Comments Locked

29 Comments

Back to Article

  • Samus - Saturday, May 13, 2017 - link

    That HGST purchase really paid off. Good job WD. You haven't ruined your mergers (or purchased crappy companies like Maxtor) like Seagate traditionally has (and yeah I'm especially talking about Samsung and LSI)
  • StevoLincolnite - Sunday, May 14, 2017 - link

    I miss the Samsung Spinpoints. They were decently quick, reliable and cheap.
  • tamalero - Sunday, May 14, 2017 - link

    Agree, those Samsung drives always ran cold, had amazing reliability and were pretty cheap.
  • jaydee - Monday, May 15, 2017 - link

    I bought a pair of Samsung Spinpoint F3 1 TB drives in June of 2012, that were second hand. Both still running strong in a RAID-1 NAS.
  • cbm80 - Saturday, May 13, 2017 - link

    Doesn't WD already sell drive-managed SMR drives? Like the 4TB passport?
  • AlphaBlaster - Sunday, May 14, 2017 - link

    It’s been a while since I’ve ranted about something, and my experiences a couple days ago triggered this post. First, a little bit of back story. Less than 3 years ago, I built a small home server that would function as a file and web test server. The storage portion required me to purchase five 3TB SATA hard drives. Being in a redundant RAID array, they simply needed to have agood price to value proposition. With all the online chatter about Western Digital being the highest quality drive manufacturer and all the drive prices being basically the same, I decided to purchase 5 WD drives. So with these new drives and an older WD drive that I grandfathered in from another machine, I assembled a 6 drive array.

    Everything went well until after 3 months the first drive failed. It actually happened right in front of me. I was monitoring the server stats when the IOwait spiked when the server was apparently doing nothing. I checked the logs and found that the drive was refusing commands. Being somewhat annoyed at the infant mortality of the drive, I decided to buy a Hitachi as the replacement. I can’t say exactly why, it might have been a bit more expensive but I don’t remember. About 6 months later, I decided to expand the array to 7 drives and purchased another Hitachi of the same model as the one prior, seeing how it fared so well. Fast forward a year and I was met with another drive failure. This one became interesting. Lastly, just this week I was migrating all the data – in fact the entire array – into another server. This move also included a switch to ZFS from NTFS (no laughing, I needed Windows support). I had to copy the data to offline storage drives, destroy the array metadata, move the drives, assemble the new array, and copy the data back. Sounds easy. I got up to the copying data part without a hitch, but I quickly noticed the data transfer rate was very low – about 1/8th the speed it should have been. I also noticed that the operation was going through a sort of cycle: 30 seconds of transfer, 5 min of idle – however the disk usage light was lit the whole time. After doing some investigating, I found that one drive busy 100% of the time. I tried to get SMART on it using smartctl and the command took almost 10 seconds to return the data. I ripped the drive out and the throughput instantly went back to normal speeds. It was replaced with a Hitachi as before.

    All this may sound like ordinary drive happenings, and that may be the case, but I can’t help but notice that WD drives are among the only ones that have ever failed on me and that being said, I have owned drives from almost every company under the sun. It isn’t just these experiences that led me to my dislike towards thSee company. I find that their products are inferior to other manufacturers. They are larger, louder, produce significantly more heat (which leads to poor energy efficiency), and I have had a few that have had bearings go bad. My spite comes from the countless bad experiences I have had with the company’s products over the last 5 years. I have since switched to all Hitachi drives and have never had a failure. I have one drive for example which has outlived 5 machines (servers and workstations) and is still chugging along. Why could I never get that reliability from WD? At my job I manage over 500 workstations with a good split of WD and Seagate drives. The ones I’m replacing 70% of the time are WDs. Something isn’t quite right.
  • asmian - Sunday, May 14, 2017 - link

    SImple question: did you buy WD RE drives? They are designed for your stated application, with 24/7 availability, higher intrinsic BER value, higher vibration tolerance, and most especially TLER (Time Limited Error Recovery) as default for use in RAID environments (drives drop out gracefully to allow the array redundancy to provide the data without stalling the array, as you describe, continuously trying to retrieve rotten data)...

    It sounds like you did not. Therefore, the issue is that you tried to use cheap consumer drives in an unsuitable environment. If you cheap out, expect unsuitable drives to fail, from whatever company. There is a reason enterprise-rated drives have a higher price - intrinsically better build and greater reliability. Just because RAID originally meant "Redundant Array of Inexpensive Disks" doesn't mean that you should go and buy the cheapest possible because RAID is covering your ass mitigating the poor reliability, or fall for the drive company marketing that somehow their "NAS-rated" consumer drives are any different from normal ones, or in the same league as enterprise drives - that's pure bullshit.

    I've never bought (or recommended to consumer clients for builds) anything less than RE disks, always from WD. Apart from one batch of 1TB REs that had unfortunate infant mortality issues, ALL I've personally installed and recommended are still trucking after many years of fully active service. You get what you pay for. The issue isn't generally the brand, it's foolish buyers expecting cheap consumer drives to have the reliability of enterprise drives in 24/7 home servers and high-uptime environments.

    If you are managing 500 obviously important workstations needing good uptime, then I (and your company should) seriously question your competence if you are cheaping out with consumer drives in a business environment. You may save them a little up front but they'll pay far more over time due to the low reliability - that is why you are wasting man-hours replacing so many. THAT is what is not quite right. Brand is almost immaterial in this regard.
  • close - Sunday, May 14, 2017 - link

    It sounds to me like the first part of the story is the "home" part, totally separate from the 500 workstations "work" part.
    Anyway this is anecdotal evidence and it's hard to extrapolate from a relatively small number of drives. For consumer drives Backblaze provides good data. You get to see how consumer drives handle the worst of conditions.
  • HomeworldFound - Sunday, May 14, 2017 - link

    Your experience isn't enough to shun an entire brand and all of their products. There are plenty of these stories for each manufacturer. Why aren't you talking to Western Digital about it? Why are you just replacing drives?

    These days you pick the right drive for the right job with the right warranty.
  • nunya112 - Sunday, May 14, 2017 - link

    I agree WD drives have been the only ones that fail on me. and the exact same thing would happen to mine. it would just sit there on full activity but be doing nothing
  • dgingeri - Sunday, May 14, 2017 - link

    I have had similar issues with Seagate drives, with nearly every drive I've had from them fail over the years, yet despite having over 30 WD drives over the last 27 years, I have only had one WD drive fail on me. I've also had about 20 Hitachi drives, including 6 4TB drives I'm running right now, and not a single failure from them.

    For 6 years, I was a sysadmin for a test lab for server level storage products. I had a LOT of drives to caretake during that time. In that time, Seagate were, by FAR, the least reliable drive I dealt with, and Toshiba pretty close behind. (I had one model of Toshiba, a 15k rpm 146GB 2.5" SAS drive, that had every single drive fail over the course of 3 years. We had 13 in every Dell R910 we had, with 21 more in a MD1220 attachment with a third of the R910s having 2 MD1220s. So, in short, we had several hundred of these drives. In three cases, when the drive failed, it failed so spectacularly that it damaged the backplane of the R910.) WD was third, but pretty far behind Seagate and Toshiba. Hitachi was by FAR, the best of the lot, with about a dozen failures out of a sample size of about 2000 1TB SATA drives.

    After my experiences in that test lab, I will not buy any other brand internal drive other than HGST. Since HGST doesn't have USB drives that I've found, I do go with WD for my external backup drives.

    One other thing I've learned from storage, never back up to another internal drive. Always backup to an external drive. You never know when a power supply issue will kill all the drives in your system. At that point, your backups don't mean squat.
  • Magichands8 - Saturday, May 13, 2017 - link

    Great! All this means they can start reducing prices, right? Implementing these new technologies to increase capacity is nice but it's increased drive costs substantially. In all the time since these have been on the market I haven't seen prices drop below about $0.35/GB.
  • harobikes333 - Saturday, May 13, 2017 - link

    Magichands... unless the market puts pressure on the prices, why would they cut their profits??
  • cbm80 - Saturday, May 13, 2017 - link

    Cramming more platters doesn't reduce costs. Neither does helium. The only benefit is that data centers can save some space and power.
  • Xajel - Sunday, May 14, 2017 - link

    HeF-HDD is still a new technology, and new technology comes with a price. in addition to that, it's really not necessary in low capacity HDD's, as these can work already without it so adding it means adding costs for nothing.

    In another hand HeF-HDD is only meant to give the ability to add more platters, denser platters while maintaining the RPM speed without worrying about heat, vibration, stability and other factors which comes with such methods of increasing density without using denser platters which take longer time.

    So when 8TB and larger HDD's becomes mainstream, then you might see HeF-HDD's becomes feasible in consumer market.

    HAMR is also coming, but again it's very difficult to make, that's why we don't have it yet ( coming next year IIRC ), this technology will increase the density also and it can work with or without HeF-HDD, but it's cost in the first place will make it expensive so as usual we will see it first in high-margin market ( enterprise/datacenters ), later it will make it's way toward consumers.
  • StrangerGuy - Sunday, May 14, 2017 - link

    Apple already proved maximizing margins/unit sold in a sufficient enough volume is a much more profitable approach than the old days of winning raw marketshare or volume sold. That means the consumer market are getting the shaft in favor for big iron enterprises who as much less price sensitive. Case in point: Ryzen is slaughtering Intel in consumer desktop CPU market yet Intel didn't even bother fighting AMD in a price war, but instead raised prices with Skylake-X. Same goes for NV aiming HPC first and gaming a distant second.
  • Beaver M. - Sunday, May 14, 2017 - link

    Skylake-X isnt released yet and the prices arent known yet either...
  • iwod - Sunday, May 14, 2017 - link

    So in the last 12 months, they shipped 10 million of these HDD, if we assume 10TB per unit sold, that is 100 Million TB, or 100 Exabytes!!!!

    At the end of 2013 Facebook reported a 300PB of Storage, even if it had grown 10x in 4 years, that is still only 3EB!. Facebook users nearly double from 2013 to 2017, even if you another 3.3x multiple that is still only 10EB!
  • ZeDestructor - Sunday, May 14, 2017 - link

    The newest number I can find is from 2015, when YouTube claimed 400 hours of video uploaded per minute. Assuming that it's increased another 100 hours per minute since then (it's been linear since they hit 100 hours per minute), that's 500 hours per minute. Some Wolfram|Alpha later, that works out to 60gbit/s = 648TB/day = 236.5PB/yr.

    All that gotta get stored somewhere, in multiple resolutions, in multiple copies. And then you have nutters like me upload at proper bitrates and allow youtube to stream to you in so-called "original" quality, so probably around the 1EB/yr. And that's just one service.

    Once you have all the other stuff.. AWS, S3, Google storage, GCE, etc... 100EB/year doesn't look that far-fetched when you're talking about what is literally half of the HDD industry's high-capacity production, does it?
  • Dug - Monday, May 15, 2017 - link

    Users are sucking the helium out of the drives, so they aren't actually being used for storage.
  • Beaver M. - Sunday, May 14, 2017 - link

    So can we expect lower prices on 8+ TB HDDs then? Even the cheapest one is still as expensive as 3 years ago. And even my 2 & 4 TB HDDs still cost as much as when I bought them 6 and 3 years ago.
    I am starting to suspect price fixing.
  • HomeworldFound - Sunday, May 14, 2017 - link

    I've been thinking like you lately. M
  • HomeworldFound - Sunday, May 14, 2017 - link

    Stupid keyboard. I've been thinking like you lately. The hard drive market has been running very slow. It wasn't until 4TB SSD's were announced that the latest jump happened. The mainstream storage manufacturers finally bought / opened SSD divisions and released their own line of SSDs.

    I'm very surprised that they aren't making 50TB drives by this point. Maybe I was being unrealistic but it was four or five years ago scientists were talking about 1000TB holographic discs.
  • voicequal - Sunday, May 14, 2017 - link

    Cloud providers are buying up all the high density HDDs which is supporting prices. I would expect HDD prices to continue to increase as they become more of a niche technology for data archival - most of which is moving into the cloud.
  • ats - Monday, May 15, 2017 - link

    Reduced competition and increased high capacity demand with reduced low capacity demand. Basically, the cloud vendors have been buying every topline capacity drive they can for the past 3-4 years effectively propping up demand for that segment. Meanwhile the rest of the market has pretty much been replaced by SSDs. That combination means that the vendors have little to no incentive to sell cheaper than the cloud vendors will pay and there is no motivation to drop max capacity prices in the wholesale/retail market.

    You can find cheap drives but they tend to be seconds/defect drives sold into the white label or portable/removable market.
  • DanNeely - Monday, May 15, 2017 - link

    SSD penetration is nowhere near as high as "he rest of the market has pretty much been replaced by SSDs". I can't find any numbers from this year (probably too soon to have aggregated even the Q1 data yet); but at the end of last year they were only projecting half of all PCs and an eighth of desktops sold this year would have them. Penetration is near ubiquitous in the enthusiast market; but outside of it sticker price is the most important factor. That is followed by bigger numbers which HDDs win because ###GB/TB goes on the sticker; not mixed (read/write and variable data chunk size) IO throughput at QD 1.

    http://www.tomshardware.com/news/ssd-hdd-shortage-...
  • sorten - Sunday, May 14, 2017 - link

    This is no laughing matter.
  • jimjamjamie - Monday, May 15, 2017 - link

    HeHeHe
  • RubiNBA - Monday, May 15, 2017 - link

    Everything went well until after 3 months the first drive failed. It actually happened right in front of me. I was monitoring the server stats when the IOwait spiked when the server was apparently doing nothing. I checked the logs and found that the drive was refusing commands. Being somewhat annoyed at the infant mortality of the drive, I decided to buy a Hitachi as the replacement. I can’t say exactly why, it might have been a bit more expensive but I don’t remember. About 6 months later, I decided to expand the array to 7 drives and purchased another Hitachi of the same model as the one prior, seeing how it fared so well.

    NBA Live Mobile Online - http://ow.ly/AHwi30aK5bD

Log in

Don't have an account? Sign up now