New Device May Revolutionize Computer Memory

Traditionally, there are two types of computer memory devices. Slow memory devices are used in persistent data storage technologies such as flash drives. They allow us to save information for extended periods of time, and are therefore called nonvolatile devices. Fast memory devices allow our computers to operate quickly, but aren't able to save data when the computers are turned off. The necessity for a constant source of power makes them volatile devices.
But now a research team from NC State has developed a single "unified" device that can perform both volatile and nonvolatile memory operation and may be used in the main memory.
"We've invented a new device that may revolutionize computer memory," says Dr. Paul Franzon, a professor of electrical and computer engineering at NC State and co-author of a paper describing the research. "Our device is called a double floating-gate field effect transistor (FET). Existing nonvolatile memory used in data storage devices utilizes a single floating gate, which stores charge in the floating gate to signify a 1 or 0 in the device -- or one 'bit' of information. By using two floating gates, the device can store a bit in a nonvolatile mode, and/or it can store a bit in a fast, volatile mode -- like the normal main memory on your computer."
The double floating-gate FET could have a significant impact on a number of computer problems. For example, it would allow computers to start immediately, because the computer wouldn't have to retrieve start-up data from its hard drive -- the data could be stored in its main memory.
The new device would also allow "power proportional computing." For example, Web server farms, such as those used by Google, consume an enormous amount of power -- even when there are low levels of user activity -- in part because the server farms can't turn off the power without affecting their main memory.
"The double floating-gate FET would help solve this problem," Franzon says, "because data could be stored quickly in nonvolatile memory -- and retrieved just as quickly. This would allow portions of the server memory to be turned off during periods of low use without affecting performance."
Franzon also notes that the research team has investigated questions about this technology's reliability, and that they think the device "can have a very long lifetime, when it comes to storing data in the volatile mode."

Read More

'Nanowire' Measurements Could Improve Computer Memory

 
The nascent technology is based on silicon formed into tiny wires, approximately 20 nanometers in diameter. These "nanowires" form the basis of memory that is non-volatile, holding its contents even while the power is off -- just like the flash memory in USB thumb drives and many mp3 players. Such nanowire devices are being studied extensively as the possible basis for next-generation computer memory because they hold the promise to store information faster and at lower voltage.
Nanowire memory devices also hold an additional advantage over flash memory, which despite its uses is unsuitable for one of the most crucial memory banks in a computer: the local cache memory in the central processor.
"Cache memory stores the information a microprocessor is using for the task immediately at hand," says NIST physicist Curt Richter. "It has to operate very quickly, and flash memory just isn't fast enough. If we can find a fast, non-volatile form of memory to replace what chips currently use as cache memory, computing devices could gain even more freedom from power outlets -- and we think we've found the best way to help silicon nanowires do the job."
While the research team is by no means the only lab group in the world working on nanowires, they took advantage of NIST's talents at measurement to determine the best way to design charge-trapping memory devices based on nanowires, which must be surrounded by thin layers of material called dielectrics that store electrical charge. By using a combination of software modeling and electrical device characterization, the NIST and GMU team explored a wide range of structures for the dielectrics. Based on the understanding they gained, Richter says, an optimal device can be designed.
"These findings create a platform for experimenters around the world to further investigate the nanowire-based approach to high-performance non-volatile memory," says Qiliang Li, assistant professor of Electrical and Computer Engineering at GMU. "We are optimistic that nanowire-based memory is now closer to real application."

Read More

New 'FeTRAM' Is Promising Computer Memory Technology

 
he technology combines silicon nanowires with a "ferroelectric" polymer, a material that switches polarity when electric fields are applied, making possible a new type of ferroelectric transistor.
"It's in a very nascent stage," said doctoral student Saptarshi Das, who is working with Joerg Appenzeller, a professor of electrical and computer engineering and scientific director of nanoelectronics at Purdue's Birck Nanotechnology Center.
The ferroelectric transistor's changing polarity is read as 0 or 1, an operation needed for digital circuits to store information in binary code consisting of sequences of ones and zeroes. The new technology is called FeTRAM, for ferroelectric transistor random access memory.
"We've developed the theory and done the experiment and also showed how it works in a circuit," he said. Findings are detailed in a research paper that appeared this month in Nano Letters, published by the American Chemical Society.
The FeTRAM technology has nonvolatile storage, meaning it stays in memory after the computer is turned off. The devices have the potential to use 99 percent less energy than flash memory, a non-volatile computer storage chip and the predominant form of memory in the commercial market.
"However, our present device consumes more power because it is still not properly scaled," Das said. "For future generations of FeTRAM technologies one of the main objectives will be to reduce the power dissipation. They might also be much faster than another form of computer memory called SRAM."
The FeTRAM technology fulfills the three basic functions of computer memory: to write information, read the information and hold it for a long period of time.
"You want to hold memory as long as possible, 10 to 20 years, and you should be able to read and write as many times as possible," Das said. "It should also be low power to keep your laptop from getting too hot. And it needs to scale, meaning you can pack many devices into a very small area. The use of silicon nanowires along with this ferroelectric polymer has been motivated by these requirements."
The new technology also is compatible with industry manufacturing processes for complementary metal oxide semiconductors, or CMOS, used to produce computer chips. It has the potential to replace conventional memory systems.
A patent application has been filed for the concept.
The FeTRAMs are similar to state-of-the-art ferroelectric random access memories, FeRAMs, which are in commercial use but represent a relatively small part of the overall semiconductor market. Both use ferroelectric material to store information in a nonvolatile fashion, but unlike FeRAMS, the new technology allows for nondestructive readout, meaning information can be read without losing it.
This nondestructive readout is possible by storing information using a ferroelectric transistor instead of a capacitor, which is used in conventional FeRAMs.
This work was supported by the Nanotechnology Research Initiative (NRI) through Purdue's Network for Computational Nanotechnology (NCN), which is supported by National Science Foundation.

Read More