Image: George Retseck
In Brief
- A radical new design for computer data storage called racetrack memory (RM) moves magnetic bits along nanoscopic “racetracks.”
- RM would be nonvolatile—retaining its data when the power is turned off—but would not have the drawbacks of hard disk drives or present-day nonvolatile chips.
- Chips with horizontal racetracks could outcompete today’s nonvolatile “flash” memory. Building forests of vertical racetracks on a silicon substrate would yield three-dimensional memory chips with data storage densities surpassing those of hard disk drives.
- RM is up against several other new kinds of memory under development.
The world today is very different from that of just a decade ago, thanks to our ability to readily access enormous quantities of information. Tools that we take for granted—social networks, Internet search engines, online maps with point-to-point directions, and online libraries of songs, movies, books and photographs—were unavailable just a few years ago. We owe the arrival of this information age to the rapid development of remarkable technologies in high-speed communications, data processing and—perhaps most important of all but least appreciated—digital data storage.
Each type of data storage has its Achilles’ heel, however, which is why computers use several types for different purposes. Most digital data today, such as the information that makes up the Internet, resides in vast farms of magnetic hard disk drives (HDDs) and in the HDDs of individual computers. Yet these drives, with their rotating disks and moving read/write heads, are unreliable and slow. Loss of data because of so-called head crashes occurs relatively often. Regarding speed, it can take up to 10 milliseconds to read the first bit of some requested data. In computers, 10 milliseconds is an eon—a modern processor can perform 20 million operations in that time.
Already a Digital subscriber? Sign-in Now
If your institution has site license access, enter here.
See what we're tweeting about
10 Comments
Add CommentBubble memory is ressurrected. Even in its "heyday," bubble memory was a flop. It was slow and expensive. Hard drives are continuing to get faster and packing data more densely - and there's no reason to believe this is going to change any time soon.
Reply | Report Abuse | Link to thisWhile volatile RAM needs power constantly, great strides are being made in battery technology. I wouldn't be at all surprised to start seeing RAM being sold with an on-board (possibly even integrated) rechargeable battery which would, for all intents and purposes, make it "non-volatile."
Is this a science article or an infomercial?
Reply | Report Abuse | Link to thisExcept that the larger hard drives are exhibiting higher failure rates than the smaller ones used to. That could be a reduction of manufacturing quality (Seagate, for example, no longer offers a five year warranty on all of the drives) or it could be because we're reaching some limits. Regardless, as the only mechanical system in a computer, hard drives have always been the most common single point of failure as well as the largest energy sink. For both reasons we've needed a change for a long time, it's hopeful we'll see it soon.
Reply | Report Abuse | Link to thisGreat article! I first read it in my print copy of SciAm and again on-line on following a link from a Sigma Xi (The Scientific Research Society and publisher of American Scientist magazine) newsletter link.
Reply | Report Abuse | Link to thisI did some market research on data storage technology back in the early 1980s, and though not a physicist or materials scientist, I worked for a company that (then) made magnetic oxides as a chemist, so I have a wee bit of background (with emphasis on the wee).
Even then the demise of the HDD was predicted!
The fact that the long touted demise of HDD has been exaggerated implies that it pays to never underestimate the power of people to push technology to its competitive limits and to exploit ever cheaper ways of improving data storage.
However, it is equally unwise to ignore competitive threats based on fundamentally new ways of thinking about a problem or need. The history of technology is rife with stories and accounts of revolutionary approaches overturning the status quo. This occurs most easily when entry barriers such as capital investment for a nascent industry to make inroads are not prohibitively high. This may well be the case here for non-mechanical storage technology to replace the ubiquitous HDD.
To respond to galaxy_man's comment on whether this was a science article or infomercial, I can only comment that he must be somewhat put off by the use of the first person in what is a rather well-written and readable narrative.
Some view science as only legitimate, when it is presented in a dry and unemotional context. I do not take that view, however, as anyone who has ever attended a Research Conference retreat such as a Gordon Conference or the Experimental Nuclear [research] Conference (devoted to NMR) know.
Scientists are people too, and they have a personable side, along with perspectives on their fields that only one who has devoted themselves to the field for long periods can have.
This article by Dr. Parkin shared some of that perspective.
Thus it was, in addition, to be wonderfully concise and well-written, as if you were at such a conference and talking informally to someone who both knows intimately of which he speaks, but who is also accomplished and well recognized as a leader in his field.
Refreshing, I say. Refreshing.
Bravo, Dr Parkin. Bravo. Well done.
Referring to this:
Reply | Report Abuse | Link to this"The bits themselves zoom along their racetrack, passing a read/write head at a fixed location beside the wire."
This theory suggests similar concept is in use on brains also:
http://onhowthebrainfunctions.wordpress.com
The difference is the use of analog waves, not bits.
Rgds,
Tomi
10 cents per Gigabyte stored on flash memory? The middlemen certainly seem to be taking a big slice of what consumers pay...
Reply | Report Abuse | Link to thisWell, Scientific American is quite wrong about phase-change memory. BAE Systems has no customers for its PRAM monstrosity and is still in qualification tests with the military. Numonyx, of course, has made no "commercial sales" of Alverstone. Just ask them to buy a chip or two (or 10,000 chips, which is the typical minimum commercial order quantity) or simply request a datasheet without a NDA! Or just ask them to show you any product on the market that uses their chip! Let's see what happens! Phase-change memory is a techno-Ponzi, and Scientific American fell for it!
Reply | Report Abuse | Link to thisMore about PRAM aka PCM and why it is a techno-Ponzi here:
http://ecdfan.blogspot.com/2009/05/how-to-spot-fake-samsung-and-pram.html
Phase-change memory will never be commercialized in volume as it is horribly slow in write, has poor density, and costs too much to make.
The concept here is quite interesting. However, the article reads a bit like an article in the June 1971 (yes, 1971) Sci Am. In that issue, you'll find an article about a cool IBM invention called magnetic bubble memory. In 38 years, will anyone remember this article? Perhaps, but we must assume that it will be someone other than IBM that has commercialized the concept.
Reply | Report Abuse | Link to thisIt would be interesting to know if this technology can avoid the effects of quantum theory? Or is nudging electrons one at a time strictly out of the range of quantum effects?
Reply | Report Abuse | Link to thiswant to be member of SciAm.......
Reply | Report Abuse | Link to this