Slashdot Mirror
A 1.2 Petabyte Hard Drive?
Angry_Admin writes "Rather than spend millions of dollars for an array of hard drives when you can have all that storage on just one drive? A story at P2P.net US inventor Michael Thomas, owner of Colossal Storage, says he's the first person to solve non-contact optical spintronics which will in turn ultimately result in the creation of 3.5-inch discs with a million times the capacity of any hard drive - 1.2 petabytes of storage, to be exact. According to the article, In the past, data storage has only been able to orient the direction a field of electrons as they move around a molecule, Thomas said. "But now there's a way to rotate or spin the individual electrons that make up, or surround, the molecule," he says. He expects a finished product to be on the market in about four to five years, adding the cost would probably be in the range of $750 each."
"Rather than spend millions of dollars for an array of hard drives when you can have all that storage on just one drive?"
1. That sentence didn't make any sense.
2. So my PETABYTES of data don't all go down the tube at once.
I'd rather have the current flash technology improved as compared to that mechanical technology. I thought that's where we were heading. I guess I was wrong.
Sounds like 1.2 Petabytes of hurt if and when that thing bytes the dust.
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PT O1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm &r=1&f=G&l=50&s1=6,028,835.WKU.&OS=PN/6,028,835&RS =PN/6,028,835
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PT O1&Sect2=HITOFF&d=PALL&p=1&u=/netahtml/srchnum.htm &r=1&f=G&l=50&s1=6,046,973.WKU.&OS=PN/6,046,973&RS =PN/6,046,973
Inventions by Michael E. Thomas under U.S. Patents, # 6,028,835 2/22/00 and # 6,046,973 4/4/00 concepts in this home page are for laboratory discussion and possible licensing and sale only.
I call BS.
Seems every few months we get a story about a wonder just a few years down the road. Most never get here, and none on the original optimistic schedule.
Where are the holographics DVDs? A few years out, which is where they were a few years ago.
OLEDs are finally showing up on small displays but remember it was only a few years ago we were promised they would supplant Plasma and LCD in 'just a couple of years?' They might do it someday, but not this year.
And so on.
Democrat delenda est
Christ, how many times are we promised phenomenal increases in storage, processing power, batteries, etc that are only "4-5 years away"? IF the technology ever materializes, it's usually a shadow of its former self, offering the standard increases we're used to (Moore's Law or thereabouts, depending on the tech). This isn't news until prototype units are done and working, as far as I'm concerned.
Meanwhile, how would you access the data? What bus would be fast enough for storage of that magnitude? How do you back it up, except to other drives of its type? What's the reliability predicted to be like (especially on such a new technology)?
Lots of questions, few answers.
I don't know what kind of crack I was on, but I suspect it was decaf.
208 KB of storage for each person on this planet
And as everyone knows, 208kB should be enough for anybody.
I would assume that one could (ab)use "electron migration" to store information, provided an easy way of resetting the electrons existed. This would have the benefit of not needing any magnetic mechanisms (which may mean you could get higher densities) but it would certainly be slower to write to, and likely to read from. I would suspect that something similar will offer much better opportunities for solid-state non-volatile storage in the future, precisely because it should be capable of far higher densities.
If I recall from engineering school, this is how flash memories work; a charge is "trapped" in the gate oxide of a MOSFET (thereby making the MOSFET conduct or not when the data is read), and with current technologies can stay there for several years. The issue (besides write speed, caused by parasitic gate capacitance) is the relatively low number of write cycles before the gate oxide begins to fail. I forget the exact mechanism, but I assume it does have to do with electromigration (as opposed to electron migration) causing the trapping layer in the gate oxide to eventually puncture through to the substrate.
Fire and Meat. Yummy.
This place is starting to have the editorial standards of the National Enquirer...
1/74th of Data's full storage capacity on Star Trek
...runs at roughly 10Mhz (defined by the protagonist as "decisions per second").
Interesting, I've never heard that one before (yup, a non-Trekkie on Slashdot). So Data's got about 90PB of storage. Seems insane, right?
It's always neat to see what sci-fi authors think is going to be some insanely huge number, and neater to see how quickly those estimates seem quaint.
I just re-read Heinlein's The Moon is a Harsh Mistress. In it, the intelligent computer, who can perfectly simulate human voice, display a real-time, photorealistic face with perfect gestures complete with animated photorealistic background scenery, store most if not all of human knowledge, and generally do everything imaginable....
I'm sure this seemed really fast decades ago, yet today it's quaint. If by some miracle we could actually keep doubling hard drive capacities forever, we'll exceed Data in less than 20 years in a single 3.5" drive.
Scary, but also fun to look forward to.
Endless arguments over trivial contradictions in books written by ignorant savages to explain thunder in the dark.
for a moment. We don't just use massive storage arrays to allow us to 'access a load of data' they also provide many other benefits. Drive mirroring/parity allows you to integrate backup into your system - one physical device fails and no data is lost.
The main issue is access speed. Most data centres are continuously supplying small amounts of data to a huge number of clients. With a single unit and with a single head that's going to be a massive problem - array can simultaneously read and supply data from the different drives at the same time.
The basic problem is: you can't identify individual electrons. No way. Not ever. When they're circling an atom they're not discernible particles per se- they're an anonymous and homogenous cloud of probability. You can apply some energy and peel one electron off, but it's not like you're picking a particular electron. It's not like a bag of marbles and you're picking a particular one of a particular color. It's more like a jar of molasses and you're scooping out a spoonful.
Also electron spin isnt something that's latched to any one electron. Electrons exchange virtual photons many millions of million of times per second, which scrambles their properties.
So to beat this dead horse again: there's absolutely nothing to this story.