Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror
×
Data Storage

Vibration Killing Enterprise Disk Performance? 159

An anonymous reader writes "Is vibration killing disk performance? ZDnet reports on research that a carbon fiber anti-vibration rack increased random read performance by 56% to 246% and random write [performance] by 34% to 88%. Vibration is a known disk problem, but this is one of the few attempts to quantify its impact — which looks to be much greater than suspected."
This discussion has been archived. No new comments can be posted.

Vibration Killing Enterprise Disk Performance?

Comments Filter:
  • Hmmm... (Score:3, Funny)

    by antifoidulus ( 807088 ) on Friday May 07, 2010 @06:34PM (#32133958) Homepage Journal
    So maybe sex in the server room IS a good idea! Not to mention all the puns you can make about racks.
    • Re:Hmmm... (Score:5, Funny)

      by Anonymous Coward on Friday May 07, 2010 @06:46PM (#32134042)

      Get off my Rack?

      In Soviet Russia, rack vibrates you?

      All your rack.....?

    • And to think, people wonder why geeks don't get laid more.

      • Yeah, because those people in school that got lots of girls didn’t constantly make jokes about sex, while the geeks constantly talked about geeky stuff...

        It’s not whether you make the jokes, but how you act while doing so. Confident, or ashamed?
        Most geeks seem to be ashamed of her noticing that they would want sex with her.
        While the playboys see it as a compliment, making her special, because amongst a thousand girls, he chose her.

        And what most geeks don’t seem to know: The can see that. T

        • Yep, that's totally true. Confidence is one of the keys to attracting women. As someone who suffered from depression and then recovered, I can tell you the change in female interest has been extremely noticeable. I haven't gained or lost any weight, changed my looks, or done anything superficial, but the interest has certainly heightened. I do joke about how we nerds never get laid, but, yes, we do get laid, and we do get women.
        • by http ( 589131 )

          Oh, and don't forget that not one girl was ever self-conscious, ashamed, or interested in something other than sex. Your attempt at sociology fails. Hint: sexist generalizations contributed to your post's downfall.
          Males have no monopoly on botching the interpretation of social cues.

    • shouting at disks (Score:4, Interesting)

      by Anonymous Coward on Friday May 07, 2010 @07:55PM (#32134778)

      So maybe sex in the server room IS a good idea! Not to mention all the puns you can make about racks.

      Also, don't shout at your disks since it increases latency, as the guys at Sun Storage learned:

      http://www.youtube.com/watch?v=tDacjrSCeq4

  • Old news is old (Score:5, Informative)

    by Anonymous Coward on Friday May 07, 2010 @06:36PM (#32133976)

    Yes, already saw some crazed guy at Sun shout into an array of HDDs and it decreased performance.

    http://www.youtube.com/watch?v=tDacjrSCeq4

    • by adisakp ( 705706 )
      That datacenter is loud as hell. I wonder how much faster their disks would perform if they could get the datacenter quieter.
  • Star Trek (Score:5, Funny)

    by robvangelder ( 472838 ) on Friday May 07, 2010 @06:52PM (#32134100)

    Did anyone else think "how is Kirk going get out of this one"?

    • by lemur3 ( 997863 )

      Did anyone else think "how is Kirk going get out of this one"?

      do not feel bad, I too thought the same thing....

      damn you kirk!!

    • Duh. He'll divert auxiliary power to the inertial dampeners. There is nothing in Star Trek that can't be solved by redirecting power there or with a strategically placed redshirt.

      Hmm, now I wonder which of the two do they use when a toilet's clogged. I'm betting it's one of the situations where a redshirt dies. But diverting power to the look has a certain appeal too ;)

  • by v1 ( 525388 ) on Friday May 07, 2010 @06:53PM (#32134110) Homepage Journal

    It's a lot easier to interfere with a moving head arm than it is to mess one up that's locked on a track, so this isn't surprising in the least for vibration to affect reads that require numerous long seeks. I'm surprised it's not worse than they've found.

    Moving the head requires accelerated head stepping to top speed, stepping to close to the track, slowing down, stopping at the destination track, waiting for the head to settle, and reading an address block to find out where you managed to land. If you find you missed the track, you have to go through the whole seek process again. (usually only once more, those short adjustment hops are pretty reliable because they're lower speed) But that really hurts your single block read time.

    Add to that the fact that the "high performance" drives are making more risky higher speed track changes, which increase the odds of missing your target and make the operation more sensitive to vibration. I've written direct HDD io code before, and sure, you can up the step speed to get very nice seek time boosts, but then you start missing your track and start getting reseeks. Usually you go with the fastest that's acceptably reliable, and that puts you on the bleeding edge of having problems, where things like vibration can run you off the deep end of the bell curve.

    It wouldn't surprise me one bit if 50% of the "high performance drive" better speed is due to faster spindle speed, and the other half is faster (riskier) seek speed.

    • Re: (Score:2, Insightful)

      by pspahn ( 1175617 )

      How difficult would it be to account for this vibration performance degradation in such a case?

      If you had some kind of vibration sensor that measured vibration levels, couldn't it then talk back to code telling it to slow down because there's too much vibration?

      • Re: (Score:3, Informative)

        by v1 ( 525388 )

        should be possible to add a sensor (probably even on silicon) that can warn of higher vibration and slow the stepping, but I bet they don't do that right now.

      • by pla ( 258480 )
        If you had some kind of vibration sensor that measured vibration levels, couldn't it then talk back to code telling it to slow down because there's too much vibration?

        I think you either overestimate the frequency of typical vibrations, or underestimate the speed at which your heads can theoretically seek.

        A 15k drive can push around 180 IOPS. A typical human scream comes out at around 300Hz. 1.6 cycles really doesn't give enough of a sample to reliably detect (much less correct for) even they most pre
        • Wouldnt it be possible to start out running balls to the wall on spin-up and do a gradual back-off if you encounter reseeks due to missing tracks?
          An algorithm for this would of course need to be created but to me this seems like the sensible way to get dynamic handling of noise.

    • What they need to do is hire some mechanical engineers that deal with vibrations. There are many many ways to deal with vibrations.

      • What they need to do is hire some mechanical engineers that deal with vibrations. There are many many ways to deal with vibrations.

        You mean, like, build an anti-vibration rack or something?

      • A lot of time and money is actually spent dealing with this problem during server development.

    • by PopeRatzo ( 965947 ) * on Friday May 07, 2010 @07:37PM (#32134600) Journal

      For the computers in my music project studio, I've always used Scythe stabilizers for my hard drives, mostly to keep the sound down.

      But now that I think about it, the drives in my raid box are like 4 years old and not one has failed. I've thought about buying new drives to replace them, but as long as they're working and everything is backed up, I haven't bothered.

      I don't know if the stabilizers have had any effect on their longevity, but the little bit of care I take with vibration dampening when I build machines sure has resulted in some quiet machines.

      • I found a 6 year old Western Digital IDE drive in our DC.

        This thing has lived through loud noise, vibration, bad power, failed AC (and so high temps) but kept on truckin'

        I think you get a proper specimen once in a great while heh.

        • 6 years is hardly ancient when it comes to disks in the DC.

          I have quite a few which are 8 or 9 inside servers - I don't believe in proactive disk replacement, I believe in wholesale server/enclosure replacement at the end of the lifecycle and on-demand servicing in between (everything is at least N+1 so I can tolerate the odd failure).

          I had about 60 disks fail over the last few years that literally failed within a couple of months of the 5-year warranty running out. Seagate must have a clock in them or some

      • Re: (Score:2, Informative)

        by textstring ( 924171 )

        The same techniques that silent PC nerds use to isolate their hard drive to keep it quiet would of course help reduce vibrations (I mean, that's the point). There is the thread detailing the techniques used to suspend/isolate HDD's at http://www.silentpcreview.com/forums/viewtopic.php?t=8240/ [silentpcreview.com].

    • Moving the head requires accelerated head stepping to top speed, stepping to close to the track

      I just want to point out that hard drives stopped using stepper motors [wikipedia.org] decades ago. They've used voice coils [wikipedia.org] since, which is basically an electromagnet and strong magnet which it deflects to various positions based on the field strength; in other words, it's continuous, not discrete like a stepper motor (though they can do microstepping [wikipedia.org] as well). OK, so in a way, a voice coil is sort of like a stepper motor with

      • by Laser Dan ( 707106 ) on Friday May 07, 2010 @08:26PM (#32135024)

        Moving the head requires accelerated head stepping to top speed, stepping to close to the track

        I just want to point out that hard drives stopped using stepper motors [wikipedia.org] decades ago. They've used voice coils [wikipedia.org] since, which is basically an electromagnet and strong magnet which it deflects to various positions based on the field strength; in other words, it's continuous, not discrete like a stepper motor (though they can do microstepping [wikipedia.org] as well). OK, so in a way, a voice coil is sort of like a stepper motor with only one phase, which is then microstepped...

        Well you are half right, but so is the GP.
        The GP's description is more accurate if you replace "stepping" with "accelerating".

        The head does not move to a position based on field strength (open loop control). It is free to move on low friction bearings, the applied field strength accelerates the head. Closed loop control is needed to make it stop at the correct position.

        • Re: (Score:2, Informative)

          The head does not move to a position based on field strength (open loop control). It is free to move on low friction bearings, the applied field strength accelerates the head. Closed loop control is needed to make it stop at the correct position.

          The servo firmware engineers I worked with went to great lengths to define and maintain (in real-time) the acceleration profiles for the actuator arm. The point is that the seek algorithm is fully closed loop.

          Originally they walked the fine line between all-out performance and reliability, but later they started slowing down the seeks (on 7200 RPM drives) to make them quieter.

          • Not just quieter, but using just enough acceleration to get the head there in time for the data to come around also reduces power consumption and -- back on topic -- minimizes contributions to environmental vibration.

            One of the things you should take away from all these papers is that "enterprise" disks have hardware and software that compensates for this type of thing, while "consumer" disks don't. If you fill a rack full of Seagate Savvio 15K disks and another rack full of Seagate Barracuda XT 2TB disks,

    • The only thing surprising is that manufacturer data didn't support this big of an impact. We tried to stay three orders of magnitude below acceptable vibration limits and assumed that was adequate. While I haven't tied an accelerometer to a rack or drive chassis, it sounds like they aren't managing internal vibration as well as they need to.

      • by afidel ( 530433 )
        Actually the disk manufacturers have this information (at least Seagate does) through their former SAN division that they sold to Xiotech. One of the big improvements in performance that they were able to get in their sealed disk packs (ISE) was through vibration reduction by placing the drives on a rigid frame and mounting them in such a way that the vibration from multiple spindles canceled. The other big performance improver was stripping two decades of compatibility code out of the firmware and talking
    • Well, okay, that's an intriguing data point. Where is the seek speed set, though? In the OS, BIOS, some kind of per-drive firmware, or is it set by the hardware itself?

      It seems like if vibration is causing re-seeking in general, you could cut the seek speed for all the drives in a data center to limit the problem. I'm kind of surprised that something as precision-oriented as (what are now) super-capacity hard drives don't have any accounting for this, when if that's the solution, they should have been ri

      • by Agripa ( 139780 )

        It is not as simple as adjusting the "seek speed". When the head it tracking a specific track, the dynamic response is limited by the small signal response of the head/actuator/servo mechanism. When the head is slewing across tracks at maximum velocity, its dynamic response is limited by the large signal response. In the later case, vibration probably has no effect at all. It's when vibration interferes with the small signal response as the head reaches the destination track or is tracking that head set

    • Re: (Score:2, Informative)

      I used to work in the HDD industry, and am very familiar with the seek algorithms used. Sorry, but your description of the move algorithm is completely wrong. Modern control implementations collect information about the actual output of the system (such as position) relative to the desired output (the target position) and act on it in real-time.

      Modern HDDs (as in, anytime in the last 15+ years) have 'servo tracks' written on the disk. These are radial spokes of distance information encoded on the disk.

  • I'm just curious why SSD is also prone to the problem.

    • by MarcoAtWork ( 28889 ) on Friday May 07, 2010 @06:59PM (#32134182)

      the article does not say that it affects SSDs, but that it affects the SSD value proposition (aka, if you can spend little $$$ on carbon rack enclosures and get a significant seek performance increase, spending the large amount of $$$ to go full-ssd might not be as cost effective).

      • by Svartalf ( 2997 )

        It depends on the application as to whether you're going to improve the value proposition. Some applications will cost nearly as much as the SSDs when you apply the vibration compensation stuff to the system. Once you get there, the temperature, power, and overall lifespan issues still are present- and you're not getting rid of those any easier than you did with just the vibration problem alone.

      • by bertok ( 226922 ) on Friday May 07, 2010 @08:48PM (#32135190)

        the article does not say that it affects SSDs, but that it affects the SSD value proposition (aka, if you can spend little $$$ on carbon rack enclosures and get a significant seek performance increase, spending the large amount of $$$ to go full-ssd might not be as cost effective).

        My over one year old SSD drive can do ~6,000 under a database workload, the next-gen consumer SSD drives are reaching 60,000 random IOPS, and there are enterprise drives that can do over 150,000 IOPS with streaming speeds over a gigabyte per second.

        This is a little like saying that Hayes has released a new 56K modem that resists line noise 50% better than existing modems, which affects the value proposition of 1Gbps fibre.

        There's also no need to go "Full SSD". The newer virtualizing SAN arrays can migrate individual blocks of data between tiers of storage to place everything on the appropriate storage depending on the need for performance.

        • by CAIMLAS ( 41445 )

          There's also no need to go "Full SSD". The newer virtualizing SAN arrays can migrate individual blocks of data between tiers of storage to place everything on the appropriate storage depending on the need for performance.

          You can also do this yourself with SSDs in ZFS. It's an awesome value proposition: a half dozen rotational disks in an array with one or two caching SSDs over a SAN = awesome.

      • by CAIMLAS ( 41445 )

        Yeah, except getting that carbon fiber rack is a big logistical change: you've either got to use it with a new rack or re-shelf an entire rack.

        Nobody in their right minds is going to go "full SSD" right now. They're not going to re-rack all their servers, either (that means downtime). The realistic response is a gradual change in either regard: in that case, it doesn't make sense to go with the CF rack unless you explicitly need the added capacity of of rotational disks.

        In that case, my experience has been

    • It shouldn't, and a quick skim didn't reveal that either.

      Are you sure you didn't just misread an acronym?

    • by InsertWittyNameHere ( 1438813 ) on Friday May 07, 2010 @07:07PM (#32134264)
      I don't think it is. I always see "no vibration/movement issues" listed as a "pro" in all SSD vs HDD reviews. Plus SSDs don't have any moving parts.

      Is you comment based on this part of the article?

      SSD value. Flash SSDs have fast random read access. But disks can improve their performance by 50% through vibration damping, that changes value proposition for SSDs.

      I think the author is saying that if you can improve performance by 50% then there's less value in moving to SSDs that previously thought.

      • by adolf ( 21054 )

        I think he's just being deliberately vague. It's an article about HDD performance as it relates to vibration, not an in-depth comparison between them and SSD.

        Taken literally, the only thing he says is that the value proposition for SSD changes. But this change could be an improvement for an application's specific needs, or a detriment -- the author does not say.

        The implied message (if there even is one) is that, perhaps, one should add "mechanical vibration" to their list of things to consider when select

  • by georgewilliamherbert ( 211790 ) on Friday May 07, 2010 @06:57PM (#32134152)

    ..at SustainIT 2010, Turner had a really good analysis. Still some gaps - figuring out what frequencies hurt the most, and how individual drive types respond to what, is necessary followon. How various vendors' drive units transfer vibration from the rack into the unit, into the drive carrier, into the drive. That sort of thing. Now that the phenomena is identified, a lot more to do on it.

    At the least, keep performance sensitive drives away from large sources of environmental vibration, such as your AC unit and so forth.

    • At the least, keep performance sensitive drives away from large sources of environmental vibration, such as your AC unit and so forth.

      I used to do data acquisition work on 6,000 hp locomotives. You want to see bad disk performance...

  • by Anonymous Coward

    ...a carbon fiber anti-vibration rack...

    What is this? Pimp My Rack?; "Next up we'll add a tinted front door, lower the rack by an inch and fit it with diamond-encrusted rack nuts."

  • Can't believe it (Score:4, Insightful)

    by lalena ( 1221394 ) on Friday May 07, 2010 @07:05PM (#32134232) Homepage
    From the article:

    that CS disk drives are more sensitive to the vibration from physically coupled adjacent disk drives.

    and

    The problem is that most civilians don’t understand the problem and are not willing to pay to solve it.

    Why should most people care about vibration caused by adjacent drives if most people only have one drive.

    The other issue from TFA is that I can't believe a different rack can cause 250% performance improvement, unless you really stacked the deck against steel racks - loose screws, hard drives not properly mounted...? I assume this means that current server racks see I/O rates that are only 40% of what is advertised by manufacturers. Are we expected to believe that no one has noticed this? What about multiple drives in a server. There is no rack separating those drives. This reads like marketing, not real research.
    http://www.greenplatformcorp.com/ [greenplatformcorp.com] is the site if you are interested and the "research" is several months old.

    • Re:Can't believe it (Score:5, Interesting)

      by georgewilliamherbert ( 211790 ) on Friday May 07, 2010 @07:12PM (#32134298)

      I've done some quick and dirty followups. Drive arrays on a concrete floor are much faster than those in a normal steel or aluminum rack with more drive arrays.

      This is real.

      You can demonstrate it with one drive array, a rack, and a solid floor. Put other equipment in the rack. Put the array in, test it. Stop testing, put it on the floor, start testing again. Put it back in the rack, test it again. The floor, test it again.

      There are some time delays involved as the drives adapt to higher and lower vibration environment - the mechanism here is the drive seems to be adopting a strategy of more error correction on reads and writes when it thinks the head's vibrating more. It will ramp that up and down as it figures out that the environment has changed.

      • Repeat your experiment using rubber washers to isolate chassis from rack. The rack can act as a tuning fork...

    • Re: (Score:3, Informative)

      by Burning1 ( 204959 )

      Why should most people care about vibration caused by adjacent drives if most people only have one drive.

      Some of the largest consumers of hard disks are enterprise companies with network attached storage, storage area networks, and RAID equipped servers.

      These companies have a very high density of hard disks, and spend a lot of money for high performance. It's not unheard of for such a company to purchase a huge array, not for capacity purposes, but for seek time and throughput.

      I'm sure they care.

    • At least some enterprise SCSI hard drives share data between each other so as to predict when each others' head movements will affect other drives. When four (for example) drives are mounted vertically under each other in a cage of any form, the sudden movement of the drive head for one drive induces some momentum in the nearby drives. Attempting to compensate for this leads to better performance whether people are aware of it happening or not, and thus higher sales.

      Whenever possible I test drives (along

    • Re:Can't believe it (Score:5, Interesting)

      by KahabutDieDrake ( 1515139 ) on Friday May 07, 2010 @08:41PM (#32135138)
      Walk over to the nearest properly mounted rack array you have, and shake it. Does it move visibly? If so, problem identified. Most racks are built to hold things up. They aren't built with much structural integrity beyond what is absolutely needed. I've seldom seen a rack with any kind of proper cross bracing, and this makes them prone to vibration transfer. You make a valid point that this is presented as a "buy this product to improve your servers" kinda thing. However, the issues with vibrations have long been ignored, and maybe that needs to change.

      My personal anecdote is: Working for a small company dealing in terabytes of data (7 years ago), they got their first disk array. Previous to that, they were using desktops to store everything around the network. So, after months of pleading, they got me the disk array I wanted, and the failure rate was atrocious. Averaged to 1 disk per 90 days. The SAN we used sat on a flimsy filing cabinet right next to a high speed printer. Not touching, but close. After a while of trying to figure out the problem, I finally sold the bosses on the idea of turning one of the closets into a server room. I installed a rack, mounted it to the wall with dampeners, and installed the SAN into it. Along with 2 1au servers, and another brand new NAS. The failure rate plummeted. The original SAN so prone to killing disks worked it's ass off for 2 more years before any of the drives failed again. As far as I know it's only had 3 disks replaced in the 5 years since then. Seems reasonable to me to assume that vibration not only plays a role in performance, but in lifetime as well.
      • I would advise against shaking random racks in your datacenter. That can make Bad Things (TM) happen :)

      • by afidel ( 530433 )
        Buy better racks then, APC makes racks that are quite rigid with the side panels on as do other manufacturers.
    • Re:Can't believe it (Score:4, Informative)

      by Angst Badger ( 8636 ) on Friday May 07, 2010 @09:32PM (#32135580)

      unless you really stacked the deck against steel racks - loose screws, hard drives not properly mounted...?

      Depending on where exactly they are, loose screws might actually help you. Tight metal-to-metal connections are much better at transferring vibrations, especially the higher frequency ones, than looser connections, where some of the vibrational energy is converted into lower-frequency vibrations. Steel is insanely good at carrying vibrations over long distances, hence the old movie trick of listening to railroad tracks for a train in the distance, or tapping on pipes in Morse code to communicate your escape plans to the inmate several cells over. (At the risk of veering off-topic, neither of these tricks work nearly as well in real life as in the movies, but they do work. Well, at least the railroad tracks do. Since the MPAA hasn't found my gargantuan mp3 collection yet, I haven't had a chance to test prison telegraphy yet.)

      One thing that has always baffled me is why racks and computer cases are made of metal to begin with. There are, of course, certain areas where you need steel or aluminum for strength or carrying waste heat, but wood or plastic would do a much better job of damping vibrations. There's a reason audio speaker cabinets are made out of crappy, soft stuff like particle board: you don't want the cabinet to resonate, and particle board does a wonderfully poor job of transmitting vibrations, which is why it isn't used in guitars, where you want strong resonance. There are also a wide variety of synthetic rubbers like neoprene and sorbothane that do a good job of absorbing vibrations. Neoprene is cheap, and sorbothane, while more expensive, is still affordable and does such a good job of deadening vibration that it feels remarkably like meat. (I happen to have a square foot of it sitting on the counter next to me, waiting to be used in some vibration-damping experiments with my scooter, but having RTFA, I think I'll try using a little bit to replace the rubber pads on the bottom of my external drive enclosures.)

      • Re: (Score:3, Informative)

        by drinkypoo ( 153816 )

        One thing that has always baffled me is why racks and computer cases are made of metal to begin with. There are, of course, certain areas where you need steel or aluminum for strength or carrying waste heat, but wood or plastic would do a much better job of damping vibrations.

        It's for RF protection in both directions. Those acrylic cases can cause problems with the PC, or with wacky noise in your stereo. I'd stick with metal, but float the drive cage, and float each drive in the cage.

        • Well, the solution would be something like a foil layer, probably on the inside where it couldn't be seen. This is pretty much exactly what Dell and HP do (or at least used to do) with their workstation cases which were plastic but had a metal layer on the inside. This of course adds cost, which is probably why it isn't common.

          • Well, the solution would be something like a foil layer, probably on the inside where it couldn't be seen

            Apple used to spray something onto the inside of Mac II cases. The problem's not insurmountable, but if you need some metal in there anyway, well, metal is a very convenient structual material with attractive modes of failure, unlike plastic. It's also more recyclable than plastic, so using more metal makes it easier to fulfill certain nations' requirements for taking back products you've produced. If you spray or stick something on plastic the cost of recycling goes up. A tiny amount of contamination in pl

  • i guess the beach boys were wrong about those vibrations.

  • Seriously? Anyone who has developed airborne systems using COTS hardware has encountered this issue with spinning media.

  • by camperdave ( 969942 ) on Friday May 07, 2010 @07:14PM (#32134310) Journal
    This is what you get when you drag a magnetic head across a surface. The sooner we get rid of mechanical storage the better. Solids are more robust, more energy efficient, quicker, denser, lighter. Cost and longevity issues are coming along. Yes, lets ditch the antiques already!
    • by Tridus ( 79566 )

      Denser? I don't see a lot of 2TB SSDs on the market.

      • by hitmark ( 640295 )

        i think you will find them being pcie cards, not 3.5" sata drives, as the drive shape limits chip numbers.

        • i think you will find them being pcie cards, not 3.5" sata drives, as the drive shape limits chip numbers.

          All of the SSD drives that I've seen are 2.5" drives. I've often wondered why no one uses some simple heat sinks and packs the chips as dense as can be into a 3.5" form factor.

          • OCZ colossus among others are 3.5"

    • The sooner we get rid of mechanical storage the better. Solids are more robust, more energy efficient, quicker, denser, lighter. Cost and longevity issues are coming along. Yes, lets ditch the antiques already!

      We have quite a way to go before that's practical for high-capacity write-intensive applications. And, as always, we won't know how much longevity to expect until we get there: the manufacturers will make wildly optimistic guesses early on, switching to bald-faced lies later, the same as they've done with every other storage medium. If you want to ditch your antiques, I'll be happy to put them to use biding my time while the bleeding edge bleeds. ;)

    • by Twinbee ( 767046 )

      I think people have kept with hard drives for so long because they were so noisy and slow. Now they're getting quieter or even silent, SSDs have become more attractive.

  • I previously had a hard drive in my Mac Pro that was vibrating like crazy; it was making the entire machine vibrate, and all for only 80GB. It wasn't the boot drive, but rather a drive where I just had some random crap.

    I decided getting rid of the vibration (and resulting buzzing sound) was more important than having a paltry 80GB more, so I copied the data off and yanked the drive. My machine seemed to boot and run quite a bit faster after wards. I was pleasantly surprised. My previous theory was that the

  • by NicknamesAreStupid ( 1040118 ) on Friday May 07, 2010 @07:56PM (#32134782)
    Plus a lube job and, definitely, Window cleaning.
  • Just in time, too (Score:4, Insightful)

    by Weaselmancer ( 533834 ) on Friday May 07, 2010 @08:03PM (#32134850)

    This is news we needed 20 years ago. SSD is going to replace mechanical HD over the next couple of years making the whole vibration issue irrelevant.

    Ah well better late than never I guess.

    • This is news we needed 20 years ago. SSD is going to replace mechanical HD over the next couple of years

      You're probably off by an order of magnitude, there. SSDs will get adopted pretty quickly for somewhat large, and very frequently accessed/updated databases, and pretty much nothing else...

      It makes sense the consumer market is adopting SSDs... SSDs vastly outperform a single HDD, but in a server environment, both throughput and seek times can be improved linearly, by just continuing to add more drives t

  • I've always suspected this, but never really cared to address the issue in any meaningful way because wtf can you do about it? I've always hated metal racks, and all the "affordable" stabilization kits do not do much to transfer the vibration from fans and other moving parts from the servers. In fact, the metal racks make for a nice little conduit to help spread out vibration from server to server. The more full the rack, the more the impact.

    Over the course of years, I've had plenty of pci cards shake lose

  • by melted ( 227442 ) on Friday May 07, 2010 @08:26PM (#32135026) Homepage

    The disks all spin at roughly the same frequency (250Hz for a 15K RPM drive), so you could get some interesting resonance patterns in that frequency band as well as in its harmonics and frequencies that you get when you subtract rotational vibration spectrum of one drive from another. You can even hear these effects if you run two 7200 RPM drives in your desktop in a quiet room (assuming you don't have a dozen fans in the case that some people like to have for some reason).

    The solutions is simple - dampen the drives to eliminate high frequency vibration transfer. Better yet, don't use screws to attach your drives at all. Use velcro.

    • by evanbd ( 210358 )
      250Hz isn't high frequency; it's fairly low, as audio frequencies go. It's normally considered in the bass or low mid region. Wavelength will be > 1m in air, several times that in the materials making up the enclosure. Damping it out effectively and efficiently is not entirely trivial, though the obvious techniques like rubber mounts and the velcro you suggest will help a lot.
  • SSD (Score:3, Interesting)

    by Ruvim ( 889012 ) on Friday May 07, 2010 @08:29PM (#32135044)
    It is going to matter only until price for SolidStateDisks becomes in line with what it should be [businesswire.com]
    • SSDs are clearly the storage technology of the future, no doubt about that. My 80GB X25-M is the best $300 I spent on on my computer.

      Right now, though, while they are still an order of magnitude more expensive per terabyte than traditional enterprise drives, addressing vibration sounds like an ingenious and cost-effective way for companies to squeeze out more performance out of available technology.
  • About the last thing you would want is a light carbon fibre reinforced plastic rack if you want to damp vibrations. A well constrained steel rack with rubber washers would do a far better job because it is far stiffer for a start and you don't want to get much motion.
    The answer is to learn from others. With high speed machine tools the trick for damping and stiffness is to use a very heavy base made from materials which damp the vibrations - such as grey cast iron where the damping effect from the graphit
    • Sorbothane is a urethane developed specifically to attenuate mechanical vibrations.
      It works.
      No other kinds of rubber will attenuate vibration nearly as well, and some will even make vibration last longer than it would without the rubber present.

      One use for it is tripod foot pads for amateur telescopes. You can go search amateur astronomy forums for user testimony about it. The only people who think that "any kind of rubber will work like Sorbothane" is the people who have not yet tried using Sorbothan
  • ..before I suspected something was not right and wrote a program continuously monitoring read speed. Then I started it, waited for the number to settle and put a disk into my CDROM. The reading speed decreased several times. After that I took the cdrom out, put it atop the case on a pile of foam rubber. That computer worked for 10 years more without any problems. The CDROM brand was Pioneer, if you care. HDDs were Maxtor and Western Digital. The story happen in 1997.
  • This shouldn't be dealt with at the rack level. It should be dealt with at the disk drive mounting bracket level, where it's far easier. There are brackets for that, [quietpcusa.com] from several suppliers. [cyberresearch.com] There are also rubber grommet kits for fans [nexustek.nl], to damp vibration from that source.

    If you buy servers in bulk, it's something to take up with your supplier. It's the sort of thing that costs only a few dollars per unit at the factory.

  • by HockeyPuck ( 141947 ) on Saturday May 08, 2010 @01:58AM (#32136900)

    Note: I'm not a server guy, i'm not a LAMP guy, I'm not a OS guy nor a DBA. I deal with two objects in the datacenter, storage devices and switches that move packets (either FC or FICON) to storage devices.

    First when I heard the term 'enterprise disks' I figured they were talking not about the drives themselves, but rather of the highend EMC, IBM, HDS variety (HP/SUN don't count b/c they rebrand HDS). There's no dampening in those arrays, they're basically racks/cabinets on wheels with casters on them to lock them down. If I was reading from a raw disk, then I could definitely see how vibration would have an impact, but with Enterprise disk arrays, there's so much cache (in the array, not on the drive) and read-ahead algorithms in place that I could see how users wouldn't notice the difference. I'm not so sure that EMC/HDS/IBM would be willing to build their disk arrays out of carbon fiber. Especially with the price conscious consumers like myself that love nothing more than my yearly meeting with my storage vendor to discuss $/GB.

    I know of some companies that put their highend (Superdome/p690) servers on earthquake pads, which in the event of an earthquake the server can stay put while the floor shifts underneath.

    I've actually experienced this problem first hand. I used to work at one of the above mentioned storage companies and we manufactured a disk shelf that had 8 drives in the front and 8 in the back. There was a metal divider in both the front and back that separated the box into quadrants. We noticed one year that there was a significant drive loss in the field and upon further investigation, we noticed that one slot in particular had an abnormally high failure rate. So we flew to one such site that had these suspect drive shelves, and measured the vibration of each disk in their disk shelves (they had about 100) using a tool that look pretty much like an accelerometer at the end of a pencil. Turns out that the drive location that had the highest number of failures, was not abnormally vibrating, but that a drive 4 spots away was. It seems that if a the drive next to the divider had a "high vibrational drive" it would set up a standing wave which would eventually cause another drive, which was perfectly fine from a manufacturing standpoint to fail.

  • Not surprising... (Score:3, Interesting)

    by John Pfeiffer ( 454131 ) on Saturday May 08, 2010 @11:45AM (#32139532) Homepage

    Luckily, all my drives are mounted through silicone vibration isolators. Gotta love Antec case design! :D http://media.giantpachinkomachineofdoom.com/2007-02/blog/images/u1_array_supersize.jpg [giantpachi...ofdoom.com]

  • I work for a small ISP and we encountered this recently.

    We bought a few SuperMicro small form factor chassis (http://www.supermicro.com/products/chassis/1U/512/SC512-260.cfm [supermicro.com]), and found that with the drives positioned directly next to a high speed prop, the performance of the disk went from a static 125Mb/sec to as low as a few kb.

    The drives we initially bought were WD 1TB Green Drives, and we thought it was initially a "Green" feature. But with thorough testing (and after replacing the drives with Barracud

The sooner all the animals are extinct, the sooner we'll find their money. - Ed Bluestone

Working...