Twenty years ago the hard drives I recall using might not have been fast in the data transfer stakes, they were certainly not lithe and sexy pieces of hardware and I prefer not to think how much I actually paid for a large and clunky bit of kit to store hardly any data at all. One thing they did have going for them was reliability though. Indeed, in the space of a decade I think I only needed to replace a single failed unit across a dozen or so computers. And that had something to do with me dropping the amber gas plasma screened, laughingly called a laptop, device while trying to live the mobile computing fantasy we had all been sold.

Fast forward to today, and in the last five years I have replaced no less than six hard drives which have mostly failed without warning, a couple within months of purchase and all for no good reason other than sloppy quality control and a market which has driven prices down at the cost of reliability.

Which is why I had to laugh when Fujitsu presented me with a press release today in order to 'big up' its latest innovation: the hard drive designed for continuous operation.

Yes, you read that right. The unique selling point of the Fujitsu MHY2 BS-series of 2.5" SATA hard drives would appear to be that they have been designed not to fall over at the slightest provocation, like actually using them a lot. OK, so the marketing push is aiming these at the bank teller ATM machines, point of sale systems and entry-level servers etc. But surely the point is that all hard drives should be engineered not to fail, designed to run for as long as the end user wants, without an expectation of disaster should you cross some unnamed time barrier which thrusts you into a continuous user bracket.

I want every hard drive I buy to feature 'high frequency access' and 'low heat generation' not to mention reduced energy consumption and integrated technology to 'minimize the effects of rotational vibration' thank you very much.

That is not, however, what I nor any other consumer appears to be getting.

What we are getting is cheap hardware produced under a 'pile 'em high, sell 'em cheap' mentality. What we are getting, is screwed.

Recent research by the Carnegie Mellon University, presented at the 5th USENIX Conference on File and Storage Technology in San Jose, looked at 100,000 hard drives and analyzed the mean time to failure. It revealed that despite the data spec sheets for the drives claiming a mean time to failure of between 1-1.5 million hours, or a mean annual failure rate of just 0.88% at the top end, the actual figures when tested were anywhere between 2% and 13%.

Unfortunately, the research was not vendor-specific so we cannot draw conclusions on which manufacturers really care more about our data than the financial bottom line, but we can conclude that they all lie about reliability rates.

At the same conference a Google study of 100,000 drives at its own server farm concluded that 56% of failures did not flag any SMART drive status issues before committing suicide, heavy data grinding did not appear to impact upon longevity, nor did a little overheating. Indeed, exposing yet another industry lie, the study revealed that drives which are cooled excessively tended to fail more often than those which were running a bit hot.

So, what can I do about this? What can you do about this? Well, for a start, you can use the power of crowds to influence your next hard drive purchase rather than the slick marketing of the manufacturer and vendor. Use Google to search for such strings as "<manufacturer> drives are useless" and "<manufacturer> drives are great" for example and base your buying decision upon what you discover. Oh, and implement a cast iron data backup strategy. I suspect you might need it whatever device you buy…

As Editorial Director and Managing Analyst with IT Security Thing I am putting more than two decades of consulting experience into providing opinionated insight regarding the security threat landscape for IT security professionals. As an Editorial Fellow with Dennis Publishing, I bring more than two decades of writing experience across the technology industry into publications such as Alphr, IT Pro and (in good old fashioned print) PC Pro. I also write for SC Magazine UK and Infosecurity, as well as The Times and Sunday Times newspapers. Along the way I have been honoured with a Technology Journalist of the Year award, and three Information Security Journalist of the Year awards. Most humbling, though, was the Enigma Award for 'lifetime contribution to IT security journalism' bestowed on me in 2011.

10 Years
Discussion Span
Last Post by WangVS

I'm curious as to which manufacturer of harddrives you commonly use. In the last 10 years, I've only had 2 drive failures. The first was a WD 8.4GB which failed within the first month but was able to have RMA and it still runs fine to this day. The other was a Seagate 200GB SATA, which had the oddest way of failing I've seen yet. Seagate's RMA tools claimed the drive worked perfectly, while 3 other SMART tools found various issues. I have yet to have any of my old SCSI drives fail on me, but like you've been saying, they're very old drives compared to these newer ones. (16GB being my largest scsi)
But even though I personally haven't had many failures, I been in the field long enough to see how often drives do fail and I'll agree that these newer drives just aren't reliable. It's like that 30 year toaster your mom had while growing up and now it has to be replaced every 2 years.


I've not had any issues the last few years either. All I've bought in the last 8 years or so have been Seagates (a couple 60GB from back then are still going strong) and WDs (still going strong as well, even after dropping them hard once).


"20 years ago" puts us virtually back into the pre-SCSI era. SMD disks, the sealed Winchesters that were a welcome improvement over 80- and 300-MB disk packs, were an utter nightmare once they got some age on them. One of my clients had about 30 SMD disks into the mid-1990s and at least one failed every month. The rebuilt Head/Disk Assemblies (HDAs) were getting so bad that the CE brought two for each repair since one might not work. I heard in the late 1990s that the last rebuilder on the planet was in France (oh, what confidence that inspires!), but more recently I heard that CDC in Minnesota had been rebuilding them more recently than that.

My experience with more modern disk drives over the last 10-15 years has generally been better, but I have used mostly commercial grade SCSI drives. Very recently Dell forced us into using SAS drives in their newer servers. I've seen more failures of the Dell-provided SCSI and SAS drives than any other types I've used in recent years even though the drive brands have been major ones.

We work with a certain kind of legacy mainframe, and many of those systems are using early SCSI drives that are now 10-15 years past their 5-year design lifetimes. The failure rates are increasing, but it's surprising how many of those older SCSI drives are still running.

I, too, have noticed that common items like appliances don't last like they used to, despite the often-touted advances in materals and engineering. Cheap seems to rule. In the late 1970s I moved into a house built in the 1920s. It still had the original water heater (copper lined) and the original oil furnace. The Hoover vacuum clearer that I grew up with in the 1950s served my family for at least 20 years, perhaps with a minor repair or two in its later years.

I think that while many people express the desire to have higher quality in many products, their purchasing belies that. They buy cheap, sending unmistakable signals to producers.

Have something to contribute to this discussion? Please be thoughtful, detailed and courteous, and be sure to adhere to our posting rules.