Backblaze’s 2022 Drive Stats Reveal Manufacturer and Model Reliability Trends
On the Backblaze blog, Andy Klein writes:
Starting in Q1 of 2021 and continuing to the end of 2022, we can see that the overall rise in the overall AFR [annualized failure rate] over that time seems to be driven by Seagate and, to a lesser degree, Toshiba, although HGST contributes heavily to the Q1 2022 rise. In the case of Seagate, this makes sense as most of our Seagate drives are significantly older than any of the other manufacturers’ drives. Before you throw your Seagate and Toshiba drives in the trash, you might want to consider the lifecycle cost of a given hard drive model versus its failure rate.
I have long enjoyed reading Backblaze’s regularly released drive stats. Although the company’s data is less relevant for normal users than it used to be, given Backblaze’s increased use of 16 TB drives that few people need, it still provides a sense of which manufacturers and drives last longer than others.
In the Slashdot discussion about these results, there’s some skepticism because “… while the smaller drives failed more often last year, they are also older…”. Naturally, older HDs fail more often than newer ones, so commenters there seem to think Backblaze may be doing a somewhat apples-to-oranges comparison. (I haven’t looked at their stats directly this year; I typically only do that if I’m actually in the market for a new drive.)
Also, huge data-centers like theirs may not have similar usage patterns to what some of our personal drives would. And yet another issue I’ve noticed in recent years is that the true brand identities and manufacturing specifics of various HD models seem to get shuffled around rather often – you may not be able to buy the actual drives Backblaze did a few years ago.
Still, despite all that, their results are always interesting, and they seem to be the main ongoing source of such HD longevity statistics.
Their graph gives me concern about Seagate drives because I just purchased a Synology NAS system with two 8Tb Seagate “Wolf” drives (configured with Synology’s hybrid RAID). But when I read the article it says an 8TB Seagate drive was the only one with zero failures (couldn’t work out the model from the codes). Oh well!
It appears they only had 79 of those drives with an average age of 14 months so I’m not sure the zero figure means too much. I’ve been bitten many times in the past by Seagate so give them a wide berth. Hopefully you’ll have better luck.
Of course we’ve had other drives fail as well, just nothing like the rate of Seagates.
While I agree that I would think long and hard about their consumer Barracuda series, I haven’t read anything bad about their IronWolf series (intended for storage servers). I also haven’t heard anything bad about their enterprise drives.
Backblaze’s post talks about the fact that older drives fail more frequently than newer ones, so while that stat should be taken with an appropriate grain of salt, they’re being upfront about it.
The main thing that makes me feel like a second-class citizen is that I can’t even begin to imagine what I’d do with a 16 TB drive. I don’t think I have more than 5-6 TB of storage in the house, between all our functional Macs and archived material.
Rarely does a study so well match my personal experience. Even though I’ve been involved with Seagate since the company’s inception, I now refuse to buy their products and stick with Western Digital.
Media server?
These are cool stats.
But for those that use BackBlaze cloud to store your backups: does it inspire confidence to know they’re managing their own data center servers, or would you feel safer if they leased space from one of the big players?
“A-hoy there. We be p!rates, me dears, riding the interweb video seas!”
…that’s why you see Reddit threads on media servers, with users with tags like “125TB, 4K HDR 60%”
Backups. My Mac has 2TB of storage, of which about 1TB is used. My Time Machine volume is a 4TB drive and has about 2.5TB available (so 1.5TB used) since it went on-line in May 2022 (about 8 months ago). If I worked with audio and video media more, then I could see it filling much faster.
If your work uses 3-4TB of data, then 16TB for Time Machine makes perfect sense to me.
Even more so if you work with virtual machines. I tend to create VMs with virtual drives of 60-100GB each, with certain ones requiring much more. (e.g. one hosting Yocto builds to run Linux on embedded hardware can consume about 300GB for each build). It doesn’t take more than a few VMs like these to fill a 4TB drive, and if you want to make backups, well then that backup device has to be even larger.
For TM most folks will be better off going for smaller drives and swapping them more frequently.
Yes by. a Backblaze subscriber.
I have two Hitachi HDDs with one aged 14 years and the other aged 12 years in external enclosures and still in daily operation and still pass all the tests. But there has been a (unreasonable) turnover of HDDs that I have purchased within the last two years no matter the brand.
I once gave Seagate drives a miss but I did buy two about five years ago and they are performing well in comparison to other brands that have a better reputation. My conclusion now is that I have no clue as to which is the best manufacturer of HDDs. I wish Hitachi was still manufacturing drives.
Yeah, I realize that those are the two big examples, but I don’t collect digital video—streaming provides vastly more video than I have time to watch anyway—and even the backups of all the Macs in the house wouldn’t require 16 TB. My iMac’s internal drive is 1 TB and my Time Machine drive is a 2 TB SSD, and that works fine for the quantity of data that I change on a regular basis.
I had two 4 TB IronWolf drives fail in my Synology just after the warranty expired. I have also had bad luck with their portable drives. I’m steering clear. Now I buy only WDC hard drives (which have now lasted several years in my Synology) and SanDisk SSDs.