picture.
Years ago, 1.5TB drives were notorious for dying more regularly than more normal densities, whereas now, it seems that 3TB drives have taken its place. Across all three vendors, 3TB drives had a higher failure rate during 2014 than the other densities, but none as bad as Seagate. In fact, the situation is so extensive that BackBlaze has promised a dedicated blog post down-the-road to talk more about it.
As for 6TB drives? BackBlaze doesn't have enough of those in use at the moment, so it's hard to gauge their durability. This is something I'm actually very interested in, as I'm planning to soon build a FreeNAS box with 6TB drives. Oh well, despite that lacking information, BackBlaze has done us all a great favor by keeping us in the loop. Good for us, not so good for Seagate, it seems.