Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×
Data Storage Upgrades IT Technology

NAND Flash Density Surpasses HDDs', But Price Is Still a Sticking Point (computerworld.com) 185

Lucas123 writes: With the introduction of 3D or stacked NAND flash memory, non-volatile memory has for the first time surpassed that of hard disk drives in density. This year, Micron revealed it had demonstrated areal densities in its laboratories of up to 2.77 terabits per square inch (Tbpsi) for its 3D NAND. That compares with the densest HDDs of about 1.3Tbpsi. While NAND flash may have surpassed hard drives in density, it doesn't mean the medium has reached price parity with HDDs — nor will it anytime soon. One roadblock to price parity is the cost of revamping existing or building new 3D NAND fabrication plant, which far exceeds that of hard drive manufacturing facilities, according to market research firm Coughlin Associates. HDD makers are also preparing to launch even denser products using technologies such as heat assisted magnetic recording.
This discussion has been archived. No new comments can be posted.

NAND Flash Density Surpasses HDDs', But Price Is Still a Sticking Point

Comments Filter:
  • by kheldan ( 1460303 ) on Monday February 08, 2016 @01:16PM (#51463369) Homepage Journal
    Have SSD's reached a point where they have a lifespan comparable to HDD's in the most extreme applications, though? For instance: Just had to replace the HDD in my DVR. It's dual tuner so it's buffering 30 minutes for each channel, perpetually. The HDD lasted for years; would a current-technology SSD last as long before it ran out of write cycles in the flash memory?
    • by gringer ( 252588 )

      Have SSD's reached a point where they have a lifespan comparable to HDD's in the most extreme applications, though?

      Yes.

      would a current-technology SSD last as long before it ran out of write cycles in the flash memory?

      Yes.

      • ..OK, are you sure about that? I'm not totally convinced. Let's be sure: If I put a 1TB SSD in my DVR, which is constantly buffering two HD video streams simultaneously, how long do you estimate the SSD would last before it ran out of write cycles? Keep in mind that a HDD will last 3-5 years in the same application.
        • by AaronW ( 33736 )

          I imagine that in a DVR a decent SSD should do just fine. Each NAND block is good for several thousand write cycles. If you completely overwrote the SSD every 24 hours that would give you years of service, though even a DVR rarely overwrites the entire hard drive in 24 hours.

        • by AnotherBlackHat ( 265897 ) on Monday February 08, 2016 @01:50PM (#51463653) Homepage

          OTA channels = 19.38 Mbps (max)
          2 channels = 38.76 Mbps = 4.845 MB/sec
          1 Terabyte SSD = 1,000,000 MB
          1,000,000 / 4.845 = 206,398 seconds, or 2.3 days
          Nand flash write cycle life : 10,000

          Total life 10,000 * 2.3 days = 23,000 days or 65 years

          If you don't like the assumptions, feel free to make your own, but I think it's clear that write cycle life isn't going to be the limiting factor.

        • by LordKronos ( 470910 ) on Monday February 08, 2016 @01:58PM (#51463705)

          The largest recording I've ever seen off of cable TV is about 8GB/hr. I know OTA broadcasts can be slightly bigger, so lets say 10GB/hr. To record that 24/7 requires about 87 TB/year.

          There was a long term test of SSDs done here:
          http://techreport.com/review/2... [techreport.com]

          Many of the drives ended up getting close to 1 PB of writes, and the best even got over 2PB. Thats enough for you to run 2 tuners 24/7 for a decade. And note, their tests were with 250GB drives. As you increase SSD capacity, longevity increases almost linearly. If you were building a DVR, you'd probably want something like a 1TB drive.

          As far as the original question of whether the SSD can outlive HDD in the most extreme application....probably at the most extreme, no. But for the vast majority of cases, including a DVR, most likely yes.

          • Re: (Score:2, Informative)

            You didn't read the report correctly. And I wouldn't plan on pushing anything past its rated limits for writes. That being said, "writes" are the limiting factor, just before general failure. Longevity is more than just writes, it is component failures as well.

            It is important to note, that spinning drives have significant drop in reliability at about 42 months (See BackBlaze stats). And while drives can last WELL into 8 years, if you're dealing with critical data, you really don't want to push it much past

          • If you're writing that much to a HDD it'll probably shake itself to pieces before the SSD expires.

            That's certainly been the experience at $orkplace when people abused nearline arrays as scratchpads.

        • by nneonneo ( 911150 ) <spam_hole AT shaw DOT ca> on Monday February 08, 2016 @01:59PM (#51463723) Homepage

          Let's work it out. A few years ago, TechReport ran an SSD endurance experiment [techreport.com] to figure out how much punishment current-gen SSDs could take before failing. Their test setup essentially involved writing random data at maximum speed for 18 months straight. The results indicated that the worst SSD in their bunch, a Intel's 335 Series, wrote about 700 TB before dying, and the best SSD, a Samsung 840 Pro SSD, went on to 2.4 PB.

          Various [xfinity.com] estimates [opposingviews.com] say you can put between 60-75 hours of HD content on a 500GB drive, so, assuming the largest possible size, that works out to about 8.3 GB/hour. Since you're writing two streams, that's 16.6 GB/hour, or 145 TB per year. For the worst drive in the bunch, that's about 4.8 years of service (right at the upper end of your HDD's service life); for the best drive, it's over 16 years.

          Keep in mind that these tests were all run on 250GB drives. Smaller drives have less flash to work with, and have to write over the same flash cells more often. Therefore, if you bought a 1TB drive, you can expect the lifetime to be easily 4x better (more if you're using a more recent drive, such as the Samsung 850 Pro) - 64 years of DVR recording should be more than sufficient.

        • by Gondola ( 189182 )

          I almost never use the live feeds, so I would happily trade that feature for a box the size of a VHS tape that performs better the rest of the time.

        • by gringer ( 252588 )

          If you're so disinterested in doing your own research on this, and believe SSDs still have this problem, no amount of explanation is going to convince you otherwise. I'm going to put you into the same basket as people who argue that Linux is not ready for regular users, and just ignore you. There's no point in arguing with someone who has such a rigid mindset. It's like trying to break concrete with a spoon.

        • SSD lives in an actual experiment are in this article:

          http://techreport.com/review/2... [techreport.com]

          The drive that did the worst failed at the 728TB written mark. These were 250 GB drives, so I would expect 1 TB drives to be able to sustain approximately four times the write volume. The means we should expect failure at about the 3.5 Petabyte mark. Two video streams should pretty much never exceed 10GB/hour. 3.5 PB/10GB =350,000 hours. That's about 40 years.

          Yeah, I think SSDs are OK for DVRs.

      • the correct answer is no.
    • Samsung Evo 840 is rated for a 28-year life span at 10GB of data write per day. That's about 100TB written. According to some tests [techreport.com], the 840 starts experiencing sector relocations (bad NAND) around 100TB; somewhere about 9 times that, it suddenly fails without warning.

      If you're constantly buffering HD video at 11GB/hr, that should give you 378 days to 100TB and maybe 9 years to sudden catastrophic failure.

      • OK, so at 100TB you'd start losing total drive capacity, right? I'll assume for the moment that the drives' controller would detect dead blocks/cells during a write operation and would perform the relocation/remapping seamlessly and not lose any data. But when it suddenly 'failed' would there be any chance of recovering anything from it? Or is it just bricked at that point?
        • by tibit ( 1762298 )

          You can't lose drive capacity, if that happens the drive is effectively dead: you have lost your data. What grandparent meant was that after 100TB of writes, the drive starts hitting flash blocks with uncorrectable errors. About 800TB later it'll have hit so many of them, that it can't find enough usable blocks to store the disk management metadata and the data stored by the user, and gives up.

        • OK, so at 100TB you'd start losing total drive capacity, right? I'll assume for the moment that the drives' controller would detect dead blocks/cells during a write operation and would perform the relocation/remapping seamlessly and not lose any data. But when it suddenly 'failed' would there be any chance of recovering anything from it? Or is it just bricked at that point?

          Depends on the failure mode and the drive design. Some of the Intel drives, for example, are designed so that once they reach their rated write limit, they switch themselves into read only mode (even if they haven't yet encountered their first error) until powered down, at which point they brick themselves. Pretty stupid design IMHO (why not just leave it permanently read-only to give you an extended chance to copy off the data).

      • by CastrTroy ( 595695 ) on Monday February 08, 2016 @01:49PM (#51463639) Homepage

        I think that the problem can be easily worked around by better designing the DVRs. Put 16 GB of RAM in there and buffer to that. You only need to write it out to the hard disk when you actually want to be recording a show. 16 GB should be enough for buffering the HD streams and allowing you to rewind shows as you're watching them.

    • by dnaumov ( 453672 )

      At the speeds a DVR would actually use to write things to the drive, modern SSDs will outlast HDDs by a pretty large margin.

    • by sims 2 ( 994794 )

      Supposedly they are pretty reliable now. http://hardware.slashdot.org/s... [slashdot.org]

      I still don't trust them. What they are supposed to do is fail in read only mode but what they often do is fail altogether.

      And afaik most data recovery services do not yet have the ability to process SSDs

      But they are faster and great in laptops because overall they are much more resistant to damage from drops.

      • by Foresto ( 127767 )

        That bothers me too, but I'm starting to think that manufacturers are deliberately avoiding a read-only failure mode for security reasons: if your drive enters a permanent read-only state, how do you erase it before recycling? I imagine having used crypto from day 0 would be your only safeguard at that point, but even good crypto gets broken eventually, so how do you safeguard the data on that read-only drive in the long term? Is physical destruction the only answer?

        On the other hand, maybe the total-failur

    • by Nemyst ( 1383049 )
      Modern SSDs will far outlast their expected lifetime. Surprisingly (not), SSDs did evolve in the decade or so since their introduction.
    • Then there's the value economics, too.

      Endurance testing have revealed modern SSDs to be remarkably reliable -- this guy wrote 7 PB to an 850 Pro. http://packet.company/blog/ [packet.company]

      But let's say the failure rate is N% higher than HDDs for a given application. But the drive itself is much faster and uses less power than a HDD. What number N is acceptable as an increased failure rate in exchange for the vastly improved performance?

      In an array, the performance increase may allow the use of single parity over double

    • by Chas ( 5144 )

      You mean like Seagate *DEAD!* drives....?

      Most decent SSDs of sufficient size already outlast these.

    • As NAND density grows, so will NOR density, and there will come a point where NOR density will match that of SSD. At that point, it will be a question of price reducing NOR, and once they do THAT, they will achieve both greater speed AND greater reliability. NOR flash does not have the cell integrity issues that NAND has.
  • Flash won already (Score:5, Insightful)

    by jones_supa ( 887896 ) on Monday February 08, 2016 @01:22PM (#51463423)
    If you look at a list of new computers, you will notice that a surprisingly large amount of PCs are already shipping with 128 GB or 256 GB SSD. That's gonna hold everything that most people need. People with bit more specialized needs (hardcore gaming, media production, virtual machines, etc.) can probably soon acquire 1 TB SSD for a price like $200. Only massive data centers will remain as users of HDDs. Flash memory companies are putting huge investments in developing the technology further, while HAMR is still a prototype in skunkworks that is struggling to be usable for mass production.
    • by creimer ( 824291 )

      Only massive data centers will remain as users of HDDs.

      Massive data centers are probably the only ones with enough money to convert HDDs to SSDs for the speed enhancements. Some of the larger capacity SSDs will probably appear first for the enterprise market.

      • Speed of the drives isn't just for Enterprise. Speed of the drives is really measured in IOPs, and just about anything meaningful can make use of the expanded IOPs available on SSDs

        Your BEST spinning disk has IOPs in the range of 700-800
        Your average SSD is now in the range of 100,000 IOPs

        No, that is not a typo.. It is how you get boot times in seconds rather than minutes.

        • "Your BEST spinning disk has IOPs in the range of 700-800"

          Sequential IOPS. Random is lucky to achieve 130.

          "Your average SSD is now in the range of 100,000 IOPs"

          That's random _or_ sequential - and even if there's a write slowdown most current SSDs drop to "only" 50-75,000 IOPs

          Then there's the warranty.
          Consumer SSDs mostly have 3-5 year warranties with some having 10 years.
          Consumer HDDs get 12-24 months at best.
          Enterprise HDDs get 3-5 years but that brings the price up into SSD territory for performance which

    • by Anonymous Coward

      We'll probably have dual discs for a while. My OS is on a small SSD but my music, photos, and videos are all on a separate 1 TB HDD since the speed difference doesn't matter.

      • by Gr8Apes ( 679165 )
        Until SSDs are less than 50% more expensive than spinning disks, spinning disks will still have a place. Fast enough for large backups, large enough to hold multiple backups, no need to spend a premium on those for performance. They will die out eventually, but it will be a few years out still before I'd start saying they're going to be dead. However for anything under 2TB you can pretty much write the obit.
        • "Until SSDs are less than 50% more expensive than spinning disks, spinning disks will still have a place."

          In the consumer arena the margin is 300%, but for anything below 1TB SSD has already won. The new generation of M2 PCIe 1TB devices are so cheap that they'll be the norm in less than 6 months.

          On the other hand I just put 24TB of SSDs into a server for scratchpad use, partly because of the speedups but mostly because we're pretty sure that the drives will have less downtime than HDDs.

    • by Kjella ( 173770 )

      People with bit more specialized needs (hardcore gaming, media production, virtual machines, etc.) can probably soon acquire 1 TB SSD for a price like $200.

      And you can get an 8TB Seagate Archive HDD for $223 at newegg today, if you need/want to store lots of data it's still cheaper by far. The real issue from the manufacturer's side is that nobody will pay a premium for anything. You get a SSD for all things performance and the cheapest, slowest HDD because for streaming huge media files you just have to be fast enough, they're mostly accessed linearly and even a video server for a big family only serves a handful of video streams at once. And a lot of people

      • The clock is ticking on spinning drives. HDD are still viable, but only for large backups. However, if all you look at is price, then you get what you pay for.

        • by Kjella ( 173770 )

          The clock is ticking on spinning drives. HDD are still viable, but only for large backups. However, if all you look at is price, then you get what you pay for.

          Sure. But apart from the personal content that you should have backed up in multiple+offsite copies for safekeeping and usually just consumes a little bit, for most people the HDD is a n'th level cache of the Internet. For many people that's even true of the SSD content, if you lose your Steam games folder or Spotify offline playlists well you can just download it again. And when it comes to digital media quantity is king, it's a lot easier to have three copies with 95% reliability than one with 99,99% beca

    • I have an old iPhone 4 with 64GB storage and an old iPad 2 with 64GB storage.
      So if both are backed up on my 256GB SSD MacBook Air, half of my storage is gone. Exagerating, yes. (But my iPhone indeed is full with about 59GB).

      When an empty Word document is sized in the MBs and and increasing pixel x pixel sizes of cameras/photos are the norm, I would not consider 256GB lot of memory.

      Now with HD movies, no idea how big they really are, but I doubt you get more than 20 on a 256GB laptop.

    • Re:Flash won already (Score:5, Interesting)

      by Nemyst ( 1383049 ) on Monday February 08, 2016 @02:11PM (#51463841) Homepage
      The thing that might dislodge NAND is Intel/Micron's new 3D Xpoint, which is supposedly much faster, allows for bit-level writes and is just as durable if not more so than NAND. It's also supposed to be available shortly.
    • If you look at a list of new computers, you will notice that a surprisingly large amount of PCs are already shipping with 128 GB or 256 GB SSD. That's gonna hold everything that most people need.

      Well, that's a bit difficult to generalize, which is a challenge that computer manufacturer's are having a bit more difficulty addressing. 128GB is fine for a browser/office suite computer, but with the OS taking 20-30GB of that (depending on OS/version/swap file size/hibernation file size), 128GB gets pretty cramped, pretty quickly, if a moderately sized iTunes library is involved. Moreover, phone backups / picture sync for images that are 10MP and higher will eat up that 128GB fairly quickly.

      256GB is abou

  • The second point is a limited number of rewrite cycles. There, FTFY.

    Oh, while we are at it, SSD tend to fail spectacularly: i.e. usually when they perish you cannot extract any information at all vs. spinning platters which usually fail gradually.

    P.S. If you wanna counter my first argument, fill your SSD up to 99% and then try to work with it continuously for quite some time. That 1% will get overwritten multiple times and your whole SSD will be prone to a failure. Those tests you've seen online all dea

    • by ShooterNeo ( 555040 ) on Monday February 08, 2016 @01:33PM (#51463527)

      That isn't how wear leveling algorithm work. Yes, once you hit 99%, every write does involve a rewrite somewhere, but those writes are not concentrated in the 1% free area. Instead, the drive controller is reading sections of already written disk and moving them around.

      • Some swapping of live data occurs, but having extra slack free space to move around in helps the algorithm better work within those constraints. In fact, Samsung provides a utility called Magician to manage Over Provisioning for extended life. It's not required, but ostensibly it does help.

      • by Bengie ( 1121981 )
        Older/cheaper SSDs do work that way. I wasn't until around the Samsung 840 or 850 that SSDs started to support wear-leveling untrimmed blocked.
    • SSD tend to fail spectacularly

      actually, the SSD controller usually spots when an error occurs and when it does, it puts the entire drive into read-only mode. on the other side, HDDs that suffer from motor burnout become unreadable.

    • Re: (Score:3, Informative)

      by Solandri ( 704621 )

      Oh, while we are at it, SSD tend to fail spectacularly: i.e. usually when they perish you cannot extract any information at all vs. spinning platters which usually fail gradually.

      Most newer SSDs are designed to fail gracefully. When they die, they become a read-only device. All your data is still accessible. Many USB flash drives are designed to fail the same way - if you've ever had a USB flash drive mysteriously become "write-protected", it probably died and set itself to read-only mode. Unfortunatel

    • by Nemyst ( 1383049 ) on Monday February 08, 2016 @02:07PM (#51463809) Homepage
      Fill up your SSD to 99% and it usually has between 20 and 40% free space to work with (more for enterprise drives, less for cheap drives). Oh wait, you've never heard of over-provisioning?
    • Re: (Score:3, Informative)

      by Anonymous Coward

      Hint: every SSD has *at least* 6% extra space for wear leveling - 1TB drives are internally 1024TiB.

      P.S. If you wanna counter my first argument, fill your SSD up to 99% and then try to work with it continuously for quite some time. That 1% will get overwritten multiple times and your whole SSD will be prone to a failure.

      P.S. Bullshit

      SMART Attributes Data Structure revision number: 1
      Vendor Specific SMART Attributes with Thresholds:
      9 Power_On_Hours 0x0032 094 094 000 Old_age Always - 28138
      177 Wear_Leveling_Count 0x0013 097 097 000 Pre-fail Always - 98
      241 Total_LBAs_Written 0x0032 099 099 000 Old_age Always - 9528109928

      That's a

    • The second point is a limited number of rewrite cycles. There, FTFY.

      Remind me to never take anything to you to get fixed. Especially if you base expert opinion on fundamental misunderstandings of technology.

  • Units (Score:4, Informative)

    by Nemyst ( 1383049 ) on Monday February 08, 2016 @01:39PM (#51463565) Homepage
    That's 4.294 Gb/mm^2 and 2.02 Gb/mm^2, respectively, for us SI folks.
  • This write up is misleading as it is comparing densities from the laboratory of one item to production densities of another item.

    Please compare apples to apples. Hard drives are more dense and cheaper than solid state drives, in addition to being far cheaper: Still, and into the forseeable future.

  • The OP rather implies that a supplier offering both convetional HDDs and SSDs of the same capacity would offer their products at prices based upon "cost of manufacture + margin" - i.e. that the retail prices would be a reflection of production costs. Sadly for consumers, this is blatantly not the case. The evidence for this is *everywhere* - for example a BluRay Movie costs no more to make, ship and sell than a DVD [maybe less, the packaging is smaller, lighter and cheaper to ship] and yet BluRay discs cost
    • by AHuxley ( 892839 )
      The unit price will stay the same but the amount of data will just go up.
      What was 60 or 100gb will grow to 500gb or 1tb for the same few hundred $ as the basic product range.
      More storage but no getting 60gb for this years very very, low price.

Retirement means that when someone says "Have a nice day", you actually have a shot at it.

Working...