Nanoscale Switches in Memory 140
Frans Faase writes "At the university of Boston, researchers are using nano-scale mechanical switches as a novel technology for building memory. These switches are extremely small, require only femtowatts of power to switch, but still can switch at speeds of 23.57 megahertz. And they are expected to become even smaller and faster and are expected to overcome the theoretical limit of 100 gigabits per square inch capacity for magnetic media."
sweet (Score:5, Funny)
Re:sweet (Score:4, Funny)
Yes, a blister pack containing six nanoswitches is $2.79.
So that means (Score:4, Funny)
One day my old vhs tapes could store all of my pr0n?
Let me see my old T160 [high-techproductions.com] at 1075 feet * 12" *
Re:So that means (Score:2, Insightful)
Anyway, no matter who you are, you have to admit there are several awesome uses of this storage capacity. Who knows, maybe the media some of us record digitally isn't illegal or perverted?
For example, my uncle happens to own an enormous amount of DVDs, CDs, records, etc.. He is a movie/song buff. He has an entire room ded
Re:So that means (Score:1)
645 teraBITS, not bytes. 81 teraBYTES.
(about 100,000 700MB movies)
What a clatter they must make. (Score:1)
do they come in olive? (Score:3, Funny)
Re:do they come in olive? (Score:5, Funny)
Re:do they come in olive? (Score:1)
There is no olive.
Re:do they come in olive? (Score:1)
Are we regressing? (Score:5, Interesting)
Re:Are we regressing? (Score:5, Interesting)
Re:Are we regressing? (Score:2)
Re:Are we regressing? (Score:3, Interesting)
70's, 80's and 90's didn't really happen.
--
Re:Are we regressing? (Score:1)
Nope, in terms of computer science, they sure didn't (well, the 80's through today).
Bugs (Score:1)
Re:Are we regressing? (Score:2)
A single nano-watt. (Score:5, Interesting)
I wonder if this technology coluld be applied to making flying smart dust with silicon cilla?
Re:A single nano-watt. (Score:1)
femtowatts? (Score:4, Funny)
Re:femtowatts? (Score:5, Informative)
Re:femtowatts? (Score:2)
Re:femtowatts? (Score:1)
Unless my wife is defining it. Then it's
One Femtowatt = one quadrillion Watts
And if the Femtowatt is experiencing that time of the month, every single one of those Watts is walking on eggshells, you better believe it.
Mechanical == Achilles' Heel (Score:4, Interesting)
Re:Mechanical == Achilles' Heel (Score:3, Insightful)
The closer they are to the atomic scale, the more durable they will be.
Just considder molecules and atoms.Memory built from single atoms will never wear. I think that is the whole idea.
But they will suspectible to errors.
Re:Mechanical == Achilles' Heel (Score:1)
Re:Mechanical == Achilles' Heel (Score:1, Interesting)
Re:Mechanical == Achilles' Heel (Score:4, Informative)
Re:Mechanical == Achilles' Heel (Score:1)
Re:Mechanical == Achilles' Heel (Score:5, Interesting)
Re:Mechanical == Achilles' Heel (Score:2)
The article describes the design as a "beam-and-pad design ... , carving the switches from wafers made of single-crystal layers of silicon and silicon oxide". I would guess the beam does not have anything we'd recognise as a hinge, an
Re:Mechanical == Achilles' Heel (Score:3, Informative)
Each of these switches is probably smaller than any particle shedded by normal wear-and-tear, and also smaller than the surface features that the whole concept of friction is based on.
Actually, friction increases with decreasing size. For nano-sized particles friction is one of the dominant effects and often cause microelectromechanical devices to fail due to "stiction"; one piece of the machinery semi-permanently sticking to another due to van-der-Waals forces.
Re:Mechanical == Achilles' Heel (Score:2)
Re:Mechanical == Achilles' Heel (Score:2, Insightful)
Re:Mechanical == Achilles' Heel (Score:2)
Tensile strain operates differently in microscopic mechanisms.
Re:Mechanical == Achilles' Heel (Score:2)
At such small sizes, even at very very high switching frequencies, I doubt the devices generate much friction or heat, meaning they won't wear out very quickly.
Radio Interference (Score:4, Interesting)
I'm no electronics whiz, but if we can start making millions upon millions of devices that can resonate at higher frequencies, what possible interference will this cause with radio-communications?
Is there an electronics nerd/engineer on here that can clarify that for me?
I am not an engineer, but... (Score:2)
eric
Re:I am not an engineer, but... (Score:1)
They claim a few "femtowatts" of power... throw a few million of them together, and all of a sudden you've got a resonating device easily in the hundreds of milliwatts... Enough to interfere with the gigahertz cell phone bands.
Re:I am not an engineer, but... (Score:2, Informative)
Re:I am not an engineer, but... (Score:2)
Re:I am not an engineer, but... (Score:2)
Just thinking in terms of backround noise today from TV's, cell phones, microwaves, etc today I doubt this new ram technology is going to be much of a big problem. Most technology which is really reliant upon radio waves has the sense to put a
Constructive Interference... (Score:1)
It seems, to me anyways, that these devices would be aligned/manufactured/constructed in some sort of grid/lattice layout that could possibly create interference by amplification of wavelengths... if these devices are spaced an equal number of wavelengths apart, or some sort of fraction thereof, you'll get frequency amplification... conversely, you could create a subtraction effect by placing these things on opposite waves of the wavelength
(Excuse my poor explanation,
Re:I am not an engineer, but... (Score:1)
The researchers clearly state that while right now these things are working at ~23 mhz, the technology supports (in the future) operation well into the mega and gigahertz bands. And while something operating at a femtowatt @ 23mhz really isn't an issue, due to the incredible length of a wave in that portion of the spectrum (from memory, a 23mhz wavelength is something like the length of a football/soccer stadium), when you get into gigahertz stuff, low power in close range
Re:I am not an engineer, but... (Score:2)
Google knows all. [google.com]
Re:Radio Interference (Score:1)
Re:Radio Interference (Score:1)
Re:Radio Interference (Score:2)
You have to get electric or magnetic fields to have any chance of interfering with radio.
Re:Radio Interference (Score:1)
When I posted this, I wasn't thinking about shielding at all; stupid me.
Call me a karma-whore if you want; not like I really care.
Re:Radio Interference (Score:1, Insightful)
RAM or Storage? (Score:1)
Re:RAM or Storage? (Score:2)
Hopefully both. And if not this technology then one day something that is fast enough and dense enough to do the job. CPU power is one area where we can increase the speed of our systems. A Gigabye/Terabyte storage mechanism that could replace RAM, operate at the speed of RAM, or something close to the clock speed of the CPU would have terrific bandwidth potential, and give us a huge increase in overall performance. And being non-volatile hopefully it wouldn't need the same power requirements as todays
Re:RAM or Storage? (Score:1)
backwards and sideways (Score:1)
Will be interesting to see if this could become commerical.
Confused (Score:5, Insightful)
Re:Confused (Score:4, Insightful)
Assuming that this technology is non-volatile and as fast or faster than the high-end hard drives, it would make the perfect replacement for the hard drive in the storage niche, which is currently one of the biggest bottlenecks (and one of the few remaining sub-systems with moving parts) in modern computers.
The technology this will have to beat is MRAM [wikipedia.org], which is both non-volatile and blazing fast.
Re:Confused (Score:2)
Re:Confused (Score:2)
Re:Confused (Score:1)
Re:Confused (Score:3, Informative)
What that means is that the cost of NVRAM drives will always remain on the same order of magnitude of cost / capacity as RAM, whereas you need better than 1/100th of that cost/capacity to compete with hard drive storage. That $70 for one GB of FLASH storage simply doesn't s
Sparks? (Score:4, Interesting)
While the article mentions these switches being extremely robust, what have they done to address some of those older issues?
Re:Sparks? (Score:2, Informative)
Nanoscale is ATOM scaled, and the mechanical issues are difficult but ENTIRELY different from the mechanical issues of machines 10000 times larger.
At the atom level, chemistry is the right frame of mind, not thermodynamics/friction/dust/surface quality.
Re:Sparks? (Score:4, Interesting)
The smaller switches considered here would probably be much more reliable.
Re:Sparks? (Score:2)
Among the points they brought up was that mechanical switches are unreliable. Sparks can fly and generate enough force to destroy the switches.
From the parent:
Consider the TI micro-mirror display chips. They are mechanical and seem to last quite a while, and a single failure creates a noticeable display flaw.
While I understand why you bring up these devices, this doesn't address the sparks concern in the grandparent post, only the mechanical reliability of continually flexing the
Re:Sparks? (Score:2)
Re:Sparks? (Score:2)
Having now read the article, it doesn't look like it. They talk of clamping both ends of the beam and then vibrating it to generate the two states (I assume by amplitude of vibration). What isn't clear is how they are then "reading" the state of the beam. What really has me confused now is if they are dependant on the vibration...how can they retain a state without power to keep vibrating? If they were flipping a swit
Re:Sparks? (Score:2)
Roger
Boston University (Score:2, Informative)
BU Alum '84
Re:Boston University (Score:1, Troll)
Worse, they never explained that simple grammer mistakes would negate the force of any written public argument. Nor, did they explain how to use widely-available computer tools to catch these little things before publishing.
I remember every day in October 1972 passing the BU main dorm near Kenmore Square where the students had placed huge ind
Re:Boston University (Score:1)
Re:Boston University (Score:2)
University of Boston? (Score:1, Informative)
Do editors even know what I'm talking about?
Re:University of Boston? (Score:2, Funny)
Or University of Indiana. Damn, nobody told me about the name change. I wonder if the expense form I'm filling out is still valid....
Re:University of Boston? (Score:2)
Get the name right (Score:2, Informative)
Its Boston University, my alma mater.
Re:Get the name right (Score:3, Funny)
Re:Get the name right (Score:4, Funny)
The University of Boston offers, among couses:
- KAR01 - Defensive Driving
- KAR02 - Offensive Driving
- KAR02 - Navigation without street signs
- ATH01 - Philosphy of Baseball - Why being #2 is better
- ATH02 - Basketball -How can so many black guys can still be call Celtics
- LIN01 - Linguistics and Pronunciation - Dialects of Somerville, Charlestown, South Boston, and Jamaica Plain compared
- BUS01 - How to get rich in the University business
- BUS02 - How to get rich in the Politics business
- BUS03 - How to get rich in the Trashy Soveniers business
- BUS04 - How to get rich by renting slum apartments to students
And many more!
Re:Get the name right (Score:2)
Re:Get the name right (Score:2)
Oh good..... (Score:5, Funny)
This is just a press release! (Score:4, Interesting)
BTW It doesn't look like one of these things would fare well if you dropped it.
Wrong reference (Score:2)
Re:Wrong reference (Score:2)
and we laughed... (Score:3, Funny)
Reason? (Score:1)
What makes the transfer rate limit for magnetic media 100 gigabit/inch? Why this specific number? Does anyone know?
Re:Reason? (Score:1)
This is based on my memory though, and not on direct reference from any authentic source though, so consider that my reasons may not be accurate.
--dv
Re:Reason? (Score:1)
It's another limitation of the BIOS API.
Re:Reason? (Score:1)
That's not transfer rate, that's density. FWIW, 100 gigabits per square inch comes out to 316,227 magnetic domains per linear inch. Those are awfully small donuts.
Why this specific number? Does anyone know?
I presume it's related to the minimum size of a magnetic domain that can be changed (magnetized either direction) without changing the adjacent magnetic domains.
Re:Reason? (Score:1)
one way to get around this (forget about IBM's pixie dust) is to do perpedicular recording. there you can pack muck higher linear densities, but
Google Feynman Drexler nanodot.org. (Score:4, Interesting)
Since Feynman's speech, things have been shrinking steadily. In the days when Feynman was a child, things were manufactured on the scale of one meter, which is approximately person size (4). At the time he gave his famous talk, technological accomplishments included vacuum tubes, which are measured on in millimeters (5). Currently, our lives are full of things that are built on a scale one thousand times smaller. Micrometers are the scale upon which today's computer components are measured (6). A thousand times smaller yet is "the scale where atoms become tangible objects" (7). This is the goal of nanotechnology, where the building of nanomachines will be realized.
The National Technology Initiative of the year 2000, which proposes funding $270 million worth of research, outlines goals that sound remarkably like an updated version of Feynman's forty-year-old speech. Using today's scientific jargon, the NNI proposes funding improved computers, bottom-up manufacturing of strong, lightweight, materials constructed out of inexhaustible resources, and nanoengineered, molecular sized medical cures (1). Bottom-up is the current technical term for building things the way in which biological systems do, "...at the molecular level, and in three dimensions" (8). Among other things, the NNI proclaims the government's intention to fund the development of technology that will allow the "shrinking the entire contents of the Library of Congress in a device the size of a sugar cube" (1). Forty years prior to the NNI, Feynman declared that it was entirely possible to put the "Encyclopedia Britannica on the head of a pin" (2). Attaining the ability to compact vast amounts of information into a small area surely will revolutionize the dissemination of knowledge and have a profound impact upon industry. However, the potential effects of these compacting technologies upon biological systems have heretofore only been fully explored in the domain of science fiction.
I couldn't help but wonder if the story that became the 1966 film Fantastic Voyage had been inspired by Feynman's speech. In "Plenty of Room" Feynman mentioned that a friend suggested "although it is a very wild idea, it would be interesting in surgery if you could swallow the surgeon" (2). Perhaps the stated objective of the NNI to employ nanoengineered gene therapies, cancer detectors and drug delivery systems," may sound more creditable than swallowing your doctor. But put forth an equally serious manner, as the NNI, was Feynman's proposal that "... small machines might be permanently incorporated in the body to assist some inadequately-functioning organ" (2). And forty-plus years ago he offered a method by which to manufacture "such a tiny mechanism."
Feynman proposed first manufacturing a full-scale precision "master-slave hand" machine. The next step would be to use this machine to make a one-quarter sized model itself. Next, he suggested, using the smaller replica of the original machine, make tools that were small enough to make a replication of the "hands" that would again be reduced to one-quarter the size of its predecessor. He hypothesis that by continuing this shrinking proc
Lotta holes, and I don't mean charge carriers! (Score:2, Interesting)
Re:Lotta holes, and I don't mean charge carriers! (Score:2)
However...
But what about reading/writing it? It's going to take not only power, but
watch out for static electricity and dust! (Score:2, Insightful)
This Reminds Me of (Score:1)
Shock tolerance (Score:1)
"It's as though a billion tiny switches cried out in terror and were suddenly flipped"
Re:Shock tolerance (Score:2)
Why don't you read the article? They quite clearly state that these switches are extremely robust and work quite well even after being dropped.
earlier nanotube-based ram (Score:2, Interesting)
It's long-term memory. That's what it's about. (Score:5, Insightful)
You drop it, it becomes unusable due to the precision required to align the HD heads and prevent collisions.
In contrast, MEMS (micro electro mecanical switches) only move back and forth. And only by NANOmeters. And we're talking about crystalline materials here (did you know that carbon nanotubes , for example, have a much greater endurance than diamonds? AND they're flexible).
Plus, nanoswitches, even when they can be "moved", have a limited and stable range of movement. And being non ferro-magnetic makes them immune to EM interference. If you flick a switch today, it requires exactly the same action in exactly the opposite direction to alter the information. But with a floppy disk... hey, just get it near to your stereo.
Of course, do you think scientists would be dumb enough not to add an "isolation" layer to deal with vibrations? But look, to alter these thingies we'd have to talk about vibrations in the megahertz scale.
So yes, in the future, I think these babies will be the replacement for flash memories and hard disks.
This is NOT persistent (Score:2, Informative)
The effect they are using is a non linearity in the restoring force of a doubly clamped beam. It is well known that if you have a nonlinear restoring force F = kx + k_3 x^3, for sufficiently large driving power, the amplitude close to resonance becomes bistable. (This system is called a Duffing oscillator).
Nit Pick (Score:1, Redundant)