Slashdot Mirror
Who Makes the Best Hard Disk Drives?
Hamsterdan writes "Backblaze, the cloud backup company who open sourced their Storage Pod a few years ago, is now providing information on drive failure rates. They currently have over 27,000 consumer grade drives spinning in Backblaze storage pods. There are over 12,000 drives each from Seagate and Hitachi, and close to 3,000 from Western Digital (plus a too-small-for-statistical-reporting smattering of Toshiba and Samsung drives). One cool thing: Backblaze buys drives the way you and I do: they get the cheapest consumer-grade drives that will work. Their workload is almost one hundred percent write. Because they spread the incoming writes over several drives, their workload isn't overly performance intensive, either. Their results: Hitachi has the lowest overall failure rate (3.1% over three years). Western Digital has a slightly higher rate (5.2%), but the drives that fail tend to do so very early. Seagate drives fail much more often — 26.5% are dead by the three-year mark."
I remember when WD caviar drives were the most replaced component on systems I serviced. Seagate was the top contender with their SCSI 10krpm drives.
I built a new gaming rig the weekend after Black Friday and had to comparison shop all the consumer hard drives on the market (read: offered by Newegg). From the reviews, Hitachi is a relative unknown, Seagates tend to last just until their three year warranty is up, and Western Digital offers a five year warranty (and a price premium to match). I ended up grabbing the WD Black. Struck by how crap seek times are on 7200 RPM TB+ sized drives.
They wrote a check to Dice?
After all this research, Backblaze still pick the highest failing drive.
"What Drives Is Backblaze Buying Now?
We are focusing on 4TB drives for new pods. For these, our current favorite is the Seagate Desktop HDD.15 (ST4000DM000)"
So what was the point in this advert again?
I live in mortal terror of the Seagate Squeak. This is an intermittent sound that their 2 and 3 GB Barracudas sometimes start to make after a while, which sounds a little like a bird chirp. It's apparently caused by crap power management on the drive.
There's actually very little information out there on whether or not it is a definitive precursor of drive failure, or just something those drives start to do after a while. However, it's so unsettling that I've ended up pre-emptively replacing two drives in my home PC which developed it.
Enterprise grade disks? The cheapest disk is not always the cheapest disk in the long run. I can buy consumer disks for my disk servers, but when they fail I have to spend time replacing them and paying for them myself. When my enterprise grade disks fail, they're under warranty and are replaced "free".
For the past 11 years, I used nothing but Seagate drives in my builds for clients. Over those past 11 years, I built something like 20 systems a month (on average) with occasional large scale orders of 200. The number of failed Seagates I could count *on one hand* YMMV clearly, but I stand behind Seagates.
So rise up, all ye lost ones, as one, we'll claw the clouds.
What's the use case for any more than 50% write?
Backup. I have two raid 6 arrays. One is backup for the other. One is 100% write, the other isn't.
Archiving and backups springs to mind.
Back-ups. For example, at my company we have thousands of DLT tapes and fill about a dozen of them a week. Every Monday, they're moved offsite to a bank safe deposit box. Other than for testing, not a single time in the seven years I've worked here have we read a tape after it was written. We have 100% write. A friend works for Backblaze, and he just confirmed that they have basically the same situation. The vast majority of their users write data that they never read back.
Good thing the government forced Blackblaze to publish statistics, then? What fuckwit modded you up?
DATABASE WOW WOW
If you RTFA, they break down the failure rates by model (no pun intended). There's a pretty huge variation between models (or at least the Seagate models). That's also what I saw in the StorageReview reliability database back when people were actively updating it (unfortunately you have to add a drive to the database to get access to it, so it was never very popular). The same manufacturer can make a gem and a stinker of a model. e.g. the IBM 75GXP (aka Deathstar) drives had one of the highest failure rates in the database. The drive which replaced it (60GXP I think) had one of the lowest failure rates in the database.
So it's more nuanced than "Seagate stinks, Hitachi rules." (Hitachi is a subsidiary of WD now, operating separately only because that was a condition China placed on them before they'd OK the merger.)
If there's one thing you can credit Seagate for, it's consistency - since the 90's the (R) for refurb on their drives has been the kiss of death, guaranteeing another failure within 3 months of receiving the replacement. While it's great they have a clearly understandable domestic RMA team, they often send you a broken drive to replace your defective drive so you now have to pay to ship two drives back.
If you politely ask them to send you a new drive since they keep sending you bad drives, they'll politely tell you they can't guarantee you a healthy drive. Typically with our servers we're guaranteed a bad Seagate SAS 10k drive with a bank of 10 drives and we're pretty much at a 100% failure rate with RMA drives and many times the RMA drives they send us are broken. Seagate (R) drives should never be installed in a server or anything reliable... heck, I'd keep Seagate drives out of anything you want to remain reliable.
1 terabyte of storage that lasts 2 years is twice as useful as 1 terabyte of storage that lasts 1 year.
Always buy whatever drive is warranted for 5 years. I pay 50% more for this! It's worth every penny. My terabyte-years are the cheapest.
I have a 20TB LAN spread out over 3-4 computers (depending on the year). The only major crashes I've had on anything under 5 years old was, ironically, the 2 WD Cavier Green's I accidentally bought (meant to buy black; got a little slaphappy with the shopping cart one afternoon). They both died within 6 months.
The choice now is: Western Digital Cavier Black. The study posted in this article will not acknowledge this as they bought the cheapest drives possible. It may make business sense with redundancy, but i do not RAID. Too expensive. (Ironic?)
-Clio
Karma: Bad (mostly from not giving a fuck)
Blog: http://clintjcl.wordpress.com
Under communism you can compare all the manufacturers' failure rates for consumer hard drives... oh wait, communist countries don't manufacture consumer hard drives. If they did, they wouldn't care if they failed, but publishing failure information would be a criminal offense. WAKE UP.
Contribute to civilization: ari.aynrand.org/donate
The Deskstar wasn't nicknamed Deathstar for nothing, back in the day...
The IBM Deskstar series was superb; faster than most other drives, excellent size/price ratio, and very reliable too. The entire OEM business cheered it, and IBM was solidly trouncing the competition, until that fatal release of the Desktar 75GXP around 2001 where a bad firmware combined with a faulty factory in Hungary started to kill the drives prematurely (too much cutting edge technology).
I owned a lot of IBM Deskstar drives and they all performed really well and none of them died before they were obsolete, including one of the affected GXP's (I think it was made in the Philipines.)
After IBM sold their storage unit to Hitachi, I started buying Hitachi branded Deskstars. Again. Superb drives, never had one fail me. To me "Deskstar" is a stellar brand name.
All the Hard disk drive manufacturers have had similar problems with certain series dying prematurely; Seagate, Quantum, WD, Maxtor have all made bad batches of hard disks, but they haven't entered public mind as a internet meme, because they didn't have a cool nickname like "DeathStar".
My comment subject might seem a bit racially biassed, but it's because pretty much all drives are manufctured in "Asia" these days. So to answer your question directly, Asians make the best drives!
Comment removed based on user account deletion
I bought a Samsung (which is really a rebranded Seagate) to use in my HTPC and less than a year later, it died. I sent it back and got a replacement, but it was a huge pain to have to reinstall Mythbuntu and XBMC, get the two programs reconfigured and communicating again, as well as re-import all of my TV shows, movies, and music and fix all of the broken metadata. Since I suspected that the drive may have been running hot, I installed temperature monitoring software for the hard drive and had it record the temp once per minute. Less than a year later, that drive began to fail. I looked at the temperature logs while the drive still worked and it was pretty steady at about 40 degrees Celsius. I thought this may have been too hot, but when I looked up the specs for the drive, it was rated to operate at up to 60 degrees Celsius. So that's two Seagate drives that failed in less than a year each. Even though I may be able to get another replacement from Seagate for the failed drive, I wouldn't bother wasting my time reinstalling and reconfiguring the HTPC apps just to have the drive fail again, so I broke down and bought a WD Green since my other WD's have been solid over the past several years.
You do know there's a vast spectrum between communist and right libertarian, right?
Slashvertising? Oh welll... I, for one, am happy to have a source of failure rate data from an entity that uses a high volume of consumer level hard drives. I hope they continue to publish these numbers because HD failure rates are a moving target. The only thing better would be to have failure rates on specific model numbers. I'm now tweaking my slickdeals alert to filter for hitachi drives.
Seagate own Maxtor. Has since 2006.
And that seems to be about when the decline in their warranty started. I've heard horror stories about Maxtor, but the two I've owned still worked when I took them off line after many years of use. In fact, I just replaced my home grown firewall which had a 15+ year old 12 GB Maxtor in it.
They are not legally required to so why would they?
Because they can put it up on blog and make money out of ads? Because it works as a loss leader for their other paid services. Because it's cheap to publish these days and if you're going to do the research for your own curiosity why not publish?
Look at OkCupid. Publishing things like this
http://blog.okcupid.com/index....
Makes me like them a whole lot. Now I'm not really in the market for a US centric dating service but if I were I'd use them. Plus they could always write a book full of this sort of stuff.
Does everyone publish all their data? No of course not. Still the trend is that people increasingly do do it for the reasons mentioned.
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
Actually if you look they do call out model numbers and even talk about which ones they won't touch and I find this VERY interesting and informative! As it happens my 30TB of space happens to contain a mix of both "reliable" and "unreliable" drives according to their testing. I run a mix of sizes from 1.5TB to 3TB using unRAID and as drives fill up they get upgraded. I have a few 1.5s that they call out as being trash (ST31500341AS) and an EARS drive of that size that should probably go ASAP since they are well into their second if not third year of use. I actually happen to be running a parity check right now and once I got past the 1.5 drives speeds increased a great deal. Once past the 2TB models things got even better so the 3TB drives appear to be much better performers. Naturally they list the ST3000DM001 as having a 10% failure rate too so I'm not exactly doing handstands! The replacement drives are all that model and I've been playing with them in another system to try and come up with something better for my needs than unRAID and so far nothing has come out much better so into the array I guess they will go here shortly.
My hat's off to Backblaze for publishing this and letting consumers know who's got decent drives and putting feet to the fire those that don't!
Build it, Drive it, Improve it! Hybridz.org
You do know China is communist, right?
Not since they took back Hong Kong, or were you not paying attention?
I'd start with ConsumerReports. Not everyone likes them but if you want that kind of data for say TVs and automobiles they claim to have it....
Build it, Drive it, Improve it! Hybridz.org
And in the last 25 years I've replaced about 200 failed drives. The Maxtors failed at about 5x the rate of the Seagates. That shows you want ancidotal evidence is worth.
To know actual reliability, you need stats on the level of Google or Amazon, that can tally failures by the 10's of thousands.
I searched on 4+ comments, and didn't see anything, so here is Google's study> (they go through a lot of drives)
If China is "communist", then I'm the pope.
China is a dictatorship wrapping itself in the trappings of communism to keep the people from executing the leaders.