Print 13 comment(s) - last by codeThug.. on Sep 19 at 4:10 PM

Nanowires store data for 100,000 years and read 1000 times faster than current data storage methods

With all the fuss about solid-state drives ever since CES 2007 -- centering on faster boot times and reducing random access times in accessing data stored on a solid-state memory module -- it is still only a matter of time until the next big discovery makes SSD technology obsolete.

A researcher at the University of Pennsylvania named Ritesh Agarwal and his colleagues have developed nanowires capable of storing computer data for 100,000 years and retrieving data 1000 times faster than current storage systems.

The nanowire technology also consumes less energy and requires less space than current memory technologies. “This new form of memory has the potential to revolutionize the way we share information, transfer data and even download entertainment," said Agarwal.

The technology is based on a self-assembling nanowire of germanium antimony telluride, which is a phase-changing material that switches between amorphous and crystalline structures -- the key to read/write memory. Agarwal also claims in tests the data writing, erasing and retrieval process lasted a mere 50 nanoseconds and consumes only 0.7 mW of electricity per bit.

The technology is said to be scalable to the terabit-level. At this time, there is no way to know when or if this technology will ever see mass production and ultimately replace current forms of flash and magnetic storage in computers.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

Blazing Speed = Blazing Hot?
By cupocoffee on 9/19/2007 12:10:07 AM , Rating: 3
Alright, .7mW = .0007W and a fast HDD today is 3 Gbs. If reading, writing and erasing take .7mW/bit on this thing, and it runs as fast as todays HDD's (granted I believe 3 Gbs is for reading) then the power usage would be:

3 Gbs = 3,000,000,000 bits/sec

.0007 * 3,000,000,000 = 2,100,000 Watts

I'm going to need a new power supply...

RE: Blazing Speed = Blazing Hot?
By 9nails on 9/19/2007 1:35:41 AM , Rating: 3
Perhaps, but that assumes a 3gb parallel read/write. Which would be astounding to say the least. But be the analogy to 3 billion heads read/writing each second on your current hard drive. Most likely this will read/write in serial at those speeds. Which would more closely match an analogy of your CD/DVD drive.

(Not to nit pick, but the fastest single drive I know of (Cheetah 15K.5 SCSI U320) reads on average at 135 MB/s, or the equiv. of 1.08 gb/s. The consumer grade SATA II bus can accommodate the speeds you quoted, but no single SATA drive that I'm aware of can fill that bus yet, and to include even under burst conditions.)

RE: Blazing Speed = Blazing Hot?
By Nightskyre on 9/19/2007 9:34:23 AM , Rating: 4
You're missing a power of negative nine.

Nano is the standard prefix for 10^-9.

This says the data retrieval lasts 50 nanoseconds.

Since a nanosecond is .000000001 seconds, you can then say it takes .00000005 seconds per bit. Serialized data transfer means you're talking about 20,000,000 bits/sec using that sustained .0007 watts you previously mentioned. Doing a little more math, that means it would require 150 watts to pull 3Gb/s. As 9nails pointed out, the fastest harddrive is actually only processing at appx 1.08Gb/s, which means you're looking at closer to 50 watts.

Of course, that's assuming linear growth on the power/performance scale, which is a big assumption.

By AnnihilatorX on 9/19/2007 9:46:49 AM , Rating: 2
3Gbps for a HDD today?
I am happy to buy your HDD!

SATA-II supports 3Gbps but a top of the range 1TB HDD only has 75MBps sustained transfer rate.

that is about 75000 * 8 = 600,000 bits/sec
600,000 * 0.0007 = 420W

Also, you need to get your facts right about SATA transfer rates:

From wikipedia:
SATA/150 communicate at a rate of 1.5 gigabits per second (Gbit/s). Taking into account 8b10b coding overhead, the actual uncoded transfer-rate is 1.2 Gbit/s, or 150 megabytes per second (MB/s)....,

Therefore even SATA-II has a transfer rate of 300MBps
which is 2.4Gbps. No HDD in the world, SCSI nor SSD can sustainably transfer info at that speed.

Hmm Interesting...
By rsasp on 9/18/2007 7:27:17 PM , Rating: 2
when will it reach consumer in a affordable price? the year 2025?

RE: Hmm Interesting...
By Randy Marsh on 9/18/2007 10:46:12 PM , Rating: 2
the same year germanium antimony telluride prices come down

RE: Hmm Interesting...
By 9nails on 9/19/2007 1:14:03 AM , Rating: 3
Please remind me of this a few months before this happens. I'd like to buy stock in the mining companies that produce german anonaminity tollride whatchamacallits.

RE: Hmm Interesting...
By codeThug on 9/19/2007 4:10:58 PM , Rating: 2
Personally I've never been to Germanium or Antimony, but I do know that lift ticket prices at Telluride are quite affordable.

How expensive will this technology be?
By neothe0ne on 9/18/2007 6:55:38 PM , Rating: 2
Sounds very nice, and 100,000 years is a long time, but how reliable would this techonology be?

RE: How expensive will this technology be?
By Alexstarfire on 9/18/2007 7:01:49 PM , Rating: 3
Hopefully very, but since they don't really specify on how it's actually done, we can't say yet.

Anyways, whatever happened to carbon-nanotubes? Weren't they supposed to revolutionize the memory industry?

By fibbeh on 9/18/2007 8:30:20 PM , Rating: 2
The problem with nanotubes is assembling them into proper ordered arrangements at the moment. We can't currently manipulate them efficiently by mechanical means since each drive would need billions of perfectly aligned nanotubes so researchers are working on methods for self-assembly.

skeptic in general
By kevinkreiser on 9/18/2007 9:05:19 PM , Rating: 3
I'm a bit of a skeptic about many things, but especially these types of news. Is it just me or does any else notice that every couple of weeks we get a post about a new revolutionary storage technique? I mean every time I read one I think about how cool it would be in practice, but that's the thing, it's never in practice, it's always just the same old "look at this research I did so I could get a paper published and speak at a conference." Of course it has to start somewhere right? But jesus, could someone just put the money into making one of these ideas a reality for the consumer?

RE: skeptic in general
By Captain Orgazmo on 9/18/2007 9:50:00 PM , Rating: 3
The problem is that these discoveries are made in university labs by researches. They do not have the money nor the means to take a completely new technology from inception all the way to production. For something like this I imagine the cost would be in the billions of dollars and over several years. Companies that do develop such technologies have to pick and choose what they develop, as they don't have money to burn, should an idea not pan out. That of course is the nature of free enterprise, and obviously all good things have a downside. This of course affects all facets of technology, from computers, to cancer drugs, to airplanes, and so on.

And don't worry these things will be made eventually; look at the relatively recent, and absolutely revolutionary invention (of its time) of the Compact Disc.

"If a man really wants to make a million dollars, the best way would be to start his own religion." -- Scientology founder L. Ron. Hubbard

Copyright 2015 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki