Slashdot Mirror
Can Ten Billion Gigs Fit In A Test Tube?
Nipple writes: "Using Nanotechnology scientists ar Rice University have been able to store 10 billion gigabytes of data on physical storage small enough to fit into a small vial. The whole story appeared in The Philadelphia Inquirer and
can be found here." No indication is given of which calculations for data density the tiny vial pictured would be able to hold that much, but the idea of all the books on my bookshelf (and yours, and yours, and yours ... in fact, all the books I ever want to read) stored inside the stylus of my 9-day-battery life, white-LED-backlit wireless anything box is pretty tantalizing.
So all you need is two of these vials and you've exceeded 64 bits of address space.
Check some articles about this in Wired and Scientific American. They are about Tour and Reed. It talks about their plan on developing molecular computers. Sounds like they are very close to coming up with transistors but have quite a ways to go to come up with wiring!
"sweet dreams are made of this..."
find / -name "lost.data" -ls
yeah, right.
I know that there are companies out there that are dealing with this sort of problem already (Zantaz archives email for a living... a lot of email) but I have no idea how much processing power it would take to find something is a reasonable time frame, or how it might be indexed.
Are these legitimate concerns or is this just another version of the "indexing the internet" problem?
--
Behold the Power of Cheese!
There's more interesting information about the binary powers here
--
Please consider making an automatic monthly recurring donation to the EFF
The web page for the Tour Group at Rice University can be found here. It has links to other articles on molecular computing. Beware, the page has a 500K picture on it, not very modem friendly.
Mea navis aericumbens anguillis abundat
Actually, the article doesn't say that they've been able to store anything like that. The article says:
Ten billion gigabytes of data can be stored in this vial, according to Molecular Electronics Corp.'s cofounder, Jim Tour.
And they say that they've demonstrated the use of a molecular form of DRAM. Now these are pretty good advances already, and I believe nanotechnology will definitely change the way we look at computing, but please, Timothy: These guys don't have a secret RAM plant set up ready to conquer the world overnight.
Not just yet, anyway.
--
Pokéthulhu
Gotta catch you all!
If they become as cheap as they claim in the article, you could have ten copies of your library. It wouldn't matter then if one was stolen.
I am more interested in how small and cheap they can make these things: with IPv6 providing enough address space for every piece of clothing you will ever own, having a gig of storage in your tennis shoes might be rather interesting... You wouldn't really have to worry taking your PDA with you anymore, if your clothes would automatically talk to each other and keep all your data handy regardless of what you wear. And those into nudism could still wear jewellery... =)
Since they are nanotechnology, would it be possible to inject them into your bloodstream? Then you would never lose your information - and you could exchange information with other people by exchanging bodily fluids. "File sharing" just wouldn't be the same anymore... ("Hi, can you get me the Smith file?" "Yeah, just lift up your skirt, will ya?")
On a serious side, naturally having strong encryption becomes even more important when you have that much storage density.
But, will Windows 2005 occupy five of these? *grin*
10 billion gigs is good for about 20 million years of MP3s or a million years of DVD video.
Provided I didn't misplace a zero there. There were lots of them.
--
This space unintentionally left unblank.
Big parts of the universe would compress pretty well, I bet.
--
This space unintentionally left unblank.
I worry about, its needing a tape autochanger the size of a house to back it up on for when I lose it down the back of the sofa.
Bob.
And it will make a hot drink which is almost totally but not quite unlike tea...
Oh please,
Do these people even know what they are talking about? Hype hype hype, but not a sensible sentence to read. It reminds me of the optical processor debacle: computing at the speed of light! while they conveniently forgot that both electrical signals in todays chips and photon in the "computer of the future" go at exactly the same speed in semiconductor materials (about 1/3 of the speed of light in vacuum) and metals (about 2/3 of the speed of light in vacuum), as they are both form of electromagnetic radiation.
Who would need such a hypothetical device, and what data would fill this black hole of a memory - and how would you index it - or transmit it, for that matter.
It would take a lifetime to fill.
Maxwell rules !
Michael
Research is what I'm doing when I don't know what I'm doing.
This seems to be another example of the focus of the computing industry: create better and better hardware but use it to run software which hasn't really changed much in the last decade.
People are always making this sort of claim, yet no-one ever provides any evidence to back it up. I thoroughly disagree with your assertion. I say that software has changed a lot in the last decade.
Ten years ago, apps were small, slow, and lacked features. I couldn't do real-time video editing in software, heck I couldn't even playback video in software. Nothing as complex as a web browser existed ten years ago. I didn't have applications like Photoshop or Gimp that allow me to perform very sophisticated image manuipulation.
Try running some 10 year old software on today's hardware. It runs faster, but it feels archaic, feature-less and flat compared with modern software.
And that's just looking at the outside of the software. If you look at the code, its also changed radically, with the introduction of object-oriented programming and large-scale software engineering.
Just seems to be another excuse to create sloppy programs/bloatware.
What you call bloatware, other people refer to as fast, stable, feature-rich software. Of course there are bad applications out there, just as there were ten years ago or twenty years ago. But there are also whole classes of application that just weren't possible a decade ago, not just because the hardware has improved, but also because we understand how to build large-scale software like never before.
Sailing over the event horizon
Anyway there isn't really a lot of point in saying look i have this much data if it isn't easily accessible.
When I blow my nose each of the dna strands in each moleucule will contain about a gigabyte of data each but until it comes with a usb interface how am i supposed to use it?
...my dog drank my homework!
LOAD "SIG",8,1
LOADING...
READY.
RUN
Furthermore, their has been a massive infrastructure shift in the last few years that, while not blatantly visible, will probably radically change the way we program and use computers.
Remember not so long ago when an application was a monolithic binary with a few config files? Now that same app is more likely to consist of a tiny frontend loader for multiple task-specific libraries, several of which may be used by multiple programs.
Back in The Day, "late binding" was a neat idea, but horridly inefficient. Given a few years to work out some of the bugs, coupled with incredible hardware advances, we now have some pretty usable object systems.
Think about Bonobo (sp?). Want online docs for your program? Throw together some HTML and call a globally-available browser. Noone really needs to write their own text editor anymore. The same holds for many other commonly-used functions.
This move has been underway for quite a while. It's been quite some time since an app needed to implement its own network stack. However, only recently have such high-level constructs been available as linkable objects.
This is part of what those CPU cycles and RAM bytes are used for. All of this linking, dynamic loading, and these powerful components take some horsepower. However, no coder in his right mind would go back to the old way without very specific reasons (embedded controllers, bare-metal recovery systems, etc.).
Dewey, what part of this looks like authorities should be involved?
please tranlate this for me
No indication is given of which calculations for data density the tiny vial pictured would be able to hold that much
Certainly!
En Français:
On ne donne aucune indication dont les calculs pour la densité d'enregistrement de données la fiole minuscule décrite pourraient tenir cela beaucoup
Auf Deutsch:
Keine Anzeige wird gegeben, von der Berechnungen für Datendichte die kleine Phiole, die dargestellt wurde, können würden, das viel anzuhalten
In italiano:
Nessun' indicazione è data di cui le calcolazioni per densità di dati la fiala molto piccola descritta potrebbero tenere quello molto
Em Português:
Nenhuma indicação é dada de que os cálculos para a densidade de dados o vial minúsculo retratado poderiam prender muito isso
En Español:
No se da ninguna indicación de la cual los cálculos para la densidad de datos el frasco minúsculo representado podrían sostener eso mucho
Have a nice day!
--
--
Mod up a post Rob doesn't like and you'll never mod again
Imagine replacing all of these volumes with a tiny vial on the bookshelf. Sterile, barren shelves. Sure, the world is now at your fingertips, but the whimsy is gone. It's just not the same.