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

 



Forgot your password?
typodupeerror
×
Data Storage Stats

No, Your SSD Won't Quickly Lose Data While Powered Down 106

An anonymous reader writes: A few weeks ago, we discussed reports that enterprise SSDs would lose data in a surprisingly short amount of time if left powered off. The reports were based on a presentation from Alvin Cox, a Seagate engineer, about enterprise storage practices. PCWorld spoke to him and another engineer for Seagate, and they say the whole thing was blown out of proportion. Alvin Cox said, "I wouldn't worry about (losing data). This all pertains to end of life. As a consumer, an SSD product or even a flash product is never going to get to the point where it's temperature-dependent on retaining the data." The intent of the original presentation was to set expectations for a worst case scenario — a data center writing huge amounts of data to old SSDs and then storing them long-term at unusual temperatures. It's not a very realistic situation for businesses with responsible IT departments, and almost impossible for personal drives.
This discussion has been archived. No new comments can be posted.

No, Your SSD Won't Quickly Lose Data While Powered Down

Comments Filter:
  • by turkeydance ( 1266624 ) on Monday May 25, 2015 @10:37AM (#49768493)
    yeah, right.
  • by Anonymous Coward

    Backup your stuff. Make it a part of your ritual. Be the data.

    • by youn ( 1516637 )

      Very true, backup is key... but if you are backing up on media that loses data, are you really backing up the data?

      • all media lost data. it's just a matter of managing the loss rate.

        • all media lost data. it's just a matter of managing the loss rate.

          Part of managing loss is understanding what makes different media lose data. For optical media, it is light. Don't leave them sitting in the sunlight. For SSDs, it is temperature. Don't store your SSDs in a hot attic, or leave them in your car parked in the sun. I put all my flash media (SSDs, SD-Cards, thumb drives) in a ziploc bag to protect them from condensation, put that inside a sealed mason jar (just in case there is a pinhole in the ziploc bag), and then put that in the bottom drawer of my refr

          • unless you've vacuum-sealed those mason jars or have a dedicated fridge with redundant power (lol), the humidity is probably doing more harm than the (semi-)controlled temperature is helping. a merely closed jar is not "sealed," even with one of those rubber rings.

            also i'm imagining your jar of thumb drives sitting in the fridge. what the fuck? do you ever use them? or is this long-term storage? if you do use them, you are of course letting the jar come to ambient temperature in a dry location each time you

          • by WaffleMonster ( 969671 ) on Monday May 25, 2015 @12:26PM (#49769213)

            I put all my flash media (SSDs, SD-Cards, thumb drives) in a ziploc bag to protect them from condensation, put that inside a sealed mason jar (just in case there is a pinhole in the ziploc bag), and then put that in the bottom drawer of my refrigerator, right next to my battery stash.

            This seems unwise. Packaging everything up in an airtight container under presumably room temperature and proceeding to put it in the fridge to "protect them from condensation" is great way to generate condensation.

            You might want to consult a psychrometric chart or invest in desiccant.

      • by Lehk228 ( 705449 )
        there are 4 options that escalate nicely for the first 3 steps, CD, DVD, BDR, Tape

        tape is horribly expensive for the drives and so unfortunately it is not a realistic option for home users. but blu ray holds 50 gigs (100 if you want to mess around with very expensive and semi-compatible XL discs)
        • there are 4 options that escalate nicely for the first 3 steps, CD, DVD, BDR, Tape

          These are not good options for home users because they all involve ongoing work. Just install a 4GB network drive (available on Amazon [amazon.com] for less than $200), and set up your home computers to do hourly incremental backups. For extra safety, buy two, keep one at your office, and swap them once a month.

          • by Lehk228 ( 705449 )
            if it's on a hard drive it's not a backup, if it's mounted at all times, it's not a backup.
            • Re: (Score:2, Informative)

              by Anonymous Coward

              Wrong. If it's a second copy of the data, it's a backup. Doesn't matter where it is, it could even be on the same device as the original, still considered a backup. You can use location of the backup to define situational appropriateness, but location does not determine status of "backup".

              • Wrong.
                Backups have to be on media that is completely disconnected. No data and no power. Otherwise it's a copy.

              • If it's on the same device, it's NOT a backup, because failure of a single device kills both copies.
            • Your first claim is false. Any storage can act as a backup, including printing things out. You are making a personal judgement over quality.
              Your second claim is false. Recycled backups are perfectly valid and highly used (for example cycled daily backups as part of a tape rota).

              What you mean in the first claim is that a hard drive may not be THE most reliable form of backup storage (but then, its not the least either).

              What you mean in the second claim is that a backup should act as a separate snapshot of da

              • Online copies are just RAID done at the file level instead of the block level. The reason that RAID is not considered a substitute for a backup is that user error or a compromise can damage all online storage. If your backups are online, they are not backups, they're just redundancy.
                • That depends on how the online copies are handled. I use Crashplan on a drive at a friend's house. It has file versioning and files deleted locally are available on the backup. The backups are encrypted and if I have a total drive failure and I can go get the drive and restore locally instead of over the network.

                  This does not deal with bare drive restores. My current plan is to not worry about it, if I lose a whole drive I'll do a new install and then restore the data files from the backup.
                  --
                  JimFive
          • by dbIII ( 701233 )
            That's just shifting the problem to someone who cares far less about the problem than you do. It leaves you one corporate restructure or changed business plan away from losing the lot and one backhoe incident away from not being able to get it when you want it.
            External USB drives are a very good stopgap or even ongoing solution for home users that don't want to mess about with optical media.
            • Plus it is a good place to dump your "photos" that you don't really care too much about if they were to be lost and you had to "take" them again.

        • Re: (Score:3, Interesting)

          by Grishnakh ( 216268 )

          Blu-Ray is total crap, as is all optical backup media.

          The biggest problem is that it's simply far too small to be useful. To back up a 1TB drive, you'll need 20 BD-R discs. That's a lot of swapping to do a full backup.

          Aside from that, they're slow, write-once, and suffer a lot from bit-rot due to poor quality media. DVD-Rs and CD-Rs had the same problems before, and those are completely useless for backup these days due to their puny size.

          The only thing that makes any sense at all for consumers is USB ha

          • by dbIII ( 701233 )

            The only thing that makes any sense at all for consumers is USB hard drives

            They have a far shorter average offline life than the optical media. For one thing the bearings have a finite life as the lubricant breaks down over time - a lot longer than it used to be but still less on average than media problems with the optical disks.
            What is viable is to shuffle stuff off those USB drives onto newer ones a few years down the track, but if you are not prepared to do that eventually the optical stuff is going to

            • What are you talking about? That's a bunch of crap. A USB hard drive is nothing more than a standard laptop hard drive in a special box with a SATA-to-USB interface board. If what you said was true, then laptop hard drives would be failing left and right; instead, they have generally excellent lifetimes, just like desktop hard drives. And USB hard drives actually do better since they're unpowered most of the time.

              • How many fifteen year old laptop drives have you tried to spin up lately? Now compare that failure rate with optical or tapes.
                As I wrote above, the life has been extended (it used to be a lot of failures in a lot less than fifteen years) but the underlying modes of failure like the one I mentioned above are still there. Another unpowered failure mode is highly polished parts diffusing together over time so the drive "sticks", once again less of an issue than it used to be due to design changes taking that
          • Actually, the other poster's suggestion of using a NAS drive makes a lot of sense too.

            Only if it's not always online, doesn't continuously backup, or isn't located on the same premises as the primary drive, otherwise you'll find your NAS will lose data at the same time when your PC gets nuked by flood, fire, theft, lightning, virus etc.

            • Oh please. The main reasons for backups (esp. for individuals) are 1) user error, and 2) hardware failure, not acts of god. An on-site NAS protects against those just fine. It protects against viruses too, since the virus isn't going to easily affect the NAS box (though even better is to simply not run Windows).

              Theft? How many people have had thieves run around their house looking for NAS boxes? What kind of thieves even bother to steal electronics these days anyway? They aren't worth enough on the us

              • by thegarbz ( 1787294 ) on Tuesday May 26, 2015 @03:55AM (#49773361)

                Oh please. The main reasons for backups (esp. for individuals) are 1) user error, and 2) hardware failure, not acts of god. An on-site NAS protects against those just fine. It protects against viruses too, since the virus isn't going to easily affect the NAS box (though even better is to simply not run Windows).

                Theft? How many people have had thieves run around their house looking for NAS boxes? What kind of thieves even bother to steal electronics these days anyway? They aren't worth enough on the used market to bother with.

                - My only ever data loss has been the result of a direct lightning strike. Or at least it would have been if the backup wasn't located at work. I lost my PC, my NAS, and most electronics in the house in one go.

                - My best friend got cleaned out by robbers who amongst other valuable things took a computer AND the NAS box.

                - As for the viruses your post is simply ignorant of the modern trends. You know there's now 4 randsomware packages (that I have heard of) in the wild that will affect NAS boxes as well. 2 of them encrypt all data on NASs, one of them wipes data from the NAS but encrypts local data, and the last used a bug in Synology's software to render the NAS useless (though reports of removing the drives and putting them in other computers solved the problem.)

                - Remember the phrase RAID is not a backup solution? That applies equally to NAS for ALL the same reasons.

                So there you go. But I'm not going to suggest anything to you. You're more than welcome to learn by yourself.

                • You and your friend are extremely unlikely. Most people never get hit by lightning, ever. Most people also never get burgled. (Your friend was not robbed, BTW, unless he was in the house at the time.)

                  As for ransomware affecting NAS, that's what you get for running Windows.

                  • No, that's what you get for anyone on your network running windows. Data loss is in the realm of unlucky. I've never had a harddisk die due to natural causes, but then none of my drives are more than 3 years old either.

                    Also it's not about luck, it's about risk management. I live in an area at the very top of a hill prone to lots of summer electrical storms. My friend lived in a low budget rental with 4 random roommates (former roommate was the thief, didn't get the data back though because the equipment got

                    • No, that's what you get for anyone on your network running windows.

                      Not a problem on my home network. Besides, that's only if you actually set your system up so that Windows has access to the NAS box.

                      My friend lived in a low budget rental with 4 random roommates (former roommate was the thief,

                      That sounds like a good lesson in choosing your roommates more carefully, or better yet not having any (and certainly not 4).

                      For $100 + a few minutes every weekend doing incremental backups, I'll go the external harddi

                • by Bengie ( 1121981 )
                  A small pocket NAS isn't much of a NAS. Let me know when someone steals your server that takes two people to lift, stored down stairs in a locked room with a metal door, and will probably cause them to slip a disk in their back if they try to walk it up the stairs unless they bring a dolly. If you have a small NAS like that, hide is somewhere. It probably doesn't take much cooling. Don't leave it out in plain sight for all to see.

                  Ransomware? Don't you mount all of your NAS drives as SAN block devices shar
              • "The main reasons for backups (esp. for individuals) are 1) user error, and 2) hardware failure, not acts of god. An on-site NAS protects against those just fine"

                It doesn't protect against burglars.

                Seriously. We've had several instances at $orkplace of equipment being stolen from staff residences and the backup drives + NAS going too.

        • Last time I checked (last week), 1TB worth of empty BDRs is about twice more expensive than a 2 TB HDD. it's better to buy 2x 1 TB drive, back-up with redundancy and keep them plugged in but powered off, or store them in two different locations.

    • Excellent advice. (but) back it up on what sort of media? not likely SSDs stored on the furnace?
    • by Anonymous Coward

      Backup your stuff. Make it a part of your ritual. Be the data.

      You'd think that, wouldn't you?, but just last week our PHB questioned why we made backups, and why did we need so many?
      That, coupled with the fact that he's making erratic IT decisions which are about to fubar our production (e.g. he was about to unilaterally switch off our firewall and both of our fileserver) made for a fun week.

      Yes, I'm looking for another job..

  • by Anonymous Coward

    I'm interested to note this source was a representative of Seagate.

    That makes sense, every Seagate drive I've ever owned lost a huge amount of data whenever it was powered off. Or left on.

    • I'm interested to note this source was a representative of Seagate.

      It is also interesting to note that Seagate's HDDs are the least reliable [backblaze.com]. Hitachi's are the most reliable. I don't know how much Seagate's lack of attention to quality in their HDDs is reflected in their SSDs.

      • I don't know how much Seagate's lack of attention to quality in their HDDs is reflected in their SSDs.

        I don't know either, and I'm not willing to find out the hard way.

  • based on a presentation from Alvin Cox, a Seagate engineer[...]Alan Cox said, "I wouldn't worry"

    Can we get these two gentlemen to agree on a statement of risk? Or maybe just a little, you know, editing from the Slashdot editors?

    • by Shoten ( 260439 )

      based on a presentation from Alvin Cox, a Seagate engineer[...]Alan Cox said, "I wouldn't worry"

      Can we get these two gentlemen to agree on a statement of risk? Or maybe just a little, you know, editing from the Slashdot editors?

      I'm wondering if the "editing" from the Slashdot editors wasn't the problem in the first place. How many Slashdot summaries wildly overstate/oversimplify/remove from proper context the real meat of a story? How many Slashdot comments essentially say, "RTFA...you'll see that [it only applies to this situation|they mean this instead of that|this was done on purpose under wildly crazy conditions to see if it could ever be true at all|this person has no credibility|this is really advertising for someone's pro

      • Nah. Blaming the editors is a long and time honored tradition at Slashdot, as you well know.

        Every once in a while I get annoyed about something and write a letter to my local newspaper. They pretty much always publish it, and (annoyingly to me) they pretty much always edit it for style, grammar, length, etc. It annoys me to be edited on a very subtle and debatable point of grammar, but I can't say they aren't doing their job. Contrast that to our Slashdot editors... (sigh).
      • by unrtst ( 777550 )

        based on a presentation from Alvin Cox, a Seagate engineer[...]Alan Cox said, "I wouldn't worry"

        Can we get these two gentlemen to agree on a statement of risk? Or maybe just a little, you know, editing from the Slashdot editors?

        I'm wondering if the "editing" from the Slashdot editors wasn't the problem in the first place. How many Slashdot summaries wildly overstate/oversimplify/remove from proper context the real meat of a story?...

        In this case, someone copy/pasted from the article, and then introduced "Alan " into the text. The linked article reads:

        Cox agreed saying there’s no reason to fret.
        “I wouldn’t worry about (losing data),” Cox told PCWorld. “This all pertains to...

        I almost hope they introduced that error on purpose, just so we'd have something else to talk about. The alternative is just sad.

    • by Anonymous Coward

      based on a presentation from Alvin Cox, a Seagate engineer[...]Alan Cox said, "I wouldn't worry"

      Can we get these two gentlemen to agree on a statement of risk? Or maybe just a little, you know, editing from the Slashdot editors?

      And what do SSDs have to do with the Linux 2.2 networking code?

  • by Anonymous Coward

    I have tried SSD's since they were a new thing. I'm a software developer. To this day I have never seen one last more than a year under my "abusive" conditions. Yes, they are awesomely fast but longevity is just not their thing.

    I still have a normal old spinning hard-drive keeping my most important data. This drive has over 83000 powered-on hours (according to SMART). In the time I have had this drive I have replaced half a dozen SSD's that weren't really used for much other than OS data. What does that tel

    • Comment removed (Score:4, Insightful)

      by account_deleted ( 4530225 ) on Monday May 25, 2015 @10:58AM (#49768635)
      Comment removed based on user account deletion
      • you'd best watch out, you only have 4226 hours left on that drive! ;-)

      • by Kjella ( 173770 )

        My first OCZ Vertex - the original one - I did nothing to optimize it and spent as temp drive for everything, including torrent downloads that I later archived. It does 1.5 years later after eating through a 10k writes/sector endurance, if I read the SMART data right it had 9.6k write average and 14k writes worst case. My replacement drive from WD I did all the basic stuff to optimize and kept my torrents to a HDD, it lasted about 3.5 years and the lifespan indicator said it should have another 1.5 years le

      • by hawguy ( 1600213 )

        What does that tell you?

        That either you're exaggerating or you do something very, very wrong with your drives? I use low-end, cheapest of the cheapest consumer-SSDs and for example this 64GB drive I have has logged 9607 hours of power-on time. Has it broken yet? No. Are there any issues with it? Nope, not a single one. I do have to ask you: what exactly do you do with those drives of yours if you manage to break them so fast?

        Unless you use a checksumming filesystem (zfs, btrfs, etc) and regularly scrub the filesystem (both data and metadata), would you even know if you lost a few blocks of data?

        • Sub question: Would you care if you lost a few blocks of data? Bit rot is a serious concern but the problem is an isolated one. If I lost a single photo on my computer I may be lightly upset, but no where near as upset as losing all the data on my drive.

          • by hawguy ( 1600213 )

            Sub question: Would you care if you lost a few blocks of data? Bit rot is a serious concern but the problem is an isolated one. If I lost a single photo on my computer I may be lightly upset, but no where near as upset as losing all the data on my drive.

            Like you said, it depends which block.

            I could have a corrupt block in the middle of a movie file and not even notice, but if I lost a block near the beginning of a large compressed archive file, I may lose the entire archive -- and if it's silent corruption undetected by the filesystem or disk drive, it may be propagated to all of my backups before it's discovered.

            I'd rather have the drive fail entirely versus slow undetected block corruption.

            • I don't understand why you think undetected block corruption is a sign of a failing drive? Bitrot these days with the densities in modern drives is kind of something that you should expect to happen. It's nothing unique to SSDs or even unique to devices which are at end of life.

              But I had similar thoughts to you too which is one of the reason my offline backups are unencrypted. It is time to give btrfs a go though. It should be relatively stable now I don't think I can talk myself out of it any longer.

      • by Anonymous Coward

        I'm a developer and we have pretty much all been using SSDs without problems at work for the last two years. However, the other day our IT guy was saying that one of the older Dells killed all 3 SSDs he put in the thing, each time the drive was replaced free under warranty, and the third drive which failed was a different brand, after that he gave the machine to a less demanding user and stuck with an HDD.

        So, I have no lack in confidence in SSDs, but that incident leaves me to believe some older SATA contro

    • ... I have replaced half a dozen SSD's that weren't really used for much other than OS data. What does that tell you?

      OS data is poisonous?

    • That as an early adopter you sure must have used some of the least reliable SSDs, with wear leveling not even quite sorted out, firmware issues or just had random bad luck. Imagine if you ever bought an IBM Death Star, Seagate 7200.11 and that bad Maxtor series as HDDs. Some other HDDs still work fine after ten years. Compared to SSDs they just have a very different kind of sword hanging above their heads.

    • by Bengie ( 1121981 )
      You purchase crappy SSDs or are purposefully unaligning your partitions and formatting with 64KB blocks and writing programs that write to your FS without using a buffer and doing

      foreach(byte b in array)
      file.write(b)
  • by Anonymous Coward on Monday May 25, 2015 @10:57AM (#49768633)

    We have a long-term (over several years) exercise where we collect large amounts of data at a very hot waterside location overseas, in an air-conditioned office. And then the racks get moved to a non air-conditioned warehouse where they sit for a year or longer. And then we come back after a year or longer, move the racks back into office space, and do the next iteration.

    We were thinking about going to SSD, just for the drive performance, and we now know that our setup is a poster child for the problem that the alarmist article described.

    Of course, we copy off the valuable data and take it home, but coming back to random corruption on our system and middleware drives could introduce some real issues.

  • I have several personal drives in a storage shed, that end up there as secondary backups when I upgrade the drive in a system. It's always about 5 degrees warmer in the shed than it is outside, so 110F+ in the summer. I'm guessing that the ones that have been out there for 5 years, both SSD and spinning platter, are probably toast now.

  • by Anonymous Coward

    In an enterprise environment the limiting factor isn't usually shelf life of the drives.

    I work for a company selling enterprise storage, and we expect our customers to use the entire write endurance of an SSD in three years. The reason is that SSD is still quite expensive per terabyte, so it's only used for workloads which really need it. That means database indexes, file caches etc. These generally involve high data turnover and even with wear levelling the whole lifetime will be used up in 3 years.

  • by BLToday ( 1777712 ) on Monday May 25, 2015 @11:30AM (#49768803)

    Lost another SSD over the weekend. Crucial m4 512GB. Lost detection of the drive by the computer (Win 7 desktop), plugged it in through a USB adaptor and it's still not detected (Windows and Mac). That's 3 in the last 18 months.

    RIP
    Muskin Chronos 120GB (Windows 7 laptop)
    Crucial m4 512GB (MacBook Pro 2012)
    Crucial m4 512GB (Windows 7 desktop)

    That being said I run everything on SSD: 2 HTPC, 2 desktops, 2 MacBooks, 2 Windows laptop.

    I can't find the common factor that causes the failures. It would just be working one day, then next day fail detection by the computer and it's all gone.

    • by Anonymous Coward

      Stop buying cheapo SSD's...

    • by tlhIngan ( 30335 ) <slashdot@worf.ERDOSnet minus math_god> on Monday May 25, 2015 @11:58AM (#49769007)

      Lost another SSD over the weekend. Crucial m4 512GB. Lost detection of the drive by the computer (Win 7 desktop), plugged it in through a USB adaptor and it's still not detected (Windows and Mac). That's 3 in the last 18 months.

      RIP
      Muskin Chronos 120GB (Windows 7 laptop)
      Crucial m4 512GB (MacBook Pro 2012)
      Crucial m4 512GB (Windows 7 desktop)

      That being said I run everything on SSD: 2 HTPC, 2 desktops, 2 MacBooks, 2 Windows laptop.

      I can't find the common factor that causes the failures. It would just be working one day, then next day fail detection by the computer and it's all gone.

      How do you turn off your computers? And do your storage drivers put the drive into low power mode prior to turn off?

      The thing with SSDs (and you don't appear to use the OEM ones like Samsung, Intel or Toshiba) is power failure can be deadly. Modern SSDs are fast and because SATA3 is a bottleneck, many sacrifice speed for data protection (if you can do 1GB/sec, and SATA3 is limited to 540MB/sec, you can sacrifice 40% of the speed in the name of data safety).

      SSDs require a bunch of tables to work - the tables manage the flash translation layer software (the software that maps physical flash blocks to what the drive exposes, including wear levelling, TRIM and other features). In data safe SSDs, those flash tables are usually write-through cached so updates are committed to media, and so media always contains a consistent table. (There are tricks done to ensure that even if the table is partially written, there's a recovery table which is an older version. Think of it like a journaling file system).

      Older SSDs cached the tables in RAM, wrote to them in RAM, then relied on a bank of capacitors to let them flush the tables from RAM to media on power loss.

      Some SSDs cache them into RAM, and don't handle power failure, which can result in failures if the tables are corrupted.

      The GOOD news is there's often a way to recover them - if you do an ATA_SECURE_ERASE command, it forces the SSD to reset the tables to a blank state and will often get them completely operational again, albeit losing all your data.

      • Standard shut-off procedure. Windows 7 => Shutdown in Start Menu. Mac OS X => Shutdown in Toolbar.

    • Never lost a single SSD. That's 4 in the last two years.

      Anecdote annihilated.

      • So, you somehow think your anecdote of not seeing something cancels someones anecdote of it happening?
        I can only hope you never ever have anything to do with statistical analysis.

        Ever seen a rape? Well, I guess they never happen then! (get the idea moron?).

        The original poster never claimed his anecodes proved anything..

  • by Anonymous Coward

    The problem is hard drives have different properties that aren't understood by pretty much everyone, including hardcore nerds.

    SSD's have not been around long enough to actually say they will last X amount of time because even the least-reliable TLC based SSD's have only been around less than a decade. Flash still works by "holding a charge" despite what kind of nonsense gets thrown about. It's not like a CD-R/DVD-R where there are holes burned into a dye substrate and the data will last forever if you can p

    • by swb ( 14022 )

      Do SAN vendors intentionally mix production runs of drives when they ship them?

      I would kind of expect them to, and it might explain why I've never seen a group of drives bought at the same time (installed in a server or SAN) fail as a group.

      Although I would kind of expect some logistical challenges if I was a SAN vendor trying to keep inventories of multiple production runs in stock for populating new SANs, especially when some single unit devices can ship with as many as 24 drives. Keeping a half-dozen un

  • Even though this sounds reassuring, I started creating par2 checksums for my family pictures (and then back up the whole bunch, of course).

    If you run OS X on the desktop, it installs nicely via Homebrew:

    $ brew install par2

    Then use as follows:

    $ cd familypics
    $ par2create par2file *

    And to verify:

    $ cd familypics
    $ par2verify par2file.par2

    It takes about 5% of extra storage. If you run Linux, you can get that back by using btrfs and mounting it compressed.

  • Although Alan Cox probably has something to say about SSDs, I don't think anyone bothered to contact him for this article.

Avoid strange women and temporary variables.

Working...