Author Topic: Hard Drive Brand Reliability Data (Read 21206 times)

mouser · « **on:** January 21, 2014, 08:19 PM »

We occasionally discuss the issue of hard drive reliability on the forum but this it the first time i can remember seeing hard data showing brands with clear differences in reliability..

Hitachi drives crush competing models from Seagate and Western Digital when it comes to reliability, according to data from cloud backup provider Backblaze. Their collection of more than 27,000 consumer-grade drives indicated that the Hitachi drives have a sub-2 percent annualized failure rate, compared to 3-4 percent for Western Digital models, and as high as 25 percent for some Seagate units.

Links:

kyrathaba · « **Reply #1 on:** January 21, 2014, 08:40 PM »

Interesting. I guess I know where I'm getting my next HD.

Shades · « **Reply #2 on:** January 21, 2014, 08:51 PM »

That graph matches with my experiences over here in Paraguay, South America. Between 500GByte and 1TByte Seagate drives are really bad, anything between 1 and 3 TByte is bad as well. However, the 500GByte, the 1 TByte and 3 TByte models perform well. Over here it is so bad that I don't even want a Seagate drive with an even number for storage capacity, even if you gave me the drive + money for free. The headache, misery and loss of time just isn't worth it.

Unfortunately, Seagate (and Samsung, if you count portable drives) is the only brand you can buy here directly. All other brands one needs to be ordered online and the extra shipping costs are prohibitive.

techidave · « **Reply #3 on:** January 21, 2014, 09:30 PM »

when it comes to laptop drives, my experience has been Hitachi and Toshiba have more problems than the WD drives that I use.

3.5" drives, I find that Samsung and WD are at the top of my list.

apankrat · « **Reply #4 on:** January 22, 2014, 03:45 AM »

One thing to keep in mind though is that these are failure rates under backblaze's usage. They are a backup provider. They may be routinely copying several very large files in parallel meaning higher than average and more random seeking, far more writing, cache thrashing, etc., which just may happen to correlate with Seagate's weak points and not those of WD. So you'd change the access pattern, you'd get a different failure profile by brand.

Jibz · « **Reply #5 on:** January 22, 2014, 04:14 AM »

Even if their use is a special case to some extent, I think a failure rate of over 10% is bad.

40hz · « **Reply #6 on:** January 22, 2014, 05:33 AM »

Even if their use is a special case to some extent, I think a failure rate of over 10% is bad.
-Jibz (January 22, 2014, 04:14 AM)

+1

FWIW, my goto brands are: Hitachi for 2" and WD (non-green) for desktop or basic server.

For commerce/enterprise servers I tend to stick with whatever the OEM ships with it. Doing so minimizes tech support arguments and warranty claims issues when dealing with the manufacturer - even if OEM-branded disks can (although not always) cost significantly more.

Time is money in enterprise. And downtime is big money!)

apankrat · « **Reply #7 on:** January 22, 2014, 07:06 AM »

Even if their use is a special case to some extent, I think a failure rate of over 10% is bad.
-Jibz (January 22, 2014, 04:14 AM)

True that. However it might be that other drives are also at 10%, but they just lucked out in BB's case :)

40hz · « **Reply #8 on:** January 22, 2014, 08:22 AM »

Even if their use is a special case to some extent, I think a failure rate of over 10% is bad.
-Jibz (January 22, 2014, 04:14 AM)

True that. However it might be that other drives are also at 10%, but they just lucked out in BB's case
-apankrat (January 22, 2014, 07:06 AM)

Still...according to BB's blog, they've had 12,675 Seagates in service, so I think it's likely to be more a case of 'trend' than 'luck' with Seagate at this point.

FWIW, the two drive brands I've seen fail more than any other make were Seagate and Maxtor - which is now owned by Seagate. Seagates usually had catastrophic 'head seek' failures whereas the Maxtor's controller boards tended to smoke with no warning.

Vurbal · « **Reply #9 on:** January 22, 2014, 01:23 PM »

FWIW, the two drive brands I've seen fail more than any other make were Seagate and Maxtor - which is now owned by Seagate. Seagates usually had catastrophic 'head seek' failures whereas the Maxtor's controller boards tended to smoke with no warning.
-40hz (January 22, 2014, 08:22 AM)

This is all from several years past but that's basically my experience as well.

Going back to the days when Maxtor was a low performance off-brand I found them to be incredibly reliable. After they grew into a major player I remember going through 5 or 6 cases (48 drives per case) of 5400 RPM PATA drives with early (within a few weeks) failure rates ranging between 20% and 40% for each and every case. Actually I'm not sure any were as low as 20%. In every case it was the electronics that failed.

For certain periods I got similar results from WD's OEM units (but never their retail models) although the sample size was much smaller - only about 30 drives.

IainB · « **Reply #10 on:** January 22, 2014, 06:08 PM »

Very interesting. Not often one sees hard data about hard drive failure-rates.
One needs to be careful with these statistics. For example, whilst the brand name is shown , the failure is relevant to that brand only by the given drive type/size within that brand name, and YMMV as these failures presumably occurred under operating conditions that might not correspond to one's own peculiar operating conditions.

I recall another post in a website a while back that also had data to back up the relatively low incidence of Western Digital drive failure as compared to Seagate.
It may be that the pattern is - or would be - noticeable across the range of brands and sizes, but the data has not been aggregated sufficiently or published for us to be able to come up with some general rule of thumb. If we did have that form of data, it would make the market more perfect (in the economic sense) and there could well be a lot of newly-educated consumers voting with their feet.

This might even be considered by some as being a good reason for suppressing the publication of such data.

mouser · « **Reply #11 on:** January 22, 2014, 06:25 PM »

Some comments to take note of from the backblaze blog:

The drives that just don’t work in our environment are Western Digital Green 3TB drives and Seagate LP (low power) 2TB drives. Both of these drives start accumulating errors as soon as they are put into production. We think this is related to vibration. The drives do somewhat better in the new low-vibration Backblaze Storage Pod, but still not well enough.

These drives are designed to be energy-efficient, and spin down aggressively when not in use. In the Backblaze environment, they spin down frequently, and then spin right back up. We think that this causes a lot of wear on the drive.

mouser · « **Reply #12 on:** January 30, 2014, 01:47 AM »

From slashdot:

"A recent study of hard drive reliability by Backblaze was deeply flawed, according to Henry Newman, a longtime HPC storage consultant. Writing in Enterprise Storage Forum, Newman notes that the tested Seagate drives that had a high failure rate were either very old or had known issues. The study also failed to address manufacturer's specifications, drive burn-in and data reliability, among other issues. 'The oldest drive in the list is the Seagate Barracuda 1.5 TB drive from 2006. A drive that is almost 8 years old! Since it is well known in study after study that disk drives last about 5 years and no other drive is that old, I find it pretty disingenuous to leave out that information. Add to this that the Seagate 1.5 TB has a well-known problem that Seagate publicly admitted to, it is no surprise that these old drives are failing.'"

nosh · « **Reply #13 on:** January 30, 2014, 02:28 AM »

I've hardly ever used drives other than Seagate and I've yet to encounter one that flaked out in the first few years of use.
OTOH, I've used a handful of WD drives and around half of them had problems a lot earlier than I was accustomed to.
Maybe it's a regional thing? More likely, just dumb luck.

40hz · « **Reply #14 on:** January 30, 2014, 08:18 AM »

I've hardly ever used drives other than Seagate and I've yet to encounter one that flaked out in the first few years of use.
OTOH, I've used a handful of WD drives and around half of them had problems a lot earlier than I was accustomed to.
Maybe it's a regional thing? More likely, just dumb luck.
-nosh (January 30, 2014, 02:28 AM)

I think it probably has as much to do with the manufacturing batch and model as it does with the brand for consumer drives. These low-cost/high-capacity drives aren't tested all that extensively. And the test sample size is kept smaller than for 'enterprise' drives in order to keep the production costs down. Far cheaper for a manufacturer to allow for a higher failure rate and simply replace drives under warranty, than it is to more thoroughly QC all the drives in a production run.

Consumer drives also have a greater likelihood of being 'accidentally' manhandled in retail shipment and storage. That may not be enough to cause a drive to fail right out of the box. But it can lead to premature failure. I've personally heard former Big Box and office store employees boast about slamming electronics around whenever The Boss wasn't looking. And there are plenty of YouTube videos of postal and other delivery drivers slam dunking boxes marked "fragile" on porches to not take that possibility into account either.

At the end of the day, I think a drive just lives as long as it does. (Much like us!) You can do some things to up the likelihood of a long productive life. But there are no guarantees. Only trends and probabilities.

Giampy · « **Reply #15 on:** February 20, 2014, 06:58 AM »

I've used a handful of WD drives and around half of them had problems a lot earlier than I was accustomed to.
Maybe it's a regional thing? More likely, just dumb luck.
-nosh (January 30, 2014, 02:28 AM)

Yesterday my WDC WD5000AAKX-00ERMAO passed away, after only 4 months of service.

RIP

app103 · « **Reply #16 on:** February 20, 2014, 05:17 PM »

I have never had a hard drive die on me yet, but recently had issues with one and decided to replace it...and it was a Hitachi:

http://www.hddstatus...erification=49B3E62B

I still have Quantum Fireballs that are running just fine.

x16wda · « **Reply #17 on:** February 20, 2014, 08:34 PM »

I still have Quantum Fireballs that are running just fine.
-app103 (February 20, 2014, 05:17 PM)

Just send 'em on over, I can run 'em in a textile plant this summer for you.... just to burn them in, ya know....

Giampy · « **Reply #18 on:** February 21, 2014, 04:05 AM »

Yesterday my WDC WD5000AAKX-00ERMAO passed away, after only 4 months of service.

RIP
-Giampy (February 20, 2014, 06:58 AM)

My trusted technician has changed that hard disk for free. He has said that among 200 hard disks WDC WD5000AAKX-00ERMAO mine is the only one to get broken.
I care to say that as I don't want to put unjustly WD in a bad light.

brotman · « **Reply #19 on:** May 22, 2014, 11:46 PM »

Has it occurred to anyone else that the failure data might be a bit skewed since the larger (e.g. 4 gb) models have been around for a shorter time than the others???

xtabber · « **Reply #20 on:** May 18, 2016, 01:07 PM »

Backblaze has published updated hard drive reliability data through the first quarter of 2016. Australian computational biologist Ross Lazarus has updated his survival analysis to include the 2016 data.

As before, HGST comes out by far the best and Seagate the worst.

But things are not quite that simple. Seagate's reliability has improved since Backblaze started publishing this in 2013 and Western Digital's declining, placing them below Seagate for newer drives. They also have far more Seagate drives than any others and are now buying fewer WD and Toshiba drives -- not because of reliability, but because they can buy Seagates cheaply while WD and Toshiba are hard to find in the bulk quantities they need.

On top of that, WD bought HGST from Hitachi in 2012 and divested HGST's manufacturing facilities by selling some of them to Toshiba.

So which drive should you purchase if you are looking for reliability? Quien sabe?

Deozaan · « **Reply #21 on:** May 18, 2016, 05:38 PM »

I recently purchased a WD Black 5TB HDD, mostly because it comes with a 5 year warranty, whereas most other drives have (IIRC) 2 year (or less) warranties.

I don't know if they're really more reliable than any other drive, but just the fact that they're confident enough in them to warranty them for 5 years when other companies only do 2 years was enough to convince me that it was worth the extra money.

MilesAhead · « **Reply #22 on:** May 18, 2016, 05:52 PM »

I recently purchased a WD Black 5TB HDD
-Deozaan (May 18, 2016, 05:38 PM)

Obliquely on topic perhaps.. but I am curious how you partition and use the 5 TB and what OS etc.. I had a few WD Blacks but I stopped at the 1 TB capacity. That was the sweet spot for the online discount house I used to purchase storage. I have seen some Windows people have weird things happen with storage larger than 2 TB. That is why I am curious.

Deozaan · « **Reply #23 on:** May 18, 2016, 06:27 PM »

It's just one massive 5 TB (4.54 TB formatted space) partition on Windows 10. It's my main data drive. My Documents, downloads, disc images, videos, games, music, virtual machines, source code repositories, etc., all go on this drive. And until recently even my Temp directory was on this drive. But now I have a 6GB RAMDisk for that.

I had reason to believe that my 2TB drive was slowly dying, so I replaced it with the 5TB drive. And by replaced it, I mean I added in the 5TB drive that I now use for the same purpose I used my 2TB drive, and the 2TB drive is now my "backup" drive.

Hard Drive Brand Reliability Data

I haven't had any problems with the large partition. But I'm running a 64-bit OS and haven't attempted any older, <32-bit software on it.