Monday, July 18, 2016

Breakthrough in data storage

Tiny Hard Drive Uses Single Atoms to Store Data

It packs hundreds of time more information per square inch than best currently available technologies, study says


By manipulating the interactions between individual atoms, scientists report they have created a device that can pack hundreds of times more information per square inch than the best currently available data-storage technologies.
The working prototype is part of a decades-long attempt to shrink electronics down to the atomic level, a feat scientists believe would allow them to store information much more efficiently, in less space and more cheaply. By comparison, tech companies today build warehouse-sized data centers to store the billions of photos, videos and posts consumers upload to the internet daily. Corporations including International Business Machines Corp. and Hewlett Packard Enterprise Co. also have explored research to reduce such space needs.
The so-called atomic-scale memory, described in a paper published on Monday in the scientific journal Nature Nanotechnology, can hold one kilobyte, the equivalent of roughly a paragraph of text.
It may not sound “very impressive,” said Franz Himpsel, a professor emeritus of physics at the University of Wisconsin, Madison, who wasn’t involved in the study. But “I would call it a breakthrough.” 
Most previous attempts at encoding information with atoms, including his own, managed roughly one byte, Dr. Himpsel said. And data could be stored only once. To store new information, the “disk” had to be re-formatted, like CD-Rs popular in the ’90s.
With the new device, “we can rewrite it as often as we like,” said Sander Otte, an experimental physicist at Delft University of Technology in the Netherlands and the lead author on the new paper.
A close-up of the scanning tunneling microscope used for building the one-kilobyte atomic memory described in a new paper in the journal Nature Nanotechnology. The machine was built by Unisoku Co., a Japanese microscope manufacturer. ENLARGE
A close-up of the scanning tunneling microscope used for building the one-kilobyte atomic memory described in a new paper in the journal Nature Nanotechnology. The machine was built by Unisoku Co., a Japanese microscope manufacturer. PHOTO: TU DELFT/SANDER OTTE
The researchers first stored a portion of Charles Darwin’s “On the Origin of Species” on the device. They then replaced that with 160 words from a 1959 lecture by physicist Richard Feynman in which he imagined a world powered by devices running on atomic-scale memory.
To build their prototype, the scientists peppered a flat copper bed with about 60,000 chlorine atoms scattered at random, purposely leaving roughly 8,000 empty spaces among them. A mapping algorithm guided the tiny, copper-coated tip of a high-tech microscope to gently pull each chlorine atom to a predetermined location, creating a precise arrangement of atoms and neighboring “holes.”
Researchers used the relationship between individual atoms and adjacent empty slots to encode bits of data. The image depicts the configuration of bits that represent the letter e, plus the command cues that tell the software how to scan the grid. ENLARGE
Researchers used the relationship between individual atoms and adjacent empty slots to encode bits of data. The image depicts the configuration of bits that represent the letter e, plus the command cues that tell the software how to scan the grid. PHOTO: TU DELFT
The team also crafted a language for their device. The stored information is encoded in the patterns of holes between atoms. The atom-tugging needle reads them as ones and zeros, turning them into regular binary code.
The researchers marked up the grid with instructions that cued the software where it should direct the needle to write and read data. For instance, a three-hole diagonal line marked the end of a file.

“That’s what I really love in this work,” said Elke Scheer, a nanoscientist at the University of Konstanz in Germany not involved with the study. “It’s not just physics. It’s also informatics.”

Turning Chaos Into Data, One Atom At a Time

Researchers stored portions of Charles Darwin’s On the Origin of Species on an atomic-memory device by rearranging the order of atoms and vacancies on a grid.

Writing the initial data to the device took about a week, though the rewriting process takes just a few hours, Dr. Otte said.
“It’s automated, so it’s 10 times faster than previous examples,” said Christopher Lutz, a staff scientist at IBM Research-Almaden in San Jose, Calif. Still, “this is very exploratory. It’s important not to see this one-kilobyte memory result as something that can be taken directly to a product.”
Reading the stored data is much too slow to have practical applications soon. Plus, the device is stable for only a few hours at extremely low temperatures. To be competitive with today’s hard drives, the memory would have to persist for years and work in warmer temperatures, said Victor Zhirnov,chief scientist at the Semiconductor Research Corp., a research consortium based in Durham, N.C.
When Dr. Otte’s team took the memory out of the extremely low-temperature environment in which it was built and stored, the information it held was lost. Next, his team will explore other metal surfaces as well as elements similar to, but heavier than, chlorine, to see if that improves the device’s stability.
“There’s many combinations to play with,” he said.

No comments: