Nano-Scale Memory Fits A Terabit On A Square Inch

prostoalex writes "San Jose Business Journal talks about Nanochip, a company that's developing molecular-scale memory: "Nanochip has developed prototype arrays of atomic-force probes, tiny instruments used to read and write information at the molecular level. These arrays can record up to one trillion bits of data -- known as a terabit -- in a single square inch. That's the storage density that magnetic hard disk drive makers hope to achieve by 2010. It's roughly equivalent to putting the contents of 25 DVDs on a chip the size of a postage stamp." The story also mentions Millipede project from IBM, where scientists are trying to build nano-scale memory that relies on micromechanical components."

Re:25 DVDs?

[captain igor]

Yes, a terabit = 125 Gigabytes, which is 31.25 DVDs

Re:25 DVDs?

[Anonymous Coward]

A Terabit is 125GB. Single layer dvd discs store 4.7GB while dual layer discs store 8.5GB.

25 DVDs is a really bad comparison since the size of a dvd could vary.

Go ahead

[killa62]

Mod me -1 redundant if you like, but for people out there, but 1 trillion b= 125,000,000,000 bytes = 116 GB, or if you're a harddrive manufacturer, its 125 GB.

More information

[ploss]

More information about the company can be found at their website, http://www.nanochip.com.nyud.net:8090 [nyud.net][Coral Cache Link].

Re:Magnetic memory = Doom

[GerbilSoft]

Warm reboots don't erase memory. Cold reboots usually don't erase memory, either. (There are still fragments of what was left before after doing a cold boot.)

And as almost all data recovery people know, reformatting a hard drive using the conventional disk formatting commands don't really erase anything; they merely create new directory structures. In order to really erase a disk, you have to use something like Eraser [heidi.ie] or `dd if=/dev/urandom of=/dev/hda`.

Re:How long.

[Spy Hunter]

It is non-volatile by nature. But it is not likely to be fast enough to replace RAM. Instead it could replace Flash memory or even (depending on cost) hard drives. The real question is, how long until it's practical to manufacture and use in mass-produced products? The answer seems to be (according to the article) 2007-2010 timeframe.

Re:25 DVDs?

[XorNand]

DVD-R / DVD+R capacities are 4.7GB. However, actual pressed, dual-layer DVDs are 8.54GB (single-sided). So the analogy is a bit incomplete. A better analogy, of course, is how many hogshead of LoC does it hold?

ATM or AFM?

[fermion]

From the article it is hard to tell what they are taking about. IBM used an atomic tunnelling microscope, a reltively complicated piece of equipment that relies on the fact that quantum particles can tunnel through a potential, to move that atoms. The ATM can either be used to create a atomic scale picture of a surface, or move atoms. An atomic force microscope is simply a physical hammer that gently taps a surface and through the change in deflection creates an image. The tip on an ATM is currently so fragile I don't think it could be used to move atoms. The lifetime of a tip is pretty short just becuase of wear, and their is not way to reliable create good tips.

So we must assume they are talking about an ATM, which a largish and complicated peice of equipment. It requires a piezoelctric device to move the tip to the proper placed on the substrate. For years, such devics kept cell phones large. The ATM requires a highly senstive feeback loop to keep the current constant. And is still requires a very delicae tip that can be easily damaged. Durable tips are probably years away and involve carbon nanotubes. Tips that have a lifetime more than a few months are probably even longer away.

It is a neat idea and probably works well in the laboratory on a vibration cancelation table. How would it work on a portable in the train or in the car? Does anyone have any real details on the technology?

Google

[blcknight]

http://www.google.com/search?q=1+terabit+in+gigaby tes 1 terabit is 128 gigabytes. That is the definitive answer from google. It's not 116, not 125.

Re:25 DVDs?

[Anonymous Coward]

Last time I checked, a postage stamp wasn't a square inch either!

Re:Go ahead

[Anonymous Coward]

1 terabit = 125 GB (gigabyte)
1 terabit = 116 GiB (gibibyte)

Networking and storage use base-10, memory uses base-2.

http://mathworld.wolfram.com/Gibibyte.html

Re:Go ahead

[bohnsack]

1 trillion bits is 125 GB, whether you're a hard drive manufacturer or not, as "G" is exactly defined as 10^9 [nist.gov]. If you're interested in representing this quantity in terms of multiples of 2^(30), as in your 116, 1 trillion bits is more correctly stated as 116.5 GiB, 116.5 gigabinary bytes, or 116.5 gibibytes. See the SI spec on prefixes for binary multiples [nist.gov] for more information.

AFM

[DaleBob]

The IBM Millipede project doesn't use tunneling microscope technology (ATM, or usually STM). It uses a modified AFM tip that can be resistively heated. The hot tip pushes into a polymer surface and creates a hole. The hole can be "erased" by heating close to the surface and the region around the hole melts and fills it in. The reading is done with cold tips using regular AFM technology.

Re:Magnetic memory = Doom

[orangesquid]

That's because a 20Gig drive usually has something like 22-24Gigs of space; the extra space is used to relocate bad blocks.

If you completely overwrite a bad block, the drive's firmware is usually smart enough to move it to a new place. Reading from a bad block until you manage to get (most of) the data, and then re-writing it, will sometimes work (due to the same mechanism).

I'm told some drives are smart enough to try to "fix" bad blocks without being forced to like this, but I don't know of any (mostly likely because there would be no way to tell, since it would be user-transparent).

Occasionally, a drive will have some sort of mechanical shock which will damage a portion of the disk; often, the disk is not "failing," but has simply become partly damaged. A true "failing" disk would mean that the disk material was corroding, or that a poorly designed drive was losing its own low-level formatting from the motor's magnetic fields (I doubt that would ever happen, though), or something pretty ugly like that.

(I don't know if what I've said is completely true; please correct me if I'm talking out of my ass. It's been a long time since I've looked this up.)

Re:Hmm

[Anonymous Coward]

Tebibit...

Re:Magnetic memory = Doom

[Coolpup]

If you RTFA you will see they are talking about FLASH MEMORY. I've never seen anyone use their flash drives as their RAM. Do you know why this is? Because flash memory is far to slow to be used as RAM.

Re:OK! ENOUGH BULLSHIT NUMBERS!!!

[Anonymous Coward]

That's a nice increase but it sure as hell isn't overwhelming.

You make a good point, but there are some issues that could make this a big deal compared to hard drives, specifically heat, resistance to shock damage, production cost, power usage, lower latency and/or higher tranfer rates. A non-volatile memory with the performance characteristics of even PC133 would have a lot of very useful applications.

Sadly their website doesn't seem to have much on any this (at least what I can grab out of Google's cache, their DNS seems to be fux0red ATM).

Re:ATM or AFM?

[notmuchtosay]

As someone already mentioned i think the ATM you refer to is usually called a STM (scanning tunnelling microscope). However an AFM does not need to operate in contact (hammering) mode. There are other techniques called non-contact/lift mode. In these mode you don't sense the repulsion from the surface. You actually drive the tip near resonance and then sense the change in frequency as the tip is pulled toward the surface.