Next-generation computer chip with two heads

November 05, 2020

It's a major breakthrough in the field of electronics. Engineers at EPFL's Laboratory of Nanoscale Electronics and Structures (LANES) have developed a next-generation circuit that allows for smaller, faster and more energy-efficient devices - which would have major benefits for artificial-intelligence systems. Their revolutionary technology is the first to use a 2D material for what's called a logic-in-memory architecture, or a single architecture that combines logic operations with a memory function. The research team's findings appear today in Nature.

Until now, the energy efficiency of computer chips has been limited by the von Neumann architecture they currently use, where data processing and data storage take place in two separate units. That means data must constantly be transferred between the two units, using up a considerable amount of time and energy.

By combining the two units into a single structure, engineers can reduce these losses. That's the idea behind the new chip developed at EPFL, although it goes one step beyond existing logic-in-memory devices. The EPFL chip is made from MoS2, which is a 2D material consisting of a single layer that's only three atoms thick. It's also an excellent semi-conductor. LANES engineers had already studied the specific properties of MoS2 a few years ago, finding that it is particularly well-suited to electronics applications. Now the team has taken that initial research further to create their next-generation technology.

The EPFL chip is based on floating-gate field-effect transistors (FGFETs). The advantage of these transistors is that they can hold electric charges for long periods; they are typically used in flash memory systems for cameras, smartphones and computers. The unique electrical proprieties of MoS2 make it particularly sensitive to charges stored in FGFETs, which is what enabled the LANES engineers to develop circuits that work as both memory storage units and programmable transistors. By using MoS2, they were able to incorporate numerous processing functions into a single circuit and then change them as desired.

In-depth expertise

"This ability for circuits to perform two functions is similar to how the human brain works, where neurons are involved in both storing memories and conducting mental calculations," says Andras Kis, the head of LANES. "Our circuit design has several advantages. It can reduce the energy loss associated with transferring data between memory units and processors, cut the amount of time needed for computing operations and shrink the amount of space required. That opens the door to devices that are smaller, more powerful and more energy efficient."

The LANES research team has also acquired in-depth expertise in fabricating circuits out of 2D materials. "We made our first chip ten years ago by hand," says Kis. "But we have since developed an advanced fabrication process that lets us make 80 or more chips in a single run, with well-controlled properties."

Ecole Polytechnique Fédérale de Lausanne

Related Data Storage Articles from Brightsurf:

Reviewing multiferroics for future, low-energy data storage
Big data and exponential demands for computations are driving an unsustainable rise in global ICT energy use.

A new ultrafast control scheme of ferromagnet for energy-efficient data storage
Using a single laser pulse that did not switch the ferrimagnetic layer, researchers demonstrated a much faster and less energy consuming switching of the ferromagnet.

Multi-state data storage leaving binary behind
Electronic data is being produced at a breath-taking rate. Around ten zettabytes (ten trillion gigabytes) of data is stored in global server farms, and that's doubling every two years.

Robust high-performance data storage through magnetic anisotropy
A technologically relevant material for HAMR data memories are thin films of iron-platinum nanograins.

Energy-saving servers: Data storage 2.0
A research team of Mainz University has developed a technique that will potentially halve the energy required to write data to servers and make it easier to construct complex server architectures.

New approach to DNA data storage makes system more dynamic, scalable
Researchers have developed a fundamentally new approach to DNA data storage systems, giving users the ability to read or modify data files without destroying them and making the systems easier to scale up for practical use.

Scientists take steps to create a 'racetrack memory,' potentially enhancing data storage
A team of scientists has taken steps to create a new form of digital data storage, a ''Racetrack Memory,'' which opens the possibility to both bolster computer power and lead to the creation of smaller, faster, and more energy efficient computer memory technologies.

Discovery offers new avenue for next-generation data storage
The demands for data storage and processing have grown exponentially as the world becomes increasingly connected, emphasizing the need for new materials capable of more efficient data storage and data processing.

Magnetic whirls in future data storage devices
Magnetic (anti)skyrmions are microscopically small whirls that are found in special classes of magnetic materials.

Laser writing enables practical flat optics and data storage in glass
Femtosecond laser machining has emerged as an attractive technology enabling appications ranging from eye surgery to direct writing in the bulk of transparent materials.

Read More: Data Storage News and Data Storage Current Events is a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to