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

 



Forgot your password?
typodupeerror
×
Technology

Terrabit Per-Square-Inch Hard Drive 146

BitGuy writes: "Physics News Update reports that current GMR (giant magnetoresistance) harddisk technology will not achieve terrabit-per-square-inch densities. Experiments with EMR (extra-ordinary magnetoresistive), which exceeds 100Gb/in^2 have been successful in the lab. There is even a diagram of the read head if you're interested."
This discussion has been archived. No new comments can be posted.

Terrabit Per-Square-Inch Hard Drive

Comments Filter:
  • Terrabit??? (Score:4, Funny)

    by bioart ( 256479 ) on Monday May 27, 2002 @08:45AM (#3590526)

    Hmm... I wonder how they got dirt and the like to such high density... did they mean Terabit?

    • Poor spelling angers Cromulus.
    • It's a Japanese thing. Mini-bonsai terraforming of the harddrive.
    • for future 3D volume holographic optical
      storage nanotechnology.

      Unfortunately, the articles doesnt mention that
      this 1 terabit/sq.in is 2D AREA technology, i.e.,
      your read or write 1 bit at a time !!! 600 mbits/sec ------ SLOOOOOOW !!!!!.

      Future Rewritable Volume Holographics can read AND write >>>>> 100 gigabits/sec

      1 bit versus billions of bits at one time !!

      This GMR technology is stupid and obsolete, is
      slow and quite frankly its an insult to have
      such primitive technology moving forward. Yuk !

      Go to this site for futue store;

      http://www.colossalstorage.net

  • Now all we need is for Palm to get their act together and make those expansion slots support hard drives :-)

    -J
  • Huh? (Score:3, Informative)

    by p3d0 ( 42270 ) on Monday May 27, 2002 @08:49AM (#3590539)
    What a bizarre way to report the story. The news is not that terrabit densities have been achieved; rather, that GMR will not be the technique that will get us there. Hardly news at all.
    • by p3d0 ( 42270 )
      Oops. That's "terabit".
    • by rde ( 17364 )
      The news is not that terrabit densities have been achieved
      Yeah; that's not news. Terrabit=piece of Earth. Not that dense at all, cosmically speaking.
      No, I'm not blaming you. I'm not even blaming He Who Posted the Story. I figure it's a new US law, or something. Like the one that requires every american to spell 'lose' with two Os.

      Suppose I'd better say something partially constructive. Here's [post-gazette.com] the story from about six months ago.
    • The story is a bit back to front, but not as bad as you suggest
      it says:

      terrabit densities

      Not using EMR
      Using EMR

      Here's a picture of the read head
    • Why not have "news" stories that say "10 GHz processors not here yet" or "2048 x 1460 LCD displays not yet ready?"

      And I got all excited when I saw that headline! "Finally," I thought, "now there'll be space to store all my pr0n!"

  • hm.... (Score:2, Funny)

    by elmegil ( 12001 )
    Terrabit....is that what I'm going to feel when I get to play Doom III?

    OH, you maybe meant "Terabit". As in "a trillion bits". This has been your obligatory spelling flame.

  • by Anomolous Cow Herd ( 457746 ) on Monday May 27, 2002 @08:51AM (#3590550) Journal
    ... current GMR (giant magnetoresistance) harddisk technology will not achieve terrabit-per-square-inch densities. Experiments with EMR (extra-ordinary magnetoresistive)...

    I'm not buying another one until it comes with Super Duper Magnetoresistance.

    • Imagine if it was called something different that had the same first letter e.g. Gargantuan
      GMR - Giant magnetoresistive
      TMR - Tunnelling magnetoresistive
      GMR - Garguantuan magnetoresistive

      Now what were those flames about AMD not using GHz any more and therefore they're ripping people off and misleading them, well, eh?

    • At least it wasn't Extra-sensory Resistance (ESR) ... Though the thought of a psychic HD is a little disturbing... OTOH, it could probably do cacheing a lot better & maybe even improve download times by knowing what you wanted in advance... :]
    • Yeh! Who makes up these names? I mean seriously?

      TERRAbit? What next? Venutiabit? Martiabit? Jupitabit?
  • it's TERA- (Score:1, Informative)

    by Anonymous Coward
    Tera is the SI prefix for trillion (10^12), not terra, which is Latin for earth.
    • Why is a high-tech industry mixing SI prefixes with English inches anyway? The next sexy target for hard disk development ought to be 1 Pb/m2 (petabit per square metre).
    • after Tera comes Peta, Exa, Zetta, and for now Yotta, a good short explanation of all these is at http://www.jamesshuggins.com/h/tek1/prefixes.htm
  • It hurts to read. The article is written backwards, in most convoluted sentences possible. Could we please be somehow warned NOT to read it?

    There is nothing to read, and there is nothing to discuss anyway (besides the standard grammatical challenges). The horrible "reporting" (if I may call it that) does not deserve reading, linking or even the IP traffic.

    • Why make up a warning system? No one who posts here actually reads the article anyway.
    • The article is written backwards, in most convoluted sentences possible

      I guess yoda feels that balance has returned to the force and as such is writing online news articles now
    • Backwards you say? Good it is not?

      Technology using tunneling magnetoresitance (TMR) or giant magnetoresistance (GMR) will probably not achieve the present goal for magnetic recording research, hard drive densities of a terabit per square inch.

      Right you are. Work better backwards it does!

      • The particular quote mentioned here is not backwards. Instead, the purpose of the comma is to set off the appositive marked in bold

        Technology using tunneling magnetoresitance (TMR) or giant magnetoresistance (GMR) will probably not achieve the present goal for magnetic recording research, hard drive densities of a terabit per square inch.

        I agree that it doesn't flow all that well, and that "the present goal..." should switch with "hard drive...", but it is not backwards in its present form. Yet.

        • I guess you missed that this 'quote' was indeed turned backwards compared to the original article. The original was

          Terabit-per-square-inch hard drive areal densities, the goal for present magnetic recording research, will probably not be achieved with giant magnetoresistance (GMR) or with tunneling magnetoresistance (TMR) technology.

          And, yes, it works a lot better this way (IMHO).

  • by isorox ( 205688 )
    After this amazing news about GMR not supporting Terabit/square inch densities, slashdot reports that pigs *cannot* fly, and bill gates *isnt* poor.
    • And of course,we are all waiting for the story which says that people do need that much porn...
    • As near as I can tell from the article, GMR technology was expected to scale to Tb/in^2 densities, but this guy A) thinks that magnetic noise will prevent that density from being achieved, and B) has a (as far as I know) new idea for a technology that is not succeptible to the same type of noise.

      For anyone who cares about things other than where they keep their pr0n and if they will have a fast enough GPU to run Quake n, this is potentially very interesting. I wish more news was like this, and less "Foo Bar Inc. has just released a new 23 terawhoosit widget, brining unprecedented levels of frobing to a consumer device" which should properly be called a press release, not news.
  • Story summarized (Score:4, Insightful)

    by peter_gzowski ( 465076 ) on Monday May 27, 2002 @08:58AM (#3590569) Homepage
    Giant magnetoresistance (GMR) and tunneling magnetoresistance (TMR) hard disk technologies cannot achieve Terrabit per square inch performance. Neither can the "extraordinary" magnetoresistive (EMR) hard disk that is proposed. It is hoped that in the near future "extra-extraordinary" magnetoresistance (EEMR) hard disk technology is developed, and then perhaps "goddamn-extraordinary" magnetoresistance (GDEMR).
    • This reminds me of a discussion we had about what the next-generation sparc chips would be called. I mean they've had microsparc, hypersparc, supersparc, turbosparc and now ultrasparc, what's next?

      I'm hoping for übersparc CPUs in my next server, but who knows...

      /August

      • "I mean they've had microsparc, hypersparc, supersparc, turbosparc and now ultrasparc, what's next?"

        Intel Itanium (TM) ...
      • electricsparc
        brightsparc
        mommasparc
        jesussparc
        googleplexsparc
        sunsparc

        Maybe they can start again with a different acronym, since the parent company is called Sun, these might be suitable, from here [glyphweb.com]
        K-type
        G-type
        F-type
        A-type
        B-type
        These being the types of star, starting with the coolest

  • by Anonymous Coward
    "Physics News Update reports that current GMR (giant magnetoresistance) harddisk technology will not achieve terrabit-per-square-inch"

    Don't worry, 999 Gb will do me fine.
  • by Speare ( 84249 ) on Monday May 27, 2002 @09:00AM (#3590577) Homepage Journal

    GMR (giant magnetoresistance) harddisk technology will not achieve terrabit-per-square-inch densities. Experiments with EMR (extra-ordinary magnetoresistive), which exceeds 100Gb/in^2 have been successful in the lab.

    After that, comes IMR (improbable magnetoresistive) where the Library of Congress fits in a square inch.

    Finally, new advancements in subatomic physics leads to LMR (ludicrous magnetoresistive), giving more bits of storage than there are atoms on the platter. The "flavor" and "color" of each quark are directly manipulated and sampled by the drive head.

  • by Alien54 ( 180860 ) on Monday May 27, 2002 @09:05AM (#3590595) Journal
    With technology advancing, I wonder about things like Solid State drives [superssd.com]. I mean, with mechanical drives getting so small, you would think the next logical step would be to eliminate the mechanicals entirely.

    I recall some things from some years ago where there were even transparent colored cubes that looked like things straight out of Star Trek, but they had problems with the registration. It was next to impossible to reseat the cube exactly correctly so that you could retain access to your data. but obviously, other solutions have worked well.

    I would love for the cost of these things to come down to something reasonable for the consumer. Recalling the old Tandy laptops that some folks still use, one of advantadge of them is their virtual indestructability, all because of the solid state memory drives inside. (admitting they are small, but they work very very well indeed)

    • Yes, let's leave the sluggish, cumbersome clockworks back in the 20th century, and get on with the solid state "drives". Oh, and while we're at it, let's grow vertically as well as horizontally. No need to stick to 2D is there?
    • Solid state might be used as intermediate fast storage, much like TRAM (transactional RAM). But it will NEVER replace mechanical. EMP from a nuke will blow away solid state battery-backed RAM, static electricity as well in hadling. There's nothing simpler than an Ultrium tape. It's just iron filings, which fails in a predictable way so data recovery services will almost always be able to do something. A lightning strike will fry your linux box and all HD controllers (even if you've got RAID-5). So if you're feeling cosy and secure with RAID-5, think again.

      Lightning is like an asteroid hit, it's rare but when it hits a power line or phone line the effect is felt for many miles around. Blowing your modem and working it's way out from there. Your line surge protector *might* protect your PSU from this happening on the mains, then again.....

    • This comes up every time a storage article is posted.

      It's shot down every time, but it keeps getting re-posted.

      Calculate the cost of the RAM in your computer, per gigabyte. Now, calculate the cost of storage in your hard drive, per gigabyte.

      Notice that the difference is several orders of magnitude.

      In order for solid state drives to be cheaper than magnetic drives, the cost of pick-your-RAM-flavour has to get a HUNDRED TIMES cheaper, while the cost of hard drives has to NOT get cheaper.

      This might happen in the far future if prices drift and keep drifting, but not any time soon.

      You can also make a good argument for it being intrinsically cheaper to manufacture hard drive platters than RAM arrays, but this has been beaten to death already.
  • Limits (Score:3, Interesting)

    by danamania ( 540950 ) on Monday May 27, 2002 @09:08AM (#3590604)
    Curiously - apart from mass data storage repositories for corporations, does anyone think we'll reach a limit to the amount of data we'll need as individuals?. While we're creeping towards (and will pass) terabyte sized drives and the ability to store every piece of documentation about ourselves, it seems to me (and this may be shortsighted) that all we have left to use is high quality media files relating to our own lives.

    How much would you record of yourself, your actions - in sound, video, feelings if you could... and would you edit it down, or keep everything you could.

    (pondering, more than posting)

    a grrl & her server [danamania.com]
    • Re:Limits (Score:3, Interesting)

      by tftp ( 111690 )
      does anyone think we'll reach a limit to the amount of data we'll need as individuals?

      Universe is infinite, probably. There would be no limit. To start with, I wouldn't mind having a local copy of a map of Earth, with resolution to 1cm, in 3D, so that I can "travel" in virtual reality. After that, I'd like to have the same for the sky (astronomy) and nearby planets... Other people could instead prefer a library of all the books, paintings, sculptures, music, photos and movies ever produced, and special publications - as super-detailed images.

      • You won't need it locally.

        Extraordinary-Internet (EI) will be there to deliver all-you-can-eat instantaneously

      • Of course, all-the-literature-ever is rather small. If 100 million books where ever produced (probably a large overstatement), a book is about 150 pages, 3000 characters per page... That makes 45 TB, packs with good compression to 1/10th, easily stored on any super computer fibre channel array (and next years HD's).

        I want the same global terrain you had, in voxel graphics (so I can explore the interior as well.)

        That's about 1000 YB (yottabytes, 10^24 (or 2^80, if we still haven't resolved the binary/decimal prefix issue)). Oh, and why are you satisfied with cm resolution? That would look really chunky, I want mm or better (multiply by 1000). In short: no, demand for memory/storage will not fall off any time soon. OTOH, if memory capacity continues to double every 18 months, we will reach yottabyte capacity in 22.5 years. Scary.

      • Well, I think there are limits - they are defined by your input channels.

        You cannot view the whole earth at one cm resolution at once, so there is no need to store it locally.

        As soon as the data transfer rate and latency for accesses to "main" memory is no longer a bottleneck, the need for local cache suddenly goes away.

        For "audio" input, this limit is almost reached for the home user.

        Unless you have golden ears (which might be inconvenient if you plan to spend the winter in Minnesota, anyways), 256Kbit will do. If that Bandwith comes at no extra cost, once the copyright issues are sorted out, one way or another, there is no reason to distribute a few hundred thousend songs several thousand copies each.

        Along comes internet radio and the virtual jukebox!

        For video input, if the display was shaped to the limitations of your eyes, something like a 1000x1000 pixels at 500 dpi resolution in the center of your vision (2 inches square), plus another 1000x1000 pixels in the periphery, will do (Think of a monitor that watches where your eyes look, and renders that part of the screen with extra resolution)

        At 50Hz with some temporal interpolation, this is something like a 3 gigabits raw or a 100Mbits compressed, even now within the reach of your LAN, if compression and Display hardware were advanced enough.

        I would guess that more than 95% of what is on harddisks these days is duplicated elsewhere, and therefore see tough times coming up for HD vendors once the networking guys catch up.
    • Curiously - apart from mass data storage repositories for corporations, does anyone think we'll reach a limit to the amount of data we'll need as individuals?.

      I wouldn't have thought so - as we synthesize intelligence we we will need to represent more complex relationships and entities.

      Of course, if the universe is finite, then there is a limit on data that can be represented. At least I think so...
      • does anyone think we'll reach a limit to the amount of data we'll need as individuals?.

        Not bloody likely in my lifetime. When I can record a month or two of H3DTV on a hard disk THEN, MAYBE, I'll be satisfied.

    • Re:Limits (Score:3, Interesting)

      by jd142 ( 129673 )

      Forget recording yourself, the desire to put my av collection right on the hard drive is what's driving me to coninually purchase more drives. I just finished filling up a 50 gig drive with a part of my cd collection. The next step would be to put my dvds on a hard drive as well. I could easily use up 1,000 gigs of data that way, and I don't have nearly as extensive a cd/dvd collection as many people.

      I'm just waiting for someone to pick a feckin' standard for dvd recordable and get the price to closer to 200 dollars, then I'll be able to do away with approximately 6/7ths of my back up cds.

    • I'm sure people will work out plenty of new ways to waste... uh, I mean use... storage. After all, there was a time the corporations got by on the "huge" 1.5 mb data store. Nowdays we can barely fit a gif in that.

      Besides, micro$haft will make sure we use plenty of hard disk space.

    • Ofcourse we'll be able to suck up all that space... we need to have it for all the copies of dvds, cds, and bootlegs that we pirated from the MPAA and RIAA.

      :-)
    • If given enough data storage space, a person will find the data to fill it, whether it be DVDs, CDs, texts, web pages, code, or what have you. Remember when Bill G said "640k should be enough for everyone", back then he was (almost) right. Now it's not unusual to see machines with 1000 times that amount of RAM and 100,000 times that amount in hard drive space. As time progresses, I don't expect this to change.

      I actually think this bit is human nature, as if you take this example and apply it into the physical realm, it almost holds true there as well. If you give a person as much space as they want, they'll fill it as they accumulate stuff. Some people make a conscious effort to cull down their stuff after a while, but many don't unless there's a huge prevailing need to.

      More pondering, but hey, you did ask. :)

  • by jb_nizet ( 98713 ) on Monday May 27, 2002 @09:09AM (#3590607)
    Common, when will you FINALLY adopt standards?
    When you finally do, I'll go drink a pint of beer and a eat a pound of cheese at the pub, two yards from here ;-)

    JB
  • by Anonymous Coward
    Dont use the word "inch" around the slashdot crowd.
  • Yah, sure, terabyte drives are coming... We all know that Maxtor had 120 TB drives back in 1987, but they incrementally release larger disks so as to keep the revenue stream coming. It's a conspiracy!
  • by egarland ( 120202 ) on Monday May 27, 2002 @09:28AM (#3590664)
    Disk manufacturers have been warning [slashdot.org] that hard drives will stop getting bigger soon. GMR will hit theoretical limits and there will be nothing to replace it that's better because of magnetic noise limitations.

    What this article is saying is that there is a new technology to move to when GMR hits it's limits. 3.5" drives won't stop at 180 GB per platter in 2 years. EMR will pick up where GMR left off and we should be able to see 1 TB per platter before they need to invent the next new technology.

  • "Stop this Progress! Stop it, I say!" --the character Theotocopulos in the screenplay of H. G. Wells' Things to Come. And later:

    Passworthy: "Oh, God, is there ever to be an age of happiness? Is there never to be any rest?"

    Cabal: "Rest enough, for the individual. Too much and too soon and we call it death. But for Man no rest and no ending. He must go on, conquest beyond conquest. First this little planet with its winds and waves. And then all laws of mind and matter that restrain him. Then the planets about him. And at last, out across immensity... to the stars.

    And when he has conquered all the deeps of Space and all the mysteries of Time [quietly, broodingly] still, he will be beginning.

  • by Aceticon ( 140883 ) on Monday May 27, 2002 @09:36AM (#3590689)
    I sugest a poll for the name of the mangnetoresistance effect that will come after giant magnetoresistance and extra-ordinary magnetoresistance:
    1. Even-more-extra-ordinary magnetoresistance
    2. Unbelievable magnetoresistance
    3. Bloody-incredible magnetoresistance
    4. All-powerfull magnetoresistance
    5. Cowboy Neal
  • 'areal' and 'InSb' are underlined (how twee :)
  • "There is even a diagram of the read head if you're interested."

    Oh, good, that's just what I wanted to make. Let's see... I'll need a hammer, chisel, Bowie knife, file...

  • Let's try to clear this up once and for all. Nintendo is based out of Japan which uses the metric system, ergo Nintendo is metric. Therefore the proper units are not 'tera/terra' etc., but are as follows:

    magnetoresistance
    Super magnetoresistance
    magnetoresistance 64
    GMR Cube

    We now return you to your regularly scheduled pedantic flamewar...

  • Way back when... (Score:4, Interesting)

    by jmichaelg ( 148257 ) on Monday May 27, 2002 @10:36AM (#3590856) Journal
    ...winchester drives came out, folks were talking about how small the read gaps were and the damage a human hair or a smoke particle could cause if it got between the read head and the platter. Since hard drives were the size of a washing machine, it was pretty amazing to think that a smoke particle could ruin it. Disk drives "fly" the heads as close as they can to the platters to minimize the area being affected by the read/write signaling.

    So at what point does the surface of "perfectly clean" material get so inherently bumpy that it's impossible to go any further without crashing into the random atom that sticks above its neighbors? Given the bumpiness induced by thermal agitation, are hard drives of the future going to have to be cooled just to get the heads in close enough?
    • we all know that cold hard drives are happy hard drives (until condensation happens), but I think requiring some sort of strong cooling will be a necessity. The limitations of hard drives will no longer be of just the read/write head.

      Rotational speeds and data density will get so high, that the vibration and heat caused by mechanical bearings will not only cause uncontrolled thermal expansion, but also create platter vibration larger (peak-to-peak) than the density of the data itself.
      Mechanical bearings will need to be replaced by either compressed-liquid bearings, or isolated magnetic bearings to eliminate the vibration.
      It's also unfortunate that the MR heads require a thin cushion of air to ride slightly above the platter's surface. That rules out running the platter inside a strong vacuum, to eliminate air friction (ever had a cdrom spin in a high-speed CD drive for a few minutes, and felt it? that heat is from friction with the air, not from the laser)

      I'm surprised these things aren't brought up when advances in disk density is discussed.
    • It's not that the material gets so bumpy it's impossible to go further - you just have to make read heads that can read the data from higher up - the two problems aren't separate. I can't find the link right now, but IBM's press material concerning glass platters includes a micrograph of the surface of a glass platter versus an aluminum platter showing the glass platter to be 60% flatter so the read heads can fly closer to the media without fear of head crashes.
    • The thermal vibration of atoms are much smaller than the atom itself. That wouldn't be a problem.
  • I hope they make that 100Tb drive i'll need in five years a fast SOB, otherwise i'll have to wait till I'll be gone for a week or two to reformat it
  • I'm waiting for LUDICROUS MR!

    Pass me those plaid HDD lights for my case....
  • This is an old article, but apparently they never got any further on the subject:
    http://www.sciam.com/askexpert/computers/computers 2.html [sciam.com]
    I suppose that if one can get mirrors to move fast (9ms to be faster than normal harddrives) and accurate, wich is probably just what they already can with normal hardrive heads moving much faster (the average access time from harddisk drives is determined by the speed the disk spins, not the speed of the heads), the only problem left is getting the data-density up..
  • by BitHive ( 578094 )
    Not just ordinary anymore! Now packed chock-full of extra-ordinary for more of that generic, everyday experience!
  • It's kind of scary looking at the thing that in five or ten years we'll be depending on to store our data, and seeing the huge irregularities in its structure. I suppose the stuff inside computers right now is like that too, but it makes you wonder how they get the precision they need, and why those flaws don't make the things break more often than they do.
  • ... will probably be

    IMR = Incredible MagnetoResistance

    and then

    RMR = Ridiculous MagnetoResistance
  • Interesting and I can't wait to get my hands on an Eval Ultra 320 SCSI drive from Seagate, which is very here and now.

    But with all this talk of growing apoferritin and magnetic proteins to get close to the illusive 1TB/inch. I just wonder at what point my hard drives going to get up and walk out my PC.

    http://www.nanomagnetics.co.uk Chasing the ACE in the UK for big drives, currently about to demo 8GB/inch a long way off that 1TB mark.

    One problem faced by any company in this game is the pace of development and often todays leading technological break through gets leap frogged by abother. Leaving some people seriously out of pocket.
  • How about stupendous magnetoresistance?
  • ...but if you guys are really that dead set on not adopting the metric system, couldn't you at least present your measurements with metric units in brackets, for...I dunno...the other few billion people who might be interested?

    Just a thought...

Trap full -- please empty.

Working...