What Hard Drive Should I Buy?

January 21st, 2014

blog-comparison-brand

My last two blog posts were about expected drive lifetimes and drive reliability. These posts were an outgrowth of the careful work that we’ve done at Backblaze to find the most cost-effective disk drives. Running a truly unlimited online backup service for only $5 per month means our cloud storage needs to be very efficient and we need to quickly figure out which drives work.

Because Backblaze has a history of openness, many readers expected more details in my previous posts. They asked what drive models work best and which last the longest. Given our experience with over 25,000 drives, they asked which ones are good enough that we would buy them again. In this post, I’ll answer those questions.

Drive Population

At the end of 2013, we had 27,134 consumer-grade drives spinning in Backblaze Storage Pods. The breakdown by brand looks like this:

Hard Drives by Manufacturer Used by Backblaze
Brand Number
of Drives
Terabytes Average
Age in Years
Seagate 12,765 39,576 1.4
Hitachi 12,956 36,078 2.0
Western Digital 2,838 2,581 2.5
Toshiba 58 174 0.7
Samsung 18 18 3.7

As you can see, they are mostly Seagate and Hitachi drives, with a good number of Western Digital thrown in. We don’t have enough Toshiba or Samsung drives for good statistical results.

Why do we have the drives we have? Basically, we buy the least expensive drives that will work. When a new drive comes on the market that looks like it would work, and the price is good, we test a pod full and see how they perform. The new drives go through initial setup tests, a stress test, and then a couple weeks in production. (A couple of weeks is enough to fill the pod with data.) If things still look good, that drive goes on the buy list. When the price is right, we buy it.

We are willing to spend a little bit more on drives that are reliable, because it costs money to replace a drive. We are not willing to spend a lot more, though.

Excluded Drives

Some drives just don’t work in the Backblaze environment. We have not included them in this study. It wouldn’t be fair to call a drive “bad” if it’s just not suited for the environment it’s put into.

We have some of these drives running in storage pods, but are in the process of replacing them because they aren’t reliable enough. When one drive goes bad, it takes a lot of work to get the RAID back on-line if the whole RAID is made up of unreliable drives. It’s just not worth the trouble.

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.

Failure Rates

We measure drive reliability by looking at the annual failure rate, which is the average number of failures you can expect running one drive for a year. A failure is when we have to replace a drive in a pod.

blog-fail-drives-manufacture

This chart has some more details that don’t show up in the pretty chart, including the number of drives of each model that we have, and how old the drives are:

Number of Hard Drives by Model at Backblaze
Model Size Number
of Drives
Average
Age in
Years
Annual
Failure
Rate
Seagate Desktop HDD.15
(ST4000DM000)
4.0TB 5199 0.3 3.8%
Hitachi GST Deskstar 7K2000
(HDS722020ALA330)
2.0TB 4716 2.9 1.1%
Hitachi GST Deskstar 5K3000
(HDS5C3030ALA630)
3.0TB 4592 1.7 0.9%
Seagate Barracuda
(ST3000DM001)
3.0TB 4252 1.4 9.8%
Hitachi Deskstar 5K4000
(HDS5C4040ALE630)
4.0TB 2587 0.8 1.5%
Seagate Barracuda LP
(ST31500541AS)
1.5TB 1929 3.8 9.9%
Hitachi Deskstar 7K3000
(HDS723030ALA640)
3.0TB 1027 2.1 0.9%
Seagate Barracuda 7200
(ST31500341AS)
1.5TB 539 3.8 25.4%
Western Digital Green
(WD10EADS)
1.0TB 474 4.4 3.6%
Western Digital Red
(WD30EFRX)
3.0TB 346 0.5 3.2%
Seagate Barracuda XT
(ST33000651AS)
3.0TB 293 2.0 7.3%
Seagate Barracuda LP
(ST32000542AS)
2.0TB 288 2.0 7.2%
Seagate Barracuda XT
(ST4000DX000)
4.0TB 179 0.7 n/a
Western Digital Green
(WD10EACS)
1.0TB 84 5.0 n/a
Seagate Barracuda Green
(ST1500DL003)
1.5TB 51 0.8 120.0%


The following sections focus on different aspects of these results.

1.5TB Seagate Drives

The Backblaze team has been happy with Seagate Barracuda LP 1.5TB drives. We’ve been running them for a long time — their average age is pushing 4 years. Their overall failure rate isn’t great, but it’s not terrible either.

The non-LP 7200 RPM drives have been consistently unreliable. Their failure rate is high, especially as they’re getting older.

1.5 TB Seagate Drives Used by Backblaze
Model Size Number
of Drives
Average
Age in
Years
Annual
Failure
Rate
Seagate Barracuda LP
(ST31500541AS)
1.5TB 1929 3.8 9.9%
Seagate Barracuda 7200
(ST31500341AS)
1.5TB 539 3.8 25.4%
Seagate Barracuda Green
(ST1500DL003)
1.5TB 51 0.8 120.0%

The Seagate Barracuda Green 1.5TB drive, though, has not been doing well. We got them from Seagate as warranty replacements for the older drives, and these new drives are dropping like flies. Their average age shows 0.8 years, but since these are warranty replacements, we believe that they are refurbished drives that were returned by other customers and erased, so they already had some usage when we got them.

Bigger Seagate Drives

The bigger Seagate drives have continued the tradition of the 1.5Tb drives: they’re solid workhorses, but there is a constant attrition as they wear out.

2.0 to 4.0 TB Seagate Drives Used by Backblaze
Model Size Number
of Drives
Average
Age in
Years
Annual
Failure
Rate
Seagate Desktop HDD.15
(ST4000DM000)
4.0TB 5199 0.3 3.8%
Seagate Barracuda
(ST3000DM001)
3.0TB 4252 1.4 9.8%
Seagate Barracuda XT
(ST33000651AS)
3.0TB 293 2.0 7.3%
Seagate Barracuda LP
(ST32000542AS)
2.0TB 288 2.0 7.2%
Seagate Barracuda XT
(ST4000DX000)
4.0TB 179 0.7 n/a

The good pricing on Seagate drives along with the consistent, but not great, performance is why we have a lot of them.

Hitachi Drives

If the price were right, we would be buying nothing but Hitachi drives. They have been rock solid, and have had a remarkably low failure rate.

Hitachi Drives Used by Backblaze
Model Size Number
of Drives
Average
Age in
Years
Annual
Failure
Rate
Hitachi GST Deskstar 7K2000
(HDS722020ALA330)
2.0TB 4716 2.9 1.1%
Hitachi GST Deskstar 5K3000
(HDS5C3030ALA630)
3.0TB 4592 1.7 0.9%
Hitachi Deskstar 5K4000
(HDS5C4040ALE630)
4.0TB 2587 0.8 1.5%
Hitachi Deskstar 7K3000
(HDS723030ALA640)
3.0TB 1027 2.1 0.9%

Western Digital Drives

Back at the beginning of Backblaze, we bought Western Digital 1.0TB drives, and that was a really good choice. Even after over 4 years of use, the ones we still have are going strong.

We wish we had more of the Western Digital Red 3TB drives (WD30EFRX). They’ve also been really good, but they came after we already had a bunch of the Seagate 3TB drives, and when they came out their price was higher.

Western Digital Drives Used by Backblaze
Model Size Number
of Drives
Average
Age in
Years
Annual
Failure
Rate
Western Digital Green
(WD10EADS)
1.0TB 474 4.4 3.6%
Western Digital Red
(WD30EFRX)
3.0TB 346 0.5 3.2%
Western Digital Green
(WD10EACS)
1.0TB 84 5.0 n/a

What About Drives That Don’t Fail Completely?

Another issue when running a big data center is how much personal attention each drive needs. When a drive has a problem, but doesn’t fail completely, it still creates work. Sometimes automated recovery can fix this, but sometimes a RAID array needs that personal touch to get it running again.

Each storage pod runs a number of RAID arrays. Each array stores data reliably by spreading data across many drives. If one drive fails, the data can still be obtained from the others. Sometimes, a drive may “pop out” of a RAID array but still seem good, so after checking that its data is intact and it’s working, it gets put back in the RAID to continue operation. Other times a drive may stop responding completely and look like it’s gone, but it can be reset and continue running.

Measuring the time spent in a “trouble” state like this is a measure of how much work a drive creates. Once again, Hitachi wins. Hitachi drives get “four nines” of untroubled operation time, while the other brands just get “two nines”.

Untroubled Operation of Drives by Manufacturer used at Backblaze
Brand Active Trouble Number of Drives
Seagate 99.72 0.28% 12459
Western Digital 99.83 0.17% 933
Hitachi 99.99 0.01% 12956

Drive Lifetime by Brand

The chart below shows the cumulative survival rate for each brand. Month by month, how many of the drives are still alive?

blog-36-month-drive-survival-rate

Hitachi does really well. There is an initial die-off of Western Digital drives, and then they are nice and stable. The Seagate drives start strong, but die off at a consistently higher rate, with a burst of deaths near the 20-month mark.

Having said that, you’ll notice that even after 3 years, by far most of the drives are still operating.

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). We’ll have to keep an eye on them, though. Historically, Seagate drives have performed well at first, and then had higher failure rates later.

Our other favorite is the Western Digital 3TB Red (WD30EFRX).

We still have to buy smaller drives as replacements for older pods where drives fail. The drives we absolutely won’t buy are Western Digital 3TB Green drives and Seagate 2TB LP drives.

A year and a half ago, Western Digital acquired the Hitachi disk drive business. Will Hitachi drives continue their excellent performance? Will Western Digital bring some of the Hitachi reliability into their consumer-grade drives?


Correction: Hitachi’s 2.5″ hard drive business went to Western Digital, while the 3.5″ hard drive business went to Toshiba.

At Backblaze, we will continue to monitor and share the performance of a wide variety of disk drive models. What has your experience been?

    UPDATE: The data has been updated to include over 34,000 drives. Please click here to read the September 2014 post.

 

Brian Beach

Brian Beach

Brian has been writing software for three decades at HP Labs, Silicon Graphics, Netscape, TiVo, and now Backblaze. His passion is building things that make life better, like the TiVo DVR and Backblaze Online Backup.
  • icabodmudd

    why run consumer drives in an industrial heavy server environment, how odd.
    I guess you never plotted cost of ownership by CLASS of SYSTEMS?
    nor those silly green drives, my guess is you never heard of NAS storage?
    and no control group (science 101) say have a top brand and quality NAS server? with matching quality drives as a control group and actually learn something,?

    Your report is like running Yugo cars in INDY 500.
    silly to a fault.

    those consumer drives are designed for 24/7 usages. if you read the data sheets it states that
    as does calling the maker of drive and telling them your APPLICATION environ, and needs.
    Just a $50 drive. (less in brick size orders)
    One other factor is usage, some servers , sit on one set of partitions, and the users bet it to death
    no consumer drive was built to do that, so….. you left out all those facts.
    heat, (during the whole life).

    vibration/shock
    and voltage spikes and surges.
    and usage factor, by platter and sectors. load factors.
    and more…

    • Kai Harrekilde

      Backblaze chose consumer drives for three reasons: Cost, cost and cost. Just because that the drives are “consumer grade” it doesn’t mean that they’re bad. This is the same that drove out MIL-SPEC components, in favour of commercial spec’ed components.

      And the information that they put out here was never intended to be a controlled scientific experiment. It’s their bloody production data. If you are unhappy with the unscientific approach, please look somewhere else. I wish you good luck in finding a good statistical dataset from a scientifically controlled experiment.

      • MetalManForChrist

        Sounds like a pretty good experiment to me. Any drive that can handle harsher-than-intended-use on a statistical basis better than another drive is obviously a better drive. Thanks for the objective data Backblaze. Sounds like icabomudd is a Seagate rep.

      • Brian Emenaker

        Google published an extremely extensive hard drive failure report a few years back, and in it, they showed there was virtually no difference in failure rate between consumer and enterprise level hard drives. This is why Google also buys consumer drives.

        • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

          If manufacturers keep tooling and variations as small as possible between models, that reduces overhead. As technology advances, metallurgy improves, finishing and machining technology improves, we should see a merging, or a closing gap, between product lines. It just makes sense for the manufacturer to have a single set of machines to produce everything they sell.

    • LordBreadCrustKitty

      Agree with Kai, this is not a test or an experiment it’s there Production statistics for failure rates, and it the best data you are going to find for “Real World” scenarios. “why run consumer drives in an industrial heavy server environment” you ask, did you even look at the business model for this company? It’s a cloud backup service. That data is just going to sit there idle 99% of the time, you don’t need anything better then so called “consumer” grade storage for that. Even EMC uses cheap disk in Data Domain, why, because it’s freakin backup data, not a SQL server with tens of thousands of IOP’s. Get your use case information straight before you criticize.

      • John

        Hang on.

        He is right, this is not based on any science since there is no control group (test) and no falsifiable hypothesis.

        All this is doing is reporting on actual data, with no explanation of it. That is like driving some cars in some unknown / un reported environment and reporting some have a different failure rate than others.

        The other irony is that the test results show HDS723030ALA640 (HGST Deskstar 7K3000, 3TB) as the best, yet the author does not recommend it, he recommends the WD Red 3TB. WTF? So the author recommends another drive over the one that he reports lasts the longest and does not say why.

        Am sorry, but I can not trust this report. This experiment will not pass peer review, and is therefore fundamentally unrealisable. That does not imply it is wrong or the results are wrong, it just means that the data can not be trusted until proper scientific controls and reviews can be shown to apply.

        • NET Sentra Cyberindo

          The author has said “Basically, we buy the least expensive drives that will work.” He also said that “If the price were right, we would be buying nothing but Hitachi drives.”

          Please read the chapter : “What Drives Is Backblaze Buying Now?”.
          Western Digital is his favorite drive to BUY because of the reason above. However he admit Hitachi is the most reliable drive in his environment.

          • John

            You did not read my post, did you?
            This suckle is not based on science, and the author contradicts himself.

            Therefore, this article is rubbish; its conclusions can not be accepted.

          • Bad

            It’s not meant to be based on “science”, just the reality of running these drives in THEIR environment.
            Or doesn’t reality count?

          • John

            Experiments do not count if they are not based on science.
            You answered your own question when you confirmed that the results are not suppose to be science.

            Thanks for playing.

            Pathetic.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            It is not science. It is statistics. Any high-school student, with a rudimentary knowledge of Excel, and good attention to detail, could reproduce the numbers, with access to the data.

          • John

            Even if only statistics, it’s still not scientific. Besides, the statistical analysis does not draw on a realistic sample size or population. Purely on the numbers it would fail to convince any skeptic.
            And the statistics (the data or results) as you claim was a result from an experiment. Or are you claiming that they made up the numbers without any testing?

            Therefore, the results are purely subjective and should not be taken as objective fact, but as biased opinion.

            Want another go?

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            I said. “It is not science. It is statistics.”

            You responded: “Even if only statistics, it’s still not scientific.”

            Another go at what? I don’t have time for you, find something useful to do, John…

          • John

            Seems like you agree then, this statistics is purely subjective and not objective and therefore can be rejected as a claim, not as fact (it’s does not apply generally).

            Thanks for playing.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            Thanks for playing what? Oh, wait that is some little quip you say to people when you cannot actually defend yourself, ahhh, amusing. Backblaze published some very useful data, for those of us who can figure out
            how to make use of it; that would be most of us probably, well except you apparently. I am curious why you are so defensive and petty. Are you a cop, you have the confrontational attitude of one… Did you have bad luck with investments? Are you a research assistant who was run out for falsifying data? I am curious. I like psychology, and people who are as defensive as you are have a story to tell, some delusions, perhaps social issues. Maybe you work for one of these lax hard drive companies, and that is
            why you are so defensive? Are you a shill for Western Digital, or Seagate, John, is
            that it? Why don’t you setup your own study and show us all how it is done?

            Too bad you cannot figure out how to put an avatar on Disqus, or understand high-school statistics. Do you understand what scientific principle is either; no doubt you feel you do. Have you ever heard of the Dunning Kruger Effect in Psychology. Go and find something useful to do, John. ;)

          • John

            My dear, you are projecting your ignorance with great confidence. Pity it does not beget any knowledge.

            Oh, and your pathetic attempt at an ad hominem cements your rejecting of science and the scientific method. Imagine if science works the way you claim; with one experiment based with vendor biased and poor statistics. Oh dear, take that panadol dear it was not based on clinical trials.

            Pathetic.

          • John

            You are still playing and with esteemed action but no excitement.
            God help us if science works the way you think and wants it to work.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            It is not science, it is a statistical report. Why do you keep calling this science?

          • John

            It is indeed you that think a flawed statistical report is science. I’m merely telling you that a statistical report is not science and nor is a flawed statistical report.

            You should not use this “statistical report’ to make informed buying decisions or to tell and convince me to.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            I would be very interested if you have another report that shows relevant results. Also the customer feedback about failures seems to loosely follow the data provided in this report by Backblaze.

            And even one statistical study, especially one that deals with so many actual physical drives, is definitely relevant to people who are making buying decisions. Unless you have some other references that offer alternative findings? And, obviously not anything coming from any of the vendors of these drives, as they could not be trusted to be unbiased.

          • John

            Ah, you are trying to shift the burden here, to make me show a report that shows your ‘statistically flawed pseudoscience’ report false. How absurd.

            But you do not stop there. Your second sentence consist of a wild claim. A claim that you completely neglected to provide evidence or your source for. How pathetic.

            In your third sentence you are yet again trying to masquerade this ‘statistically flawed pseudoscience’ report as science by stating it is relevant to people that buy stuff based on evidence that it works. But you admitted before that this stats is not science. WOW, what a dichotomy you find yourself in, Harvey Two-Face.

            And in your 4th sentence you are back at trying to shift the burden (as per your opening sentence). Kieron, you are having a circular argument by begging the question.

            And you last sentence is grounded in your emotional opinion again, trying to masquerade it as fact.

            Kieron, people that accept this article as fact are delusional. Do you suffer from amnesia, or are your cognitive functions crowed by your delusion? Either way, you do project your ignorance with confidence. Pity, it does not beget any knowledge.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            I have no, as you put it, ” ‘statistically flawed pseudoscience’ report”. I have never attempted ANY such report on hard drive reliability. I neither have the time, nor resources and access to the amount of hardware required to show a meaningful result. This report is (scroll up, and note the browser address) offered by Backblaze, to which I have no affiliation whatsoever. How am I shifting the burden? I just browsed here looking for something on drive reliability. You are the one making claims without backing them up, not I.

            This is the only large-scale published data pertaining to drive reliability that I could find online. However, you claim it is false! Have you considered, if the data is false, this company is opening themselves up to litigation? Therefore, anyone reading these comments would assume you have alternate data to back your claims. Some statement by you, at least making useful claim Backblaze’s report is false and incorrect? The bottom line is you either have something worthwhile, a link perhaps (we hope), something tangible, or nothing (which is what I suspect).

            No matter how much effort and time you put into using amusingly unnecessary verbiage, it does not change that bottom line; what references YOU have for saying their report is false? So, do you have anything? Or, are you just trolling (my gut feeling)?

          • John

            Yet you rely on this ” statistically flawed pseudoscience’ report” as your basis for your argument. Pffff

            You claim: “This is the only large-scale published data pertaining to drive
            reliability that I could find online. However, you claim it is false! ”

            I did not claim it’s false. What is quite clear is that it’s not good science and the sample population is not wide enough. Yet, you claim this report is true, but fail to justify why. The burden of evidence is on you, not me.

            If you use this report to make buying decisions then you are a fool.

            You did not even try and address you dichotomy you find yourself in. Indeed, that is your trolling.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            Well, it is good enough for me, and many other people. I trust the numbers from Backblaze, and I seriously doubt they are lying. The numbers of drives they buy and replace, pretty much assures they know what they are talking about. If you can point the readers of this article to something better, then by all means please do so!

          • John

            Yes, this “statistically flawed pseudoscience” (aka this article) is good enough for you. But it’s not how science works. Imagine if science works this way; one vendor does a test and the rest of humanity needs to adhere to it. How absurd.

            And no, I do not need to give you anything. And your argument that this report is better than no report is fundamentally flawed. You fail to understand the biased built into this report. if you indeed want a decent report, search for it, or learn about science and do your own report, send it for peer review and then use it if it’s not rejected.

            But hey, people like you buy products after being exposed to it’s advertisement. And your excuse for not buying the competitor product, or to reseach the products is flawed when you state you have not seen an add for the compensator product.

            Dude, wake up.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            What is this obsession you have with science? Also, you do know what Backblaze is and what they do, right? You do understand they do not sell hard drives, (“…one vendor does a test …”) because it seems you are confused. Also what advertisement (“But hey, people like you buy products after being exposed to it’s advertisement”)? Backblaze states that one drive is better, then goes right on to state they buy the cheap drives anyway. So, where is this ad and who is this “vendor” you are talking about?

            You are some sort of shill for Seagate or Western Digital perhaps? It does not matter, as whoever you are is irrelevant anyhow. Did you even read the article carefully? Likely you skimmed it, formed a negative conclusion about the content, assuming no one else had read it either. So much for science and unbiased opinions. *laughing* Every one of your replies makes me laugh because I have met your type before. You communicate using good language skills, but sadly say nothing anyone can actually use. If you worked for me (would never happen), you would have been gone long ago because of your attitude and inability to mesh with others efficiently. You are quite entertaining though, which is why I continue to bait you…

          • John

            Yep, all you can offer now is an ad hominem grounded in your personal attack. Here is s revealing quote:
            ‘When the debate is lost, slander becomes the tool of the loser’. That was from Socrates, ironically summing up your agley behavior.

            Why do you keep on defending this pseudoscience report (this article) when you already admitted it’s flawed (not based on good science)?

            And no, there is no obsession with science. You do not even know what science is. If you knew, then you would understand that there can’t be an obsession. You are merely projecting your ignorance with confidence. However, it doe snot beget any knowledge. I’m merely opening you eyes to understand how to distinguish fact from opinion and belief. This article (this statistically flawed pseudoscience report) is opinion and not fact. Yet you admit it, but still masquarade the ‘opinion’ as ‘fact’ by grounding it in your belief. That is a serious dichotomy you find yourself in. I think your reason filter is broken because it let’s all the crap get thru. You adopted someone’s reason filter and did not realize it is broken. That can be changed, but should give you pause if you are serious about the foundations of your beliefs. It would be hard work for you, extremely hard for you, but it can be done. You need to consider closing down your museum of mistakes and wake up to reality.

            Here is a lesson for you:
            Belief does not change reality, it’s the other way around. You need to change your beliefs to fit reality.

            With your last sentence you merely admit to being a troll. I leave you with another quote:
            “Never argue with an idiot. They will only bring you down to their level and beat you with experience.” – George Carlin

            I need to stop arguing with you, immediately. I refuse
            to go down to your level and to be beaten by your troll experience.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            ‘When the debate is lost, slander becomes the tool of the loser’. – Socrates

            Oh, now I am REALLY laughing. All you have done, in nearly every paragraph of every post, is attack and slander. Your failed attempts at belittlement and utter lack of common sense concerning the value of the data on this page, no doubt have provided many people with considerable amusement.

            I am not “a troll” but I will admit to trolling people when it amuses me. But then, I do that in person far more than online. Also, I use my real name and photo, unlike most online; hint hint. :P

          • John

            Am not arguing with you. I am merely refusing to be beaten by your troll experience.

          • AnOilMan

            Actually you’re the troll. Undeniable.

          • John

            Troll.

          • John

            You are indeed a troll.

          • John

            Trolling.

          • John

            How do you know that?

          • John

            How do you know that?

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            What would constitute a wide enough data set for you, every desktop in use? Do you honestly think scientists gather data from that many sources? Maybe a dozen, perhaps a hundred in very rare occasions. This gathered data is from thousands of drives. Where do you propose we go to find more that that? And, who is going to pay for, and authenticate the results? So, we have what we have. Tells me far more than the outside of the box, or the 20 word blurb that is given on hard drive sites.

            I still think you are a shill from Seagate…

          • John

            Straw man.
            Maybe you should learn what constitutes a good experiment and then come back when we can deal in fact.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            LOL

          • John

            troll.

          • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

            Or the way you do? You actually show by your replies that you have almost no reading comprehension, and you have no idea what I mean, or the article means. I believe you are a shill for one of these drive manufacturers who comes out looking bad.

          • John

            Nice ad hominem from you.
            As they say: when the debate is lost, libel becomes the tool of the loser.
            I refuse to accept your defamation as your response to my facts.
            Wake up to reality, you need to change your beliefs to fit reality, not the other way around.

          • John

            You believe?
            To believe means that you are holding cognitive content as true in spite of evidence or not. You are merely taking the focus away from your content and placing it on you. You are saying that you are not sure of yourself, you have not checked the evidence for your claim.
            Therefore, you are still playing with esteemed action but offers no facts, only drivel.

  • Alex Chen

    Do you change default drive parameters in any way? For example, I know
    WD green drives have 5 seconds default sleep timeout. This would be
    very bad for data center use, I think.

    • http://www.bulkerz.com Rodrigo Morera de Souza

      Search in Google and Download: “wdidle3.exe”, It Works.
      :D

      • Alex Chen

        I know. I asked if they did it or not.

        • den

          assume they did, otherwise their IT would be pretty hopeless. *grin*

  • Manish Kumar

    I have a simple Question for you .. I just need to buy a 1 TB external
    drive in around $75. Which brand should I buy. Plz reply ASAP

    • Kartman13

      Buy HGST touro mobile 1.5 TB. Its just 5300rs in SnapDeal and you get 1500 GB which is very useful as 1 TB hdd is doesn’t give you full 1 TB usage. Its a rock solid hdd with 3 year warranty.

      • Mohammed Anees

        Problem is, a large number of Hitachi and Toshiba hard disks are being sold with refurbished/defective drives from other countries at a very low price. They come with a 1 year warranty (whose online warranty registration almost always fails)instead of the 3 year warranty that Toshiba/Hitachi provide. When you buy online, like on Snapdeal, you can’t be sure what the seller will ship because they have the same name and branding as the regular ones. Best to buy it at a local store and check the warranty printed on the box instead of buying online. Otherwise buy a WD.
        Stay away from the really cheap versions Toshiba and hitachi drives. They WILL fail. (Hitachi and Toshiba drives themselves are great when you find the genuine ones)

  • new guy

    what kind of drive should I get for gaming with. BF 4, farcry 4, leage of leagends.

    • wreid23

      an SSD preferably an EVO series from samsung

      • Kruger

        ^^ what wreid23 said. Plonk a Samsung 840 Evo Pro SSD into your gaming rig and you’ll noticed an overall improvement in PC speed & multitasking functionality. The Samsung is around 15-20% cheaper than a OCZ Vertex IV and performs just as well for gaming purposes. Of course… if you’re game is (for example) GPU intensive (like BF4 can be) don’t expect it to turn an entry level graphics card into a high spec one. On my high spec rig I only felt that an improvement on overall Windows multitasking and game(&stage) loading times. Other than that it was on par with the raptor 10k rpm drive I had before that.
        Sometimes simply tweaking your existing video card drivers (profiles) and internal game settings, can gain you an additional 10-20 frames/sec. If you’re trying to squeeze out more than that, it may be time for an upgrade.

    • ToothyGrinn

      A rectangle one

  • Nitesh

    Thanks for the info!

  • Aviad

    How is the Seagate Barracuda 2TB?

  • Waseem Yousaf

    Samsung has introduced the latest T1 Miniature external SSD. The Size of SSD is not more than a business card.

  • Chris

    It is a sad shame this article does not show the failure rate of brand new drives that fail upon opening the packaging and connecting them to your computer system. I have opened three batches of Western Digital 3TB and 6TB Green hard drives and each batch there has been a minimum of 1 or 2 drives that was found to be manufactured defective and failed to initialise which had to be returned to Western Digital or their retailer only for them not to supply a new drive as you would expect but to supply a reengineered and recretified drive with someone elses data on these drives which are immediately registered to the sender? I now have 20TB of someone elses data supplied by WD and registered to me?

    • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

      Yes, I have experienced that in the past, and it seems a little dishonest on the part of Western Digital… If the drive is less than 30 days old, they should honour the warranty with a new drive, not a refurbished one. It almost seems like a scam to stick a few dead drives in a large box, gambling that you will not do anything at all, or forget, and if you do, just send you a refurbished unit. Yeah, that leaves a negative feeling in the mind of the customer.

  • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

    This is an informative and extremely useful article. Probably one of the THE most useful articles on the Internet for system admins. To have access to this level of statistical data, that actually names companies and models, is an absolutely invaluable dataset. Most companies would be charging a lot of money for this data.

    So, to the whiners; LOL! You are just displaying: A) your personality flaws for the world to see, and B) your ignorance! It is quite entertaining however. So, carry on. *grins*

  • Milos Ivanovic

    First of all I want to thank you for taking the time to actually collate all this data and make it available for the general public at no additional cost. Few companies would bother going that far.

    It’s a long shot, and I would say unheard of in the industry, but it would be incredibly interesting to be able to get statistics at this magnitude on how failure rates correlate with the place of manufacture of the drives. For example, Seagate has “site codes” written clearly on the drive labels which specify the factory ID where their drives were manufactured. It would certainly mean something if it’s found that, for example, there is only a 1% annual failure rate for Seagate drives manufactured in Thailand (TK), with the remaining 15% in China (SU, WU).

    What do you think? If it’s not a total burden to add this additional piece of information, I think you may come to terms with some surprising conclusions.

  • alafrosty

    I just took a look at the the raw data, and I think you have a problem with the ST4000DM000 drives as well. I could tell more clearly with more recent data, but looking at the drives taken offline (not just the ones that failed), there are a *lot* of them that aren’t making it past 5000 hours.

  • Ronald Perez

    I fully agree with this. Without the formal statistical analysis found in the excellent article I had the same feeling about hard drive reliability with brands and models. I manage a 300 PCs call center with a 60 something servers, and my experience match this results. Hitachi is the most reliable and Seagate is the less, when you see it as a long term experience. I stop using Samsung long ago because some models had Sudden death syndrome, so there may be a tie with the most failures. Western Digital for me is second best in the call center and I buy them because hitachi is somehow difficult to buy in my country. I really appreciate your data, the time you invested in your research, but most of all your courage to publish the facts.

  • Robb C

    Toshiba drives have been and continue to be rock-solid performers for us.

    • http://www.facebook.com/Kieseyhow Kieron Seymour-Howell

      That is what I ended up going with. I use Toshiba for the mechanical drives, and Intel for the SSDs.

  • johnkristian

    You say you run all these bingo consumer drives in a raid? What kind of raid? SW raid or HW (controller?)? RaidZ?