Storelab, a Russian player in data recovery, recently released a long-term study comparing hard drives from a number of different vendors. The controversial report offers interesting data and provides a snapshot of HDD reliability and defect reasons.
Reliability is hardly what consumers think most about when they decide to buy a storage product. Capacity, price, and speed receive most of the attention. The fact of the matter is that reliability is just assumed. This makes potential customers an easy target for marketing departments.
Furthermore, few people actually know what specific hard drive they get when buying an OEM PC because vendors tend to screen the identities of individual components. Many customers also get lulled into a false sense of security by warranty periods. A five-year warranty does not guarantee five years without hard drive failure, it just means you get a new drive should it actually fail. Statistical surveys of hard drive reliability are as rare as independent, long-term tests, but we found at least some insight through this analysis that we're discussing today.

Practical Relevance and General Validity
The Russian company Storelab, an Eastern European market leader in professional data recovery, is known for analysis and professional guidance published on its website. Storelab recently published the results of a long-term study based on its own operations and the observed failure rates of certain hard drives from leading manufacturers.
Storelab found some interesting results, and although the study is not large or scientific enough, it provides a useful insight into the hard drive market. While the hard drives from one manufacturer lasted 3.5 years on average, comparable models in capacity, features, and price from another manufacturer only lasted 1.5 years. Consumers might "only" lose their home photos and video (no small event when it means a decade of memories erased), but in the commercial sector, unexpectedly high failures can paralyze companies and destroy months of work. Even if safety measures have been taken to secure the data, premature drive failures mean increased costs due to labor, replacement, and interrupted operations. For these reasons, Storelab decided to publicly evaluate the failure data it had collected in order to demonstrate which manufacturer reaches the lowest failure rates.
We want to point out that Storelab's data consists exclusively of hard drives sent in for data recovery. This says nothing about the total number of claims handled directly by the retailer or manufacturer. But it does allow for detailed analysis of actual errors. It also needs to be said that drives sent in for data recovery have not necessarily been treated badly. An advantage of this analysis is the long-term factor, which usually only plays a minor role in market statistical surveys. With this all in mind, let's take a close look at the company's results!
Again, this is personal study - for example I have used only one Hitachi drive during this period, so I can't make statistics on the base of one.
Which is true, but only according to a limited, but still relevant (and one of the few available) study.
But, it correlates pretty well to what I experienced: the only failed Hitachi drives I got (and I got a bunch of them) were due to:
- manufacturing fault (and RMA worked well and fast) creating faulty sectors,
- old age (10 years) creating faulty sectors.
Never once, in the two dozen Hitachi (and IBM) drives I had to manage, was a failure related to hardware fault - even dropping a (not running) 2001 drive from one meter height on solid concrete failed to damage it.
I tried Maxtor, Toshiba, and WD drives, and most of these eventually developed unrecoverable hardware faults leading to massive data losses; Hitachi drives caused, at worst, a few files to be lost due to irrecoverable sectors (many times, I got SMART warnings before the data became impossible to rebuild).
I got some Samsung drives recently; they're cheap, and they added shock resistance capabilities in the 2.5" versions, I'll see if they do better than the Hitachi ones (I do backups on Hitachi drives still). If this report is any indication, Samsung notebook drives in RAID 1 array for a desktop machine should allow for reliable operation - the one I have carried around for 2 years in my fanny pack seems sturdy enough, too.
Seems a waste of time.
Which is true, but only according to a limited, but still relevant (and one of the few available) study.
But, it correlates pretty well to what I experienced: the only failed Hitachi drives I got (and I got a bunch of them) were due to:
- manufacturing fault (and RMA worked well and fast) creating faulty sectors,
- old age (10 years) creating faulty sectors.
Never once, in the two dozen Hitachi (and IBM) drives I had to manage, was a failure related to hardware fault - even dropping a (not running) 2001 drive from one meter height on solid concrete failed to damage it.
I tried Maxtor, Toshiba, and WD drives, and most of these eventually developed unrecoverable hardware faults leading to massive data losses; Hitachi drives caused, at worst, a few files to be lost due to irrecoverable sectors (many times, I got SMART warnings before the data became impossible to rebuild).
I got some Samsung drives recently; they're cheap, and they added shock resistance capabilities in the 2.5" versions, I'll see if they do better than the Hitachi ones (I do backups on Hitachi drives still). If this report is any indication, Samsung notebook drives in RAID 1 array for a desktop machine should allow for reliable operation - the one I have carried around for 2 years in my fanny pack seems sturdy enough, too.
Could this be part of the reason why Seagate has more than its fair share of drives sent in?
Again, this is personal study - for example I have used only one Hitachi drive during this period, so I can't make statistics on the base of one.
Yeah WD gave me a slightly upgraded HD with 40 more GB but all that info and files that were lost was irreplacable. The problem, was a faulty pin, or the reader that looks like a raptors nail! The device would just click and click and my HDD wouldnt read.
I know a lot of people who had problems with that particular "CC Fly", maybe that is why I don't like Seagate. Nowadays I'm turning to WD, I really want to see how are they doing.
Btw I have 3-4 40GB Maxtors that are still very much alive and working for about 9 years now
IMO 7200.12 with CC34 firmware is not that great either. FYII have another Seagate 7200.12 CC35 firmware and it has been ok, so it might be another firmware fiasco that Seagate doesn't want to acknowledge after the last generation "Seagate" case.
I was rather astonished to see the ratings on recent drives: when parked, up to 1000G!
One thing for sure: backup, backup and backup!
i did have a WD blue 640GB drive fail and got an advanced replacement. that is a great feature and i have bought several WD drives since then.
the longest running drive i have is an old Quantum Bigfoot from 98'. 5-1/4" 12GB monster that sounds like a vacuum cleaner when it spins up but it still works. not that i ever use it