Slashdot Mirror

← Back to Stories (view on slashdot.org)

New Seagate Shingled Hard Drive Teardown

Posted by Soulskill on from the made-out-of-actual-shingles dept.

New submitter Peter Desnoyers writes: Shingled Magnetic Recording (SMR) drives are starting to hit the market, promising larger drives without heroic (and expensive) measures such as helium fill, but at a cost — data can no longer be over-written in place, requiring SSD-like algorithms to handle random writes.

At the USENIX File and Storage Technologies conference in February, researchers from Northeastern University (disclaimer — I'm one of them) dissected shingled drive performance both figuratively and literally, using both micro-benchmarks and a window cut in the drive to uncover the secrets of Seagate's first line of publicly-available SMR drives.

TL;DR: It's a pretty good desktop drive — with write cache enabled (the default for non-server setups) and an intermittent workload it performs quite well, handling bursts of random writes (up to a few tens of GB total) far faster than a conventional drive — but only if it has long powered-on idle periods for garbage collection. Reads and large writes run at about the same speed as on a conventional drive, and at $280 it costs less than a pair of decent 4TB drives. For heavily-loaded server applications, though, you might want to wait for the next generation. Here are a couple videos (in 16x slow motion) showing the drive in action — sequential read after deliberately fragmenting the drive, and a few thousand random writes.

93 comments

  1. Interesting idea, nasty downsides by complete+loony · · Score: 1

    Sounds like an interesting idea to increase capacity, but the downsides are huge. This really needs filesystem level optimisations to get any performance out of it. For rarely modified bulk storage, this should be fine.

    --
    09F91102 no, 455FE104 nope, F190A1E8 uh-uh, 7A5F8A09 that's not it, C87294CE no. Ah! 452F6E403CDF10714E41DFAA257D313F.
    1. Re:Interesting idea, nasty downsides by cheater512 · · Score: 3, Insightful

      Oh look here are some SSD optimised file systems already. Incidentally they apply to these drives rather well.

    2. Re:Interesting idea, nasty downsides by YukariHirai · · Score: 1

      Yeah. Also, this would be way more prone to data loss when there's a sudden power cut than a more traditional hard drive.

    3. Re:Interesting idea, nasty downsides by Yomers · · Score: 1

      If they'll add ~32 gigs of ssd cache for delayed writes (and faster reads as a bonus, and reliability in case of power failures) - it'll be overall winner.

    4. Re:Interesting idea, nasty downsides by fraxinus-tree · · Score: 2

      ... except in some cases where these filesystems make some wrong assumptions about the wear leveling mechanism below and perform worse than a generic, disk-oriented fs.

    5. Re:Interesting idea, nasty downsides by fraxinus-tree · · Score: 3, Insightful

      The main downside is the disk becoming much more algorithmically complex (read: bug-prone) for a less than a radical improvement in performance.

    6. Re:Interesting idea, nasty downsides by Cramer · · Score: 2

      Because that's worked so well for the seagate hybrid drives. (hint: no it doesn't)

    7. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 0

      I stopped buying Seagate after they lowered their long warranty period, because "it was an industry standard."

      Which company out there has the longest warranty period? Toshiba maybe?

      Disk drives are only for large or long term storage going forward, like Tape drives were. Even if these drive out perform other hard drives in performance, can they catch up to SSD in seek times, and if not, where will they fit into the price / performance scale (in different setups)?

      They may have some technology, but they aren't going forward in a good direction. Short term profit maybe, but it may stunt their long term growth.

    8. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 0

      This also has the down side that writing to the same sector does not immediately overwrite the previous data on the disc, since the write will go into the persistent cache instead and will then eventually overwrite the original data when the garbage cleaning occurs (I think the presenter says ~15 minutes).

      But again, this is fine, because Seagate knows their customers buy bigger drives.

    9. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 3, Insightful

      Why not just go full SSD?

      For much the same reason we still use tape. Sometimes read/write speed isn't nearly as important as $/TB.

    10. Re:Interesting idea, nasty downsides by Peter+Desnoyers · · Score: 4, Insightful

      Drive performance is kind of like airplane legroom - people gripe about it, but in the end they ignore it and buy the cheap ticket.

      Shingled drives aren't better - they're bigger, and that's what people pay for. WD's 10TB helium drive is shingled, and I would guess that every drive over 10TB will be shingled for the foreseeable future. By the time HAMR and BPM come out, SSDs will probably have killed off the high-performance drive market, so those technologies will probably be released as capacity-optimized shingled drives, too.

    11. Re:Interesting idea, nasty downsides by AK+Marc · · Score: 1

      SSD for boot/OS/swap, and slow spinner for data gives 99% of the performance for 99% of people. And cheap spinners are much cheaper than cheap SSDs. Sadly, there was a time, 2-5 years ago when you could find laptops with spare mSATA slots, and a spinner in them. Put an SSD in the mSATA slot, and biggest drive possible in the spinner slot, and get huge storage for cheap cost.

      Though, one of the ones I got then, had the mSATA already holding a 20GB SSD, set up as a cache for the slow spinner. It runs surprisingly well, especially if you run the same things repeatedly. 100% of the performance of SSD for 95% of what you do. And cheaper than the 256GB/1TB I am running in my laptop.

      --
      Learn to love Alaska
    12. Re:Interesting idea, nasty downsides by Minupla · · Score: 1

      This. Even auditors have stopped blinking at me when I say "No tapes, we just have another data center like this one and a big ol pipe and XYZ data backup solution attached to the disks at the other end."

      When auditors stop blinking, you know it's hit mainstream.

      Min

      --
      On the whole, I find that I prefer Slashdot posts to twitter ones because I don't get limited to 140 chars before
    13. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 1

      Security. A data center that uses SANs and devices like Avamars is just asking for an intruder to log on and purge their devices of all data. All it takes is something like blowing the disk images (target LUNs) away, or random dd-ing garbage onto their sectors.

      With tape, a bad guy will have to have to have physical access to grab a batch, stuff it in the tape silo, then tell the silo to sequentially erase every single unit there.

      Problem with SANs/NASs and backups on hard drives is simple: They are great for dealing with environmental hazards, but all someone has to do is hack root or Administrator rights, purge the damn thing, and it is gone. Replicas to other geographic regions? The deletion and overwriting gets propagated.

      This is one big IT problem: RAID != backup. Snapshots != backups. The last Sony hack shows that the bad guys are starting to destroy stuff, and ransomware is just in its infancy. There is even ransomware out there that exploits domain privs and actually attacks Active Directory to encrypt schema and the user information, which would shut an entire organization down. Without backups, that organization is just plain fucked.

      So, if one wants to ignore basic security, fine, just use deduplication and an array of hard drives. However, if one wants to be able to handle corporate regulations and show that data is archived, tape is the way to go, as it is cheap, reliable [1], and can be used as WORM media.

      [1]: In one organization, I worked with tens of thousands of LTO media pieces. The number that failed with soft errors in the years I worked there, I could count on one hand (no, not in binary, so think less than 5), and I never had a hard error from a tape.

    14. Re:Interesting idea, nasty downsides by AK+Marc · · Score: 4, Interesting

      I've worked for multiple Fortune 500 companies. All used spinning tape. And nearly everywhere I've worked has used tape. It's cheaper and easier to buy tapes. You sound like a salesman, but I've never seen the numbers work for an off-site storage, Tapes are cheaper than hard drive storage, and more controlable (having them physically stored where you want, restoring only what you want, good for lawsuits).

      --
      Learn to love Alaska
    15. Re: Interesting idea, nasty downsides by DigiShaman · · Score: 2

      There is even ransomware out there that exploits domain privs and actually attacks Active Directory to encrypt schema and the user information, which would shut an entire organization down.

      All users should be a member of the Domain Users group along with other security groups isolated to the share access they need. Only IT sysadmin should be using AD accounts that are Domain Administraors and Schema for utilitarian reasons only! Even sysadmins should have their own Domain Users AD account. Sadly, many sysadmins still conflate the roles with one account; very bad!!

      --
      Life is not for the lazy.
    16. Re:Interesting idea, nasty downsides by mlts · · Score: 1

      If this technology becomes commonplace, I can see this used as a third tier of storage, between normal HDDs and tape, either used as a live landing zone until it gets copied to tape, or perhaps used in concert with a higher tier landing zone, where the data is written onto the platters already deduplicated, aimed at staying there for long term storage.

      Even operating systems are starting to become storage tier aware. Windows Server 2012R2 can autotier between SSD and HDD, and Windows Server 10 has improved on that.

      What would be ideal would be some drive maker to come up with some way of creating cartridges of drives, in a RAID configuration. Something like iMation's RDX... except each cartridge having 2+ drives in them, so each unit has not just RAID, but can be scrubbed to find and correct bit rod when the garbage collector goes about its business. This would completely replace tape, but also offer the benefits of tape, as in being offline and out of the reach of a bad guy doing "rm -rf/" on every SAN and NAS he can find with his newfound domain admin rights.

    17. Re:Interesting idea, nasty downsides by Smauler · · Score: 1

      I had reason to boot up an old XP system I set up for my mum to play games recently (my graphics card died). I timed the boot : 13 seconds from bios to desktop.

      I used to have 15 second boots on my system with Vista too, it's now closer to 30 with all the cruft I now have.

      Does anyone need much quicker than that on a desktop system?

    18. Re: Interesting idea, nasty downsides by Billly+Gates · · Score: 1

      I would say 98%

      I have never heard of one who doesn't. One client still uses punch cards. if it ain't broke don't fix it!

      --
      http://saveie6.com/
    19. Re:Interesting idea, nasty downsides by AK+Marc · · Score: 1

      Are you talking about SSD or HDD?

      --
      Learn to love Alaska
    20. Re:Interesting idea, nasty downsides by Smauler · · Score: 1

      Just simple HDD. With the Vista installation I've striped the hard drives.

      I personally was astonished how quick XP booted. I always remembered a dearth of incomprehensible chuntering. It seems I set up a decent system. I repeat, 13 seconds to desktop.

    21. Re:Interesting idea, nasty downsides by camperdave · · Score: 1

      I switched my laptop from a HDD to a SSD. Before the changeover, it used to be that once the login screen came up, I could log in. Now it boots much faster, but now when the login screen comes up, I have to wait for 30 seconds or so until the keyboard becomes responsive. (And before you suggest it, I have updated the bios to the latest version with no effect.)

      --
      When our name is on the back of your car, we're behind you all the way!
    22. Re:Interesting idea, nasty downsides by Enter+the+Shoggoth · · Score: 4, Interesting

      Who still uses tape? Seriously, no data centric company on the planet still uses tape, its easier and cheaper to throw a bunch of large drives and a big fat pipe to offsite storage than deal with a tape robot.

      People still using tape are doing so because they haven't moved on and like pain or are just ignorant of the alternatives.

      Google, probably the most "data centric" company on earth, that's who!

      http://highscalability.com/blo...

      http://www.theregister.co.uk/2...

      --
      Andy Warhol got it right / Everybody gets the limelight
      Andy Warhol got it wrong / Fifteen minutes is too long.
    23. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 0

      This is one big IT problem: RAID != backup. Snapshots != backups.

      Correct, but this is not a problem.

      You had backup on tape before? Well, you can have backup on a disk array too. Of course, a backup won't be a snapshot - it will be a backup with several generations and ability to fetch back files that got deleted in error. The only difference between backup on disk and backup on tape is that you don't wait so much to get stuff back from your disk-based backup. Tape has horrible seek time.

    24. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 0

      Weird timings. My machines boot in 10s, and I can log in immediately when the boot screen comes up. 1s more, and I'm in. What wrong are you doing to get 30s of unresponsiveness?

    25. Re:Interesting idea, nasty downsides by Aqualung812 · · Score: 4, Insightful

      No tapes, we just have another data center like this one and a big ol pipe and XYZ data backup solution attached to the disks at the other end."

      So, you're not protected against malicious destruction of data? Pretty sure that requires an air gap.

      --
      Grammer Nazis - I mod you "troll" unless you actually add something on-topic. Yes, I know I have mispellings in my sig.
    26. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 1

      As someone whose job it is to support the data storage needs of some of the _largest_ data-centric companies on the planet, I assure that they *all* use tape and that you have *no* idea what the fuck are you are talking about.

      But this is Slashdot, so what did I expect?

    27. Re:Interesting idea, nasty downsides by Anonymous Coward · · Score: 0

      its easier and cheaper to throw a bunch of large drives and a big fat pipe to offsite storage than deal with a tape robot.

      There are limits to this - we are a relatively small shop, so we're only backing up ~25TB a night, but even that taxes the amount of bandwidth we can get at our location. A "big fat pipe" is not always going to be the answer, especially once you start getting into really large amounts of data.

    28. Re:Interesting idea, nasty downsides by rs79 · · Score: 1

      Raid drives are transparently removable and can be placed in a bank vault.

      --
      Need Mercedes parts ?
    29. Re:Interesting idea, nasty downsides by Aqualung812 · · Score: 1

      I somewhat agree, because "transparent" isn't exactly true. Popping a drive and putting a blank one in means a resync spike as the new one is filled. With multi-TB drives, that resync time isn't exactly trivial, and a drive failure during resync is certainly a possibility, as there is more activity.

      Going with something that can handle multiple failures is helpful, such as RAID6, but you're still increasing risk just by performing your backup operation. In fact, part of the reason RAID6 exists is that there were enough cases of a RAID5 drive failure causing a second failure during the rebuild, and losing the whole array.

      Using a removeable drive as a target for backup software, and treating is just like a tape, though, is a very effective solution and takes care of air gap concerns.

      The parent post didn't explain any procedure other than a DC sync, which only protects against physical failure at the main DC.

      --
      Grammer Nazis - I mod you "troll" unless you actually add something on-topic. Yes, I know I have mispellings in my sig.
    30. Re:Interesting idea, nasty downsides by Minupla · · Score: 1

      Depends on your risk scenario planning. But yes, it does. A full rundown of our data integrity program would exceed the tl;dr scope on Slashdot, as well as violating NDAs :).

      In general though I'd point out that disk based vaulting technologies have advanced considerably in the last few years and if I were providing advice to someone I'd point out that there are cloud based solutions which are write-only type solutions if your risk tolerance permits the use of third parties to store your data (e.g. CrashPlan). Avamar may also be an option depending on costs and resources.

      That's where the professional part of IT professional comes in. You weigh your risks and have an honest discussion with your partners on the business side without fear mongering and you all decide on what your risk tolerance is, and have those discussions regularly (hint: Google's risk tolerance was different when they were in a garage then as a publicly traded company :)).

      --
      On the whole, I find that I prefer Slashdot posts to twitter ones because I don't get limited to 140 chars before
    31. Re:Interesting idea, nasty downsides by camperdave · · Score: 1

      As I said, I upgraded the HDD to an SSD. Obviously, the extra time needed to boot via the hard drive masked the keyboard lockout.

      --
      When our name is on the back of your car, we're behind you all the way!
    32. Re:Interesting idea, nasty downsides by sjames · · Score: 1

      You can do 3 drives /mirror to considerably reduce your risk. Even when you pull a drive for the vault, you still have a mirrored volume.

  2. backup storage by iamwhoiamtoday · · Score: 1

    This seems that it would be beautiful for a backup server that backs up every few weeks.

    1. Re:backup storage by bughunter · · Score: 1

      It also sounds good for a video server. I have one attached to my PC-based DVR, with playback clients in other rooms. For 99.9% or more of the data, it's very large files (100s of MB to multiple GB) that are written once and read many times until deleted.

      However, since this server is also a backup server, its a RAID array. I wonder if this Shingled format has any effect on RAID performance. A lot of "green" drives do not work well in this RAID setup, causing stuttering video playback when they continually try to go into energy saving modes.

      --
      I can see the fnords!
    2. Re:backup storage by WorBlux · · Score: 1

      I would think so at least for drive-managed devices as writing to a virtual sector doesn't neccesarily write to that physical sector, seek and read times could be inconsistent throwing the RAID read and write algorithms into dismay. With host-aware or host controlled drives, you could probably tune for better performance matching logs in the FS and stripes on the RAID to fit evenly into the shingles tracks.

  3. Drive needles by phorm · · Score: 3, Interesting

    One thing I've tended to wonder, why have a single read-write needle on conventional drives (especially in multi-platter situations). Why not have two needles, one on either side so they can't touch.
    Alternately, why not a "track" that runs across the drive with shuttles on either side to perform the reads/writes. You could have two perpendicular tracks to increase performance

    1. Re:Drive needles by fraxinus-tree · · Score: 2

      I had the same idea long ago. It ended up that the platters and the spindle are a small part of the disk in terms of cost. So instead of such a complex setup you can just buy twice more disks and get the performance you want.

    2. Re:Drive needles by vadim_t · · Score: 3, Informative

      More than a head per side? It's been attempted, and turned out it's not really worth it. It's a lot of extra complication for not that much benefit. Heads are expensive and generate heat, so it works out to close 2X the price anyway, plus an increased change of failure. Easier and safer to just add another drive.

      These days there are SSDs too.

    3. Re:Drive needles by Anonymous Coward · · Score: 0

      > One thing I've tended to wonder, why have a single read-write needle on conventional drives (especially in multi-platter situations).

      It was tried like 30 years ago.
      Extra complexity leads to increased costs and more failures.

    4. Re:Drive needles by mlts · · Score: 1

      About 10-15 years ago, some drives used to have two read/write stacks of heads, each independent from the other. This was killed due due to people wanting cheap drives.

    5. Re:Drive needles by AaronW · · Score: 1

      I remember seeing a Conner hard drive like that many years ago. Tom's Hardware discusses it.

      --
      This post is encrypted twice with ROT-13. Documenting or attempting to crack this encryption is illegal.
    6. Re:Drive needles by phorm · · Score: 1

      Conner. There's a name I haven't heard in awhile.

      It's interesting that they also did try the sweeping-servo (similar to a CD-ROm drive) read/write head design as well, once upon a time.

    7. Re:Drive needles by Anonymous Coward · · Score: 0

      I worked with drives that had clusters of movable heads in the late 70's. The problem is that it causes aerodynamic stability problems as the heads flew over magnitic media. Two heads per platter doubled the bust rate but there still was the latency as the carrige was moved. To get around this problem drives were connected in strings with each drive stepping to the next track as each head read one rev worth of data. With large strings and buffers write speeds were very fast for the time. Sort of an early RAID. see. IBM 3350, 3370 and 3390. Huge drives for their times.

  4. Log Filesystem - as hardware? by ezrec · · Score: 1

    This reminds me of the original 'Log Filesystem' research in the 80s, back when drive geometry was known to the OS, and the OS could take steps to optimize for it.

    The Log Filesystem concept was to write all data sequentially to disk, and update metadata during idle times.

    The basic research influenced a number of filesystems, such as NetApp's WAFL, Sun's ZFS, Linux's JFFS2, etc.

    Interesting to see the concept implemented 'in hardware'.

  5. meh by Anonymous Coward · · Score: 0

    The problem with those helium filled hard drives is they float away as soon as I take them out of the package.

  6. 144p video? by Anonymous Coward · · Score: 1

    Why is the camera a potato?

  7. How big is it? by hawguy · · Score: 1

    The summary says "Reads and large writes run at about the same speed as on a conventional drive, and at $280 it costs less than a pair of decent 4TB drives", one of the 7 links in the summary mentions a 5TB model. 5TB for the price of two 4TB drives doesn't sound that great.

    1. Re:How big is it? by jtownatpunk.net · · Score: 2

      The 8 TB drive is the one being priced.

  8. Nope..... by wbr1 · · Score: 4, Insightful
    From TFA

    The real question is whether or not Seagate can maintain similar full drive performance compared to a non-SMR drive.

    No.The real question is longevity. Per backblaze and my own anecdotal experience, Seagate drives already have a higher failure rate. Looking at this, any firmware bug or flaw could result in massive data loss of an entire 'band' if written incorrectly.

    I understand that in any environment backups are crucial, but I live in the real world. A world where small and medium size business (for good or ill) neglect IT until it bites them. At least with regular drives recovery is often possible with block for block copies, and baring that a clean room has a good chance of recovering crucial data.

    If a user has a performance need, I can suggest an SSD or SSD+HDD config with appropriate redundancy and backups. For pure space, large HDDs in an appropriate RAID or ZFS work fine. Per TFS, this is not ready for heavily drive loaded server configs yet, and i do not see a need in residential, or small biz or workstation use where other solutions are far better. To me this is currently a product looking for a solution, and one that is risky to data to boot.

    --
    Silence is a state of mime.
    1. Re:Nope..... by ckatko · · Score: 1

      While I pragmatically agree, it is not Seagate's fault nor burden to take care of companies that don't treat their data as important. In the same way, it's not a clothes-washer company's fault if you never change the lint trap and it catches on fire. (causing 15K fires a year in the US)

      There are clear, defined, industry standard ways to use a product and if you refuse to do so because you are a cheap and lazy, the ramifications are solely your own.

      Can't afford to replace large hard drives? Get smaller ones, or play the game of data roulette.

    2. Re:Nope..... by wbr1 · · Score: 1

      I never said it was Seagares fault or burden. A users actions are thier own. That said Seagate does have a much higher failure rate, and this tech could greatly complicate data recovery in a failure scenario. Given this I can not in good conscience recommend it to my customers.

      --
      Silence is a state of mime.
    3. Re:Nope..... by Anonymous Coward · · Score: 0

      Pretty much this as well.

      Any experience I have had with anything Seagate has always been a disaster.
      I still have drives from the 90s that outlive their metallic trash.

      Friends don't let friends buy Seagate.
      You get what you pay for.
      If you want to make huge arrays very quickly and cheaply without too much care for the data since you have reliable backups, that is the only use-case for Seagate.
      And even then I still wouldn't.

    4. Re:Nope..... by Anonymous Coward · · Score: 0

      Per backblaze and my own anecdotal experience, Seagate drives already have a higher failure rate.

      I would qualify that to "Seagate's 3TB drives already have a higher failure rate" - I have had over 50% failure on those over 2 years, but their 2TB & 4TB drives seem fine.

    5. Re:Nope..... by guruevi · · Score: 1

      It's actually due to a bad firmware bug (I've replaced 12 of them in the last few months) which still doesn't seem to have fixed the issue completely.

      If you issue a SMART command at the exact same time something is writing to the disk, data is read back corrupted (the read times out or something, the controller kicks it out due to inactivity). If you are monitoring your SMART eg. every hour, you would eventually start having 'bad disks' in a seemingly random fashion. In my case, there could be months between bad disks and other times 2 or 3 would fail at once.

      Disabled SMART monitoring on my controller -> no more bad disks (although once a disk has encountered the bug, it will continue to report faulted through SMART).

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
    6. Re:Nope..... by goarilla · · Score: 1

      So you disable one of the most important technologies to monitor your disks.
      Aren't you afraid you are sticking your head in the sand that way.
      What's left: log messages, iostat output ?

    7. Re:Nope..... by guruevi · · Score: 1

      Yes, to both.

      SMART has in my career not made one difference to notice whether a drive was bad. SMART only shows things after the drive has gone bad already. A rise in the number of read/write retries in a large disk array is immediately noticeable and you don't need SMART to tell you that a drive is bad (and just the timeouts will usually cause the bad drive to be kicked out). Bad data is immediately noticeable (and likewise, will cause the drive to be kicked out) if you use ZFS, even before SMART catches on to an ECC fault.

      If they return any relevant SMART data at all, most of the 'important' data is locked behind manufacturer specific codes. SMART is good for desktop drives, useless otherwise.

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
  9. pOINTLESS vIDS by DrunkenTerror · · Score: 1

    Those two videos have less pep than an empty soda can.

  10. Because you're an idiot? by Anonymous Coward · · Score: 4, Informative

    Find me an 8TB SSD that is even within spitting disttance (hell, within ICBM distance) of $300 and you win the prize, otherwise the suggestion is sless. Hint: Not today, not next year. Possibly this decade . The cost has come down a ton, but it was absolutely astronomical before.
    The systems I'm buying now will be obsolete by the time SSD can even think about touching hard drives in terms of capacity per $. Typically, the ONLY reason to go full SSD now for large storage capacities is because you absolutely need the performance and are willing to pay essentially "whatever it costs" (at least 8x+ the price) because it's that important to get the IOPS. Maybe by the end of next year we'll get it down to "only" 4x the price (not counting that though because price per GB for large capacity hard drives still continutes to fall, balancing out a part of the cost reduction in SSDs).

    1. Re:Because you're an idiot? by sabri · · Score: 1

      The systems I'm buying now will be obsolete by the time SSD can even think about touching hard drives in terms of capacity per $. Typically, the ONLY reason to go full SSD now for large storage capacities is because you absolutely need the performance and are willing to pay essentially "whatever it costs" (at least 8x+ the price) because it's that important to get the IOPS. Maybe by the end of next year we'll get it down to "only" 4x the price (not counting that though because price per GB for large capacity hard drives still continutes to fall, balancing out a part of the cost reduction in SSDs).

      You're right, but also somewhat wrong.

      "Raw capacity" is indeed quite expensive. However, the increased speeds of flash have made it possible to provide in-line data reduction services. Data reduction is a widely used term for two techniques: de-duplication and compression. This works on a block level. When the host writes a block, the system will look in its table to see if that block has already been written. If so, it will simply write a pointer. If not, it will be forwarded to the compression engine and stored. In the storage industry, a data reduction ratio of 1:6 is accepted as generally achievable, with higher rates possible on large virtualization clusters. This means that the effect cost per GB is reduced dramatically and even reaches the price of high-end disks. So your 80TB raw capacity array will store ~300TB of data, depending on the data reduction ratio.

      On top of that, since flash performs so well, other features such as continuous data protection by (a)synchronous replication are possible with very little performance degradation.

      If you still doubt this technology, go have a look at the technological advances in hard drives, versus the technological advances of flash in the last 5 years. Hard drive vendors have very little to innovate while flash/rram are at the lower end of a hockey stick figure when it comes to innovation and price reduction. Within the next 5 years, flash/rram will replace disk, both in the enterprise as well as the consumer market.

      I'd love to see your all-disk array do 2M IOPS, something that all flash arrays are capable of today. Again, at the price that (with dedupe and compression) comes darn close to your disks. Even legacy storage vendors are increasingly investing in solid state technology. Investing in disk is equal to investing in Greek government bonds.

      --
      I'm not a complete idiot... Some parts are missing.
    2. Re: Because you're an idiot? by DigitAl56K · · Score: 1

      Compressing data for important backups is not a good idea. A one-bit error and entire files or worse become garbage.

    3. Re:Because you're an idiot? by Anonymous Coward · · Score: 0

      Again, at the price that (with dedupe and compression) comes darn close to your disks. Even legacy storage vendors are increasingly investing in solid state technology. Investing in disk is equal to investing in Greek government bonds.

      Disks are not some investment to pay off in a year or two. They are storage space to be used almost immediately. I can't wait a couple more years for a new technology to come in line with what I need now. And in the case of things like hard drives, a change in technology is in many ways a drop in replacement, with no sunk costs of older components. Especially in cases like at my job where storage solutions are purchased by the rack, and we've been do as much compression to data as possible for years already.

    4. Re: Because you're an idiot? by Anonymous Coward · · Score: 0

      also don't get how people will apply such compression to SSDs but for some reason these won't apply to regular HDD?

      So 6:1 compression on a 1TB SSD gives you a 6TB SSD for $500.

      A 6:1 compression on an 8TB HDD gives you 48TB for $300.

      So to get SSD to HDD at same compression rates we're sitting at... $4000 vs $300.

    5. Re:Because you're an idiot? by tlhIngan · · Score: 1

      If you still doubt this technology, go have a look at the technological advances in hard drives, versus the technological advances of flash in the last 5 years. Hard drive vendors have very little to innovate while flash/rram are at the lower end of a hockey stick figure when it comes to innovation and price reduction. Within the next 5 years, flash/rram will replace disk, both in the enterprise as well as the consumer market.

      I'd love to see your all-disk array do 2M IOPS, something that all flash arrays are capable of today. Again, at the price that (with dedupe and compression) comes darn close to your disks. Even legacy storage vendors are increasingly investing in solid state technology. Investing in disk is equal to investing in Greek government bonds.

      SSDs follow Moore's law - double the transistors, double the size. You can cheat with TLC and more, but the smaller cells aren't helping. Hard drives aren't restricted to Moore's law which is why their prices can tumble.

      As for 2M IOPS, if you're doing that, you need SSD, no question. But if you're doing movie editing where raw speed and basically long stripes of writes take place, you may be better off going with hard drives because you value storage over IOPS. When a CGI scene can start to require several TBs of assets to be loaded for rendering (seriously, the modern movie is already pushing 200+TB of assets just for CGI), 2M IOPS isn't so critical.

      And if your business needs 2M IOPS and 8TB of storage at 2M IOPS, there are companies willing to accommodate you, and sure you're going to be paying tens of thousands, but if you're pushing that sort of workload, that's probably well worth the investment.

      Anyhow, I can see SMR drives serving as yet another tier in the storage hierarchy - you have the SSD for the hot files, the cooler files get put on a huge regular storage array, and the SMR drives serve as cold files long term near-availability storage, and then you have tape for offline storage. The SSD gets used for the files where they're busy so it's fast, the hard drives serve up older files that are referenced only once in a while, and the SMR serves as a archive pool for files accessed rarely or not at all, just before being dumped to tape.

      Or, perhaps SMR drives can be used in high capacity applications where it's mostly read only. Think like a digital download store (music, movies), or Netflix. Updating a single file on an SMR is painful, but if you're going to write a big movie to the disk and everyone's going to be reading from it, a large cheap SMR drive fits the bill because it's going to be read from far more than written to.

    6. Re: Because you're an idiot? by amorsen · · Score: 1

      If you block-level deduplicate a file on an HDD, and even a small fraction of the blocks from an otherwise sequential file are replaced by pointers, you have completely destroyed read performance for that file. Block-level deduplication is not a viable technology on hard drives except for very specific use patterns.

      In contrast, almost no workloads suffer when doing block-level deduplication on an SSD.

      --
      Finally! A year of moderation! Ready for 2019?
    7. Re:Because you're an idiot? by goarilla · · Score: 1

      Find me an 8TB SSD that is even within spitting disttance (hell, within ICBM distance) of $300 and you win the prize

      I think that will be available in 6 years.
      Bear in mind though that this SMR drive is an exception. The other manufacturers are still producing 4-6 TB drives for that price.

    8. Re:Because you're an idiot? by tlhIngan · · Score: 1

      Find me an 8TB SSD that is even within spitting disttance (hell, within ICBM distance) of $300 and you win the prize

      I think that will be available in 6 years.
      Bear in mind though that this SMR drive is an exception. The other manufacturers are still producing 4-6 TB drives for that price.

      Well, $2000 gets you spinning rust 8TB HDDs from the other guys with fancy helium and no SMR right now. You can easily buy 4x 1TB SSDs for that price.

      But you can bet even those hard drives will probably drop to $300 range soon enough (probably a couple of years). Moore's law will see the price halve in that time, so you can get an "8TB" SSD (8x 1TB) for $2000 after the spinning rust version dropped to $300-ish.

    9. Re:Because you're an idiot? by goarilla · · Score: 1

      Maybe, But I doubt if the demand for bigger spinners is high enough for that.
      The prices haven't really come down since the 2010 flood and we lost Samsung as a competitor.
      8 TB might be the last (premium) drive of the spinners.

    10. Re: Because you're an idiot? by guruevi · · Score: 1

      There are also few workloads where block-level dedupe is any better than block-level compression. Most people don't store the same data over and over (yes, disk images of virtual machines, but even there the images stop converging soon after deployment).

      With block-level compression I get ~30% for what is basically 'random' data (user home directories and medical imaging data, 100TB). Block-level deduplication would only give me ~15%.

      Compression does internal de-duplication already (one of the easiest ways of compressing is finding duplicates and eliminating them, the hardest part of compression is finding the best way of doing it).

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
    11. Re:Because you're an idiot? by guruevi · · Score: 1

      The demand won't stop, we'll always need more cheap storage and until we can make chips as fast and cheap as we can coat a piece of metal with magnets, we'll have spinning rust. The spinning rust will get heavier and slower though as we are nearing the limits of what is physically possible but that's another decade or 2 away from now.

      Eventually SSD will outperform hard drives in cost on other levels (energy usage, price per cubic feet) just as hard drives did to tape and their robots (robots are too expensive to operate and take up too much space for anyone that has their infrastructure in less than half a rack).

      Look at tape, even it's demand has not stopped and they're still inventing larger and larger versions because if you can spare a rack to hold your robot, you could hold several PB's worth of data whereas a rack of hard drives barely holds 1PB.

      --
      Custom electronics and digital signage for your business: www.evcircuits.com
  11. I want by kuhnto · · Score: 1

    1 minute and 46 seconds of my life back. Please?

    --
    "A 'person' is smart. 'People' are dumb, panicky animals and you know that."
    1. Re:I want by Anonymous Coward · · Score: 0

      And yet you clicked on it...

    2. Re:I want by wonkey_monkey · · Score: 1

      That'll learn yer not to read YouTube descriptions.

      --
      systemd is Roko's Basilisk.
  12. firmware by Anonymous Coward · · Score: 0

    Does it come with the latest NSA firmware release?

    1. Re:firmware by hcs_$reboot · · Score: 0

      Do we have any tangible proof of that, like a reverse engineered firmware of a tampered HD? Showing exactly to what extent a firmware actually spies on its host computer?

      I really wonder if those who started that rumor don't want us to be more worried about our privacy that we need to be.

      --
      Slashdot, fix the reply notifications... You won't get away with it...
  13. garbage collection? by Anonymous Coward · · Score: 0

    What if you never delete anything, then there's no garbage.

  14. Lost me at "Seagate" - I just don't care by Anonymous Coward · · Score: 0

    Lost me at "Seagate" - I just don't care.e
    They've screwed their reputation as far as I'm concerned - like in 2007. Tried them again in 2014. Screwed again.

    If they last until 2024 ... I _might_ try seagate again. Maybe.

  15. Mismatch by SuperKendall · · Score: 1

    SSD for boot/OS/swap, and slow spinner for data gives 99% of the performance for 99% of people.

    That would be great except 99% of people don't want more than one disk.

    Hell, *I* don't want more than one disk, and I can ably manage them. But there's no way I can afford the SSD it would require to store everything I have (never mind the backups).

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
    1. Re:Mismatch by AK+Marc · · Score: 1

      99% of people don't care. I set up my mother's computer with the OS on C and everything else (and all system defaults) to D, and she didn't even know she had multiple disks. My wife didn't know she had two disks (one SSD cache and one HDD), she just knew it worked and wasn't too slow.

      --
      Learn to love Alaska
    2. Re:Mismatch by hab136 · · Score: 2

      There are several solutions that take an SSD and hard drive and present one logical drive to the OS.

      Windows: Intel Smart Response
      OS X: Fusion Drive
      Linux: bcache/flashcache

    3. Re:Mismatch by justthinkit · · Score: 1

      The idea of setting up my mom with an SSD is pretty hilarious. I only just got an SSD, to speed up editing 1TB audio files. Only got a 3 times speed up, FWIW (on an, I suppose, old Q6600 system). Oh, did I mention that Nana is 90?

      --
      I come here for the love
    4. Re:Mismatch by AK+Marc · · Score: 1

      I'll never be without one again, even if it's just a tiny one for Cache, or the minimum size for an OS/Swap/boot partition. Just that little bit makes a massive difference, and makes almost all the difference. The cost isn't that high.

      --
      Learn to love Alaska
    5. Re:Mismatch by justthinkit · · Score: 1

      I agree. For me, with audio editing, an SSD turns loads & saves into "word processor speed" actions, instead of "I have time to get a coffee" ones.

      I haven't been brave enough to trust the SSD with my swap file, but am getting there. In part because the OS gets badly swapped out when you work with very large files -- stupid design. If it didn't, I wouldn't bother.

      Similarly, I will be moving browser cache directories pretty soon. Opera (v12) seems to get particularly sluggish when things are all swapped out. Strangely, Firefox is almost no different. Ah, the voodoo science of modern computers.

      --
      I come here for the love
    6. Re:Mismatch by justthinkit · · Score: 1

      Make that 1GB...

      --
      I come here for the love
    7. Re:Mismatch by AK+Marc · · Score: 1

      Windows doesn't use swap files well. In Win, since XP, I've just turned off swap. Works much better then, just need to have lots of RAM. Other OSs, I leave it on.

      --
      Learn to love Alaska
    8. Re:Mismatch by justthinkit · · Score: 1

      Unfortch on this machine I only have 4GB of RAM, so have to have a swap file as well -- or else, random-loss-of-data pays a visit.

      BTW, I moved my Opera browser cache to the SSD a couple of hours ago. Insert VBG here.

      --
      I come here for the love
  16. amazing, for a buggy whip by Anonymous Coward · · Score: 0

    imagine, something with actual moving parts for data storage in the early-21st century.

    1. Re:amazing, for a buggy whip by camperdave · · Score: 1

      Those poor harness racers, all put out of work because nobody makes buggy whips anymore.

      --
      When our name is on the back of your car, we're behind you all the way!
  17. Danger by SuperKendall · · Score: 1

    9% of people don't care. I set up my mother's computer with the OS on C and everything else

    She sure will care if there's every a problem (and there will be a problem) with one of those drives.

    As a rule of thumb it's way better not to double someones possible failure rates if they don't know themselves how to recover from it...

    I've spent my life helping people get set up technically so they never need to talk to me again - at least not about their systems. It creates a lot less work for yourself, unscheduled works that generally comes at very bad times.

    --
    "There is more worth loving than we have strength to love." - Brian Jay Stanley
  18. Impatience by Anonymous Coward · · Score: 4, Funny

    How impatient must one be to tear down a Seagate hard drive before it breaks down?

  19. No data-overwrite-in-place = ransomware relief by davidwr · · Score: 1

    data can no longer be over-written in place, requiring SSD-like algorithms to handle random writes.

    Good, now when my clients get hit by ransomware there is still hope that the "over-written" file can be recovered.

    --
    Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.