Hard Drive Stats for Q3 2017

By | October 26th, 2017

Q3 2017 Hard Drive Stats

In Q3 2017, Backblaze introduced both 10 TB and 12 TB hard drives into our data centers, we continued to retire 3 TB and 4 TB hard drives to increase storage density, and we added over 59 petabytes of data storage to bring our total storage capacity to 400 petabytes.

In this update, we’ll review the Q3 2017 and lifetime hard drive failure rates for all our drive models in use at the end of Q3. We’ll also check in on our 8 TB enterprise versus consumer hard drive comparison, and look at the storage density changes in our data centers over the past couple of years. Along the way, we’ll share our observations and insights, and as always, you can download the hard drive statistics data we use to create these reports.

Q3 2017 Hard Drive Failure Rates

Since our Q2 2017 report, we added 9,599 new hard drives and retired 6,221 hard drives, for a net add of 3,378 drives and a total of 86,529. These numbers are for those hard drives of which we have 45 or more drives — with one exception that we’ll get to in a minute.

Let’s look at the Q3 statistics that include our first look at the 10 TB and 12 TB hard drives we added in Q3. The chart below is for activity that occurred just in Q3 2017.

Hard Drive Failure Rates for Q3 2017

Observations

  1. The hard drive failure rate for the quarter was 1.84%, our lowest quarterly rate ever. There are several factors that contribute to this, but one that stands out is the average age of the hard drives in use. Only the 4 TB HGST drives (model: HDS5C4040ALE630) have an average age over 4 years — 51.3 months to be precise. The average age of all the other drive models is less than 4 years, with nearly 80% of all of the drives being less than 3 years old.
  2. The 10- and 12 TB drive models are new. With a combined 13,000 drive days in operation, they’ve had zero failures. While all of these drives passed through formatting and load testing without incident, it is a little too early to reach any conclusions.

Testing Drives

Normally, we list only those drive models where we have 45 drives or more, as it formerly took 45 drives (currently 60), to fill a Storage Pod. We consider a Storage Pod as a base unit for drive testing. Yet, we listed the 12 TB drives even though we only have 20 of them in operation. What gives? It’s the first step in testing drives.

A Backblaze Vault consists of 20 Storage Pods logically grouped together. Twenty 12 TB drives are deployed in the same drive position in each of the 20 Storage Pods and grouped together into a storage unit we call a “tome.” An incoming file is stored in one tome, and is spread out across the 20 storage pods in the tome for reliability and availability. The remaining 59 tomes, in this case, use 8 TB drives. This allows us to see the performance and reliability of a 12 TB hard drive model in an operational environment without having to buy 1,200 of them to start.

Breaking news: Our first Backblaze Vault filled with 1,200 Seagate 12 TB hard drives (model: ST12000NM007) went into production on October 20th.

Storage Density Continues to Increase

As noted earlier, we retired 6,221 hard drives in Q3: all 3- or 4 TB hard drives. The retired drives have been replaced by 8-, 10-, and 12 TB drive models. This dramatic increase in storage density added 59 petabytes of storage in Q3. The following chart shows that change since the beginning of 2016.

Hard Drive Count by Drive Size

You clearly can see the retirement of the 2 TB and 3 TB drives, each being replaced predominantly by 8 TB drives. You also can see the beginning of the retirement curve for the 4 TB drives that will be replaced most likely by 12 TB drives over the coming months. A subset of the 4 TB drives, about 10,000 or so which were installed in the past year or so, will most likely stay in service for at least the next couple of years.

Lifetime Hard Drive Stats

The table below shows the failure rates for the hard drive models we had in service as of September 30, 2017. This is over the period beginning in April 2013 and ending September 30, 2017. If you are interested in the hard drive failure rates for all the hard drives we’ve used over the years, please refer to our 2016 hard drive review.

Cumulative Hard Drive Failure Rates

Note 1: The “+ / – Change” column reflects the change in the annualized failure rate from the previous quarter. Down is good, up is bad.
Note 2: You can download the data on this chart and the data from the “Hard Drive Failure Rates for Q3 2017” chart shown earlier in this review. The downloaded ZIP file contains one MSExcel spreadsheet.

The annualized failure rate for all of the drive models listed above is 2.07%; this is the higher than the 1.97% for the previous quarter. The primary driver behind this was the retirement of all of the HGST 3 TB drives (model: HDS5C3030ALA630) in Q3. Those drives had over 6 million drive days and an annualized failure rate of 0.82% — well below the average for the entire set of drives. Those drives now are gone and no longer part of the results.

Consumer Versus Enterprise Drives

The comparison of the consumer and enterprise Seagate 8 TB drives continues. Both of the drive models, Enterprise: ST8000NM0055 and Consumer: ST8000DM002, saw their annualized failure rates decrease from the previous quarter. In the case of the enterprise drives, this occurred even though we added 8,350 new drives in Q3. This brings the total number of Seagate 8 TB enterprise drives to 14,404, which have accumulated nearly 1.4 million drive days.

A comparison of the two drive models shows the annualized failure rates being very similar:

  • 8 TB Consumer Drives: 1.1% Annualized Failure Rate
  • 8 TB Enterprise Drives: 1.2% Annualized Failure Rate

Given that the failure rates for the two drive models appears to be similar, are the Seagate 8 TB enterprise drives worth any premium you might have to pay for them? As we have previously documented, the Seagate enterprise drives load data faster and have a number of features such as the PowerChoiceTM technology that can be very useful. In addition, enterprise drives typically have a 5 year warranty versus a 2 year warranty for the consumer drives. While drive price and availability are our primary considerations, you may decide other factors are more important.

We will continue to follow these drives, especially as they age over 2 years: the warranty point for the consumer drives.

Join the Drive Stats Webinar on Friday, November 3

We will be doing a deeper dive on this review in a webinar: “Q3 2017 Hard Drive Failure Stats” being held on Friday, November 3rd at 10:00 am Pacific Time. We’ll dig into what’s behind the numbers, including the enterprise vs consumer drive comparison. To sign up for the webinar, you will need to subscribe to the Backblaze BrightTALK channel if you haven’t already done so.

Wrapping Up

Our next drive stats post will be in January, when we’ll review the data for Q4 and all of 2017, and we’ll update our lifetime stats for all of the drives we have ever used. In addition, we’ll get our first real look at the 12 TB drives.

As a reminder, the hard drive data we use is available on our Hard Drive Test Data page. You can download and use this data for free for your own purpose. All we ask are three things 1) you cite Backblaze as the source if you use the data, 2) you accept that you are solely responsible for how you use the data, and 3) you do not sell this data to anyone: it is free.

Good luck and let us know if you find anything interesting.

Andy Klein

Andy Klein

Director of Product Marketing at Backblaze
Andy has 20+ years experience in technology marketing. He has shared his expertise in computer security and data backup at the Federal Trade Commission, Rootstech, RSA and over 100 other events. His current passion is to get everyone to back up their data before it's too late.
Andy Klein

Latest posts by Andy Klein (see all)

Category:  Cloud Storage
  • Chi Kim

    Would you mind posting the tables in html? I’m a visually impaired reader, and my screen reader cannot read the image. Thanks!

  • Pingback: The Advent of the 12 TB Hard Drives - History-To-Share Ceramic Outdoor Memorial Plaques()

  • Pingback: The most reliable hard drives in the world – Gamers Alpha()

  • Pingback: Are enterprise hard drives more reliable than mainstream models? Report says no – Just another WordPress site()

  • regs

    Still no HGST NAS (HDN) and Toshiba N300 series in stats?

  • Ios5wasthebest

    I have 3 10T helium drives that drive me crazy by constantly making noises while idle.

    I’ve contacted both HGST and Seagate and was adviced to return to place of purchase. I did replace the Iron Wolf but the replacement still does the noise.
    I’ve read on forums that these noises are actually sign of background surface scan.

    Do any of you own Seagate or HGST in quiet home environment to confirm the same noises, and if yes, what can be done to shut them off?

    These are the sounds (seagate enterprise, seagate ironwolf, hgst ultrastar)
    https://youtu.be/Eqi1vBSofQA
    https://youtu.be/sAsGlwPqFUc
    https://youtu.be/yXGp8rwScM8

  • zxoiesrus

    I am very much looking forward to reading the stats on 10 and 12tb drives as time goes on. It is a good start, but I agree only 1200 units and just 1 year or less is not a good enough measure for reliability prediction.

    On the other hand, it’s not looking good for seagate 4 and 5tb drives or the 6tb wd red (EFRX) over the longer period.

    It’s also looking a little scary for the 8TB drives all decreasing like that and over not a particularly long period either.

  • gregge

    Do you wipe and sell older drives that are still working? (I could use a few 4TB drives with a bunch of life left in them.) What about the older pod chassis? Upgrade the boards etc. or total replace with newer designs?

    • I agree – I’d be interested in purchasing old pods, even if I need to buy drives to populate them with.

      • Milk Manson

        Pretty sure they shred them.

      • gregge

        I don’t care about the old Pods, I want some cheap, old hard drives. If a Pod is booted from a DBAN CD-R or USB stick, will that wipe all the drives connected through the SATA port multipliers?

  • OllieJones

    Great stuff as usual. Thanks. You’ve come a long way from the drive-shortage days when you sent us customers out to buy drives for you at Best Buy. I for one never doubted you would.

    Do you have any sense of whether drive temperature makes a difference? Have the drives with higher failure rate suffered temperature spikes or endured higher average operating temperature?

    I suppose your drives spin 24×7. Do you know whether the actuators on each drive tick away 24×7 as well, or do your drives and pods have noticeable active and idle periods?

    Are you buying COTS drives or are your vendors getting wise to your policy of publishing reliability stats and sending you the best drives they make? (Hey, I wouldn’t complain if they did: you have my data on those drives. But it might skew your reliability stats.)

    Thanks again.

  • Pingback: Backblaze HDD Report: No Reliability Difference Between Consumer, Enterprise Drives | USASpeaks.com()

  • Pingback: Backblaze HDD-Bericht: Keine Zeitliche Unterschied Zwischen Consumer -, Enterprise-Festplatten | Furkann.NET | Technik Nachrichten()

  • Pingback: Backblaze HDD Report: No Reliability Difference Between Consumer, Enterprise Drives - TachJokers()

  • emyzne

    What do you do with the old drives that you retire?

    • Hank Barta

      There seem to be a lot of “refurbished” drives on Ebay. I suspect most of them are retired drives from a cloud storage company (and which have passed the manufacturer’s diagnostic tests.) I can’t imagine that it is really cost effective to refurbish a hard drive for a fraction of the retail price. No way to know if these come from Backblaze or some other company. I suspect they sell them to a “refurbisher” for a few dollars or less each.

      • Milk Manson

        most refurbs are warranty returns…

        • Hank Barta

          I wonder if there are more warranty returns than retired drives.

  • Pingback: В сервисе Backblaze не увидели разницы в надёжности между серверными и домашними HDD Seagate – Charge-up.ru()

  • Sherwood_Botsford

    I can see some merit it splitting this up into batches and report the average days per drive. Is the failure rate random, in which case we would see an exponential decrease in the survivors of a given batch. I would be interested in seeing the failure distribution as a function of time. Is there merit in replacing a working drive just because it’s old?

  • Pingback: Low Security Pillow Storage | TechSNAP 343 | Jupiter Broadcasting()

  • pdwalker

    I really enjoy these reports.

    For years I had my own opinion on drive reliability based on anecdotal data. It’s nice to see it validated with massive real world samples.

  • Yuki Fuyui

    Seagate proving once again how useless they are despite their innovation and bullying Samsung into selling their amazing harddrive department who managed to cram 2TB into 9,5mm space.

    God I wish Toshiba would release 3 and 4TB 9,5mm 2,5″ harddrives.

    • Hank Barta

      Not strictly true. I’ve been working with ZFS on Linux and only yesterday set up a RAIDZ2 with 6 X 2TB drives. One of the drives I used was a Seagate I retired a couple years ago as the reallocated sector count hit about 3000. My intent was to work through the replacement of a failed drive. I just checked and the drive went offline about 2:30 PM – somewhere past the halfway point in a ‘scrub’ operation. In this case a Seagate drive was useful to support work with a ‘real world’ failure.

  • Pingback: Deze harde schijven zijn het betrouwbaarst | ZDNet.be()

  • Christian Opfer

    On the failed drives are you warranting them to the manufactures? What is your procedure disposing of the retired drives?

  • BeaveVillage

    I guess in the end the question is: Which hard disk should we avoid and which hard disk is the most reliable?

    • That’s hard for us to say. Our environment is different than most folks have at their home, so you can view our results as either a “best case” or “worst case” scenario. The important thing to remember is that all drives do fail, so you should be prepared for that!

      • Hank Barta

        I had a discussion with a co-worker about the service your drives see. My expectation is that they get written to and (less frequently) read from both on a lower frequency than the HDD in my desktop. This is, afterall, offline storage. My co-worker speculated that the drives see heavy read/write activity because so many users are involved. If you care to reveal this, I’d be interested to hear your thoughts.

        Thanks for publishing the information that I do find here!

        • Milk Manson

          They have mentioned previously that their drives tend to fill up then sit around smoking cigarettes while waiting for something interesting to happen. Or something to that effect.

          • Hank Barta

            Thanks – I didn’t see that comment but it makes sense.

    • My answer to this has just been “anyone but seagate.” None of the other failure rates ever seem as bad, and while many of the seagate models seem to do fine there always seems to be some variety with big problems. I guess with failure rates of around 4% I wouldn’t pay twice as much for an alternative but thankfully there’s plenty of comparable choices.

      • Lathe of Heaven

        And yet, kind of weird that Seagate is what these guys use most…

        • Well, when you have a massively redundant system setup the calculus is different, right? A 4% failure rate for me, where most of my drives are not in a RAID array, has a more significant cost in time. I’ve then got downtime and the hassle of a restore. So I’m going to simply pay a few percent more for a slightly better probable outcome.

          Backblaze, on the other hand, has much less immediate downside to a drive failure. So 4% if it saves them 5% on a drive purchase may well be no big deal.

          • Lathe of Heaven

            Well, I’ve bought Seagate external HDDs for many years, and overall I’ve never had the trouble that many others have, and most are still running. BUT… as soon as they dropped their warranty to ONE BLOODY YEAR!!! And WD had basically the same drive warrantied for 3 years, well, that was that… Any company who will not back their ‘flagship’ 8TB drive for more than ONE frigg’n year (especially when they began my first drives were covered for FIVE years!) they are NOT going to get any more of my business. Anyone who buys it with only a 1 year warranty is not being terribly bright in my lowly and wretched opinion… (the only exception would be, and I think I did this with my last 4 & 5TB Seagate purchases, would be to throw on an added 3 year Squaretrade unconditional warranty for like $15 or so…)

  • Pingback: Are enterprise hard drives more reliable than mainstream models? Report says no - Read The Net()

  • Pingback: Deze harde schijven zijn het betrouwbaarst - TechPulse()

  • Pingback: Deze harde schijven zijn het betrouwbaarst | Smart Business()

  • Pingback: Big hard disks may be breaking the bathtub curve - Texas Network Solutions LLP()

  • Pingback: Big hard disks may be breaking the bathtub curve – Wireless Bud()

  • Pingback: Big hard disks may be breaking the bathtub curve()

  • Pingback: Report shows enterprise hard drives are no more reliable than mainstream models()

  • Pingback: Report shows enterprise hard drives are no more reliable than mainstream models - Versatile Blogger()

  • Pingback: Large-capacity hard drives are holding up well in data centers | The Viral Info()

  • oliversl

    Having some sort of graph with days running/number of drivers/number of failures should look really nice, but I don’t have an idea on how to made it.

    • You want a https://en.wikipedia.org/wiki/Survival_function graph. Looking at the format of their data, it wouldn’t be easy because it’s daily, not cumulative. So you would have to iterate through the data for each drive until the day it is marked as dying & removed, to get the lifespan of that drive or whether it is still alive, and then with that you could do a survival curve.

  • DaktariD

    The stat that interests me most (and which I don’t see here) is average lifespan. Number of failures is interesting for sure, but what is the average lifespan of the various drives? This number would be especially interesting if the “died in infancy” drive counts are kept separate. I.e. excluding drives that died in the first month, what are the mean & median lifespans of each type of drive?

    • commenter

      Sounds like “Drive Days” divided by “Drive Count”.

      Edit: Well, not exactly as not all drives have reached their end of life yet :-/

      • Exactly. You can’t compute the average until every drive is dead (what if the last drive lasts a million years?). You can, however, compute some quantiles based on how many have died so far. If 51% of your drives are dead by day X, then you know the median lifespan is X.

    • Andy Klein

      One technique we seen is to create a “Kaplan-Meier” survival graph. It is generally applied in biological settings, but a couple of folks we know of have done this with our data. Using this method, the graphs show the life expectancy of the different drive models we have used.
      The most recent person (Simon Erni) published his findings here: https://hackernoon.com/applying-medical-statistics-to-the-backblaze-hard-drive-stats-36227cfd5372.

      • DaktariD

        Thanks. That’s exactly what I was looking for – tho I’d never heard of the K-M survival graph.

  • Pingback: Big hard disks may be breaking the bathtub curve - RYC HOME()

  • Pingback: Big hard disks may be breaking the bathtub curve - techsqrd.com()

  • Pingback: Big hard disks may be breaking the bathtub curve – NEWS Magazino The Online Daily News Portal #NewsMagazino()

  • Pingback: Big hard disks may be breaking the bathtub curve | techgeek()

  • donjjcarroll

    Is there any way to tell exactly what model hard drive is inside an enclosed product. For example, a local Best Buy store has the following drive on sale. Seagate Backup Plus Hub 8TB 3.5″ USB 3.0 External Desktop Hard Drive (STEL8000100). I am wondering if the enclosed HD would be a model that is on your Backblaze Quarterly listing.

    • tannerpowell

      I always wonder this. I wish it turned into a Wirecutter article at the end, but maybe an apples-to-apples hardware comparison isn’t possible.

    • DAVe3283

      I am not aware of a way to know without connecting the specific enclosure to a computer. But if you do that, you can run any number of S.M.A.R.T. utilities (I prefer Crystal Disk Info) to see exactly which hard drive is inside.

      I suspect the manufacturers intentionally don’t specify the internal drive used so they can switch what goes in it as prices change, to minimize their costs.

    • Alec Martin

      There’s no way to tell, unfortunately. If you take that drive out and attach it to a host computer using SATA, it will report itself with a model number used for the external drive it was sold as. As @DAVe3283Martin:disqus said, manufacturers may routinely change the bare drive model, while selling the external drive with the same part number. To make matters worse, the bare drives’ firmware may even report the same part number when attached using SATA, even though they’re completely different drives. Disclosure: I work for Seagate and I am not speaking on behalf of my employer.

      • donjjcarroll

        Friday, October 27, 2017, 4:14 PM

        Thanks you Alex and the others who replied or commented on my query re hard drives. I am not a “computer wiz” when it comes to hard drives but, because photography is a hobby of mine, I am annually requiring more external HD storage space for my photos and I always like to buy external HD’s with a good reputation. I have experienced over the years a few external HD failures, one just this year for a 2017 purchase, a Lacie 6TB T2 unit that was quickly replaced by Seagate by the way.

        I will soon need an 8Tb external so that’s why the BackBlaze article interested me but I needed a way to identify the better external HD’s when I go to the market place. Based on some of the comment offered up, this may prove problematic.

        Perhaps, I need to acquire an external box of some kind and then source the internal hard drive separately based on Backblaze’s quarterly reports?

        • Yat Yi

          fwiw, I have warranty replaced 3 of 4 8TB Seagate external backup plus hubs. All failed when over 90% capacity, and all failures were bad blocks / excessive reallocation (that means the drives were still readable, but some files got corrupted). I switched to the 8TB WD easystore drives from best buy – on sale they are around the same $180 price point. The enclosures contain WD red drives.

          I use these drives on a backup server in windows storage spaces (simple). That is generally the use case for these slow smr drives – infrequent writes. In my case, it’s infrequent reads also.

          • Lathe of Heaven

            Yep, I was concerned here because it looks like these guys don’t even use WD’s large drives (8TB) I’ve switched to them too just because to me 3 year warranty sounds a HECK of a lot better than 1 year warranty :) But, the fact that these guys don’t bother with them is disconcerting…

          • Joey Famiglietti

            I think the reason for this is that their needs don’t align well with WD’s offerings. WD doesn’t offer their performance ‘black’ series in the 8 or 10TB sizes, and the standard ‘red’ drives spin slower. ‘red pro’ and ‘gold’ drives are available, but at a measurably higher price per unit than other manufacturers.

            Backblaze addresses reliability needs through volume, so at $90/drive difference, buying 100 is drives is a $9,000 delta that can instead be addressed through purchasing greater quantities of drives.

          • Lathe of Heaven

            Thanks Bro! makes sense…

        • cjacja

          Sounds like you are a candidate for RAID. Then buy more smaller drives and when one fails the system rebuilds with no down time. You can buy multiples of whichever size drive has the best Terabyte to cost ratio.

    • Sherwood_Botsford

      Your best bet is to buy an empty external box with power supply for the drive, and to buy a drive for it. Based on these reports, I bought 4 HGST 4 TB drives, and installed them in my Mac Pro as two sets of mirrored drives. I also have Newer Tech external drive cases that hold two drives each.

  • Pingback: Big hard disks may be breaking the bathtub curve - IT AND US()

  • Pingback: Big hard disks may be breaking the bathtub curve – Scientist.ml()

  • Pingback: Big hard disks may be breaking the bathtub curve | Israelopadeyi.com - Wealth And Tech Digest()