Slashdot Mirror

← Back to Stories (view on slashdot.org)

Holographic Storage Overview at CNET

Posted by timothy on from the smoosh-smoosh-smoosh dept.

encebollado writes: "CNET has an article about how holography is being used to create next generation storage devices. The researchers promise they'll beat out DVD by an order of magnitude." Actually, it's an overview with four separate articles -- no bets on when the technology covered will really be available though.

5 of 119 comments (clear)

  1. "an overview with four separate articles" by jeffehobbs · · Score: 4, Funny


    Actually, it's an overview with four separate articles...

    It's interesting to note that articles about holography can be broken/cut into multiple pieces, and each piece will retain a full and exact copy of the original article.

    ~jeff

  2. my eggs are in the IBM basket by fons · · Score: 5, Interesting


    interresting timeline

    - IBM has lot's of hard-disk related technologies patented
    - IBM has a relativly flourishing HD business
    - IBM sells said HD activities (except R&D)
    - IBM breaks storage records in lab with new technology

    => I'm betting IBM will come out with a new kick-ass storage technologie shitin the next 5 years

    1. Re:my eggs are in the IBM basket by fons · · Score: 4, Funny


      yeah, and I could use a kick-ass spell checker

      (In my defense, it's late and English is not my mother tongue)

  3. You do have a hard limit. by Christopher+Thomas · · Score: 5, Insightful

    Also, there are no fixed limits on how much information you can store on hologram - you can always store something more, which will lower the quality of the rest of stored information, but you won't hit any fixed maximum number of bits, like with standard types of memory. Saying that it "[beats] out DVD by an order of magnitude" is totally ignoring the most fundamental features of holographic memory.

    The problem is, if I don't care about the quality of the data retrieved, I could use /dev/null to store data and /dev/random to retrieve it and claim as much space as I want.

    If you need to get your data back intact - i.e. with enough fidelity for you to rebuild the original data without loss - there turns out to be a hard limit to how much you can store with a holographic storage medium. The exact limit varies based on the geometry of the setup and of the holographic medium, but can be calculated. You can also measure it directly for any real system, which is presumably what the company involved did when citing storage densities for its prototype.

    So, while the accessing method is very different, the storage limit for holograms scales in the same way as storage limits for other types of device (in this case, with the volume of the holographic film IIRC).

  4. Cube storage on Silica 48E by Anonymous Coward · · Score: 5, Interesting

    I work for Tilc heavy industries and whatever I say here is not endosed by our corporation.

    We have been researching into holographic storage since way back in 94. Dr. Erwin Gupta, our senior researcher is known for his work on such storage (AFAIK he was one of the main proponats of DivX (hehehe)).

    We are using Silica 48E in crystalline format to form a 3D latics that can store and information. Now for the nice part, our sturcute XI can (3 cm - 2 cm - 2 cm ) can store informations in the magniature of 78TB. Information retrival is only hapmpered by speed of axuliary pheripherials. But our test machines are now running with optical conntections to the cube, and thus storage and retrival is trivial.

    The only problem we have is, how to format such huge quantities of information. By format I mean, how to place it in a viable layout. I feel this is one area where we have constantly lacked. In the past we've been using standard methods of storage (Unix filesystem type layout and b-tree, cMax cube), but this lead to storage being shrunk down by a magnitude of 15. For information that is stored in bulk (eg: Large archives that are interlaced raw, this is not an issue).

    Our partner, IBM has also been very interested in the Silica 48E strucute, they are also going into research with us and I felt this was one of the reason for the closure of their HD shops (since they've felt the limit of HD's being reached and thus they are moving to better media).

    Silica 48E can be mass produced cheaply, the storage opens to almost limitless quantities (Oopes.. Sorry If I'm doing a Gates 64k again). Currently we are stress testing the crystals and we have put the entire library of congress (Storage A-F (Pre 94)) in 2 crystals with an induced Cmax filesystem. Retrival is an issue here, cause such a large archive needs better tools (ours is only hardware).

    I dont see crystals going into consumer use in the near future, give it 5-8 years and you'd probably seem them at the high end. Given that, it's cheaper to produce on Silica 48E cube than make 3 DVD's :> The military,government and archives are our first customers, we also have strong interest from world libraries and other such entities. IBM might go into production of Silica 48E at their Phillipines plant staring Q3 04.