- Researchers have developed new computer logic expected to boost energy efficiency to 100 times that of existing technologies.
- The new technology uses mechanical strain to process information and combines multiple information-carrying approaches.
- MEFA logic technology can significantly reduce the energy required for computation while maintaining speed and functionality.
- The project is part of a DARPA program exploring new computing methods, aiming to surpass CMOS systems.
US researchers are developing a new type of computer logic that uses nanoscale mechanical strain to process information, ultimately promising energy efficiency up to 100 times greater than existing technologies. The project received $10.6 million in funding from the Defense Advanced Research Projects Agency (DARPA) and is led by Dr. Asif Khan, an associate professor in the Georgia Tech School of Electrical and Computer Engineering, along with his team. The funding comes from the "Fast and Curious" program, which aims to create logic circuits at least 100 times more efficient than current state-of-the-art technologies.
Khan stated: "Whether it's charge, light, magnetic field, or strain, computing doesn't care how information is conveyed, as long as it reliably represents ones and zeros." He added: "Our approach explores a different path by fusing multiple information-carrying modalities."
How the New Computing Technology Works
Traditional computer chips improve efficiency by moving electrical charges through billions of transistors. These transistors act like tiny switches, turning on and off to represent binary 1 (yes) and 0 (no). However, the research team believes that mechanical strain may offer another way to process data. Their technology, called Mechanically Enhanced Ferroelectric Activation (MEFA) logic, uses ferroelectric materials that slightly expand or contract when voltage is applied. This motion is amplified and transferred to an adjacent semiconductor channel to control current flow and switch between logic states.
Dr. Khan noted: "A new computing technology only becomes truly useful if it can ultimately be incorporated into the broader semiconductor ecosystem. We've considered manufacturability and integration from the very beginning." According to Khan, this approach can significantly reduce the energy required for computation while maintaining the speed and functionality needed for practical systems.
Background and Goals of the DARPA Program
The project is part of a DARPA program exploring new methods of computation, with the goal of investigating technologies that can surpass current complementary metal-oxide-semiconductor (CMOS) systems. CMOS currently forms the basis of nearly all modern computer chips, from smartphone processors to powerful computing hardware used for artificial intelligence. CMOS is a small battery-powered memory that stores critical system information, such as date, time, and hardware configuration, and tracks basic input/output system (BIOS) settings to control how a computer interacts with its hardware components.
DARPA's "Fast and Curious" program challenges scientists to create new devices, materials, and computing architectures that can operate more efficiently while remaining compatible with advanced semiconductor manufacturing. This effort, led by Georgia Tech, also includes researchers from Rice University and the University of Southern California, as well as scientists from Northrop Grumman Space Park Foundry. The team aims to advance MEFA technology from individual devices to increasingly complex circuits. Ultimately, they hope to demonstrate a practical computing platform that significantly reduces energy requirements. Khan summarized in a press release: "What's exciting isn't just the device physics, but the possibility of combining extremely low energy consumption with the speed and functionality required for practical computing systems."
Item Specification Processor MEFA Logic Energy Efficiency Improvement 100 times
Potential Impact of MEFA Technology on Future Computing
The development of MEFA technology represents a significant breakthrough in computing technology, particularly in terms of energy efficiency. As demand for energy efficiency continues to grow, this technology can not only reduce energy consumption but also provide a new direction for future semiconductor technology while maintaining computational speed and functionality. DARPA's support demonstrates the government's emphasis on innovative computing architectures, which could drive industry-wide advancement and promote more environmentally friendly computing solutions.