Development of extremely energy efficient microprocessors

image: photo of the AQFP MANA microprocessor matrix. MANA is the world’s first adiabatic superconducting microprocessor.
seen Following

Credit: National University of Yokohama

Researchers at Yokohama National University in Japan have developed a prototype microprocessor using superconducting devices that are approximately 80 times more energy efficient than the advanced semiconductor devices found in microprocessors in high computer systems. performance today.

As today’s technologies become more and more integrated into our daily lives, the need for more computing power is growing. As a result of this increase, the power consumption of this growing computing power increases dramatically. For example, modern data centers consume so much energy that some are built near rivers so that flowing water can be used to cool machines.

“The digital communications infrastructure that supports the information age we live in today currently uses around 10% of the world’s electricity. Studies suggest that in the worst case scenario, if there is no fundamental change in the underlying technology of our communications infrastructure, such as the hardware in large data centers or the electronics that drives communications networks, we could see its electricity consumption reach over 50% of the world’s electricity by 2030, ”said Christopher Ayala, associate professor at Yokohama National University and lead author of the study.

The team’s research, published in Journal: IEEE Journal of Semiconductor Circuits, details an effort to develop a more energy-efficient microprocessor architecture using superconductors, devices that are incredibly efficient, but which require certain environmental conditions to operate.

To tackle this power problem, the team explored the use of an extremely energy-efficient superconducting digital electronic structure, called the adiabatic quantum flow parameter (AQFP), as a building block for ultra-efficient microprocessors. – low power and high performance, and other computing hardware for the next generation of data centers and communications networks.

“In this article, we wanted to prove that AQFP is capable of convenient high-speed and power-efficient computing, and we did this by developing and successfully demonstrating a prototype 4-bit AQFP microprocessor called MANA ( Monolithic Adiabatic iNtegration Architecture), the world’s first adiabatic superconducting microprocessor, ”said Ayala.

“The demonstration of our prototype microprocessor shows that AQFP is capable of all aspects of computation, namely: data processing and storage. We also show on a separate chip that the data processing part of the microprocessor can operate up to a clock frequency of 2.5 GHz, which is comparable to today’s computer technologies. We even expect it to drop to 5-10 GHz as we improve our design methodology and experimental setup, ”Ayala said.

However, superconductors require extremely cold temperatures to function properly. One would think that if one takes into account the cooling required for a superconducting microprocessor, the power requirement would become undesirable and exceed current microprocessors. But according to the research team, this was not the case, surprisingly:

“The AQFP is a superconducting electronic device, which means that we need an additional power supply to cool our chips from room temperature to 4.2 Kelvin to allow the AQFP to go superconducting. But even in taking this additional cooling cost into account, the AQFP is still approximately 80 times more energy efficient compared to the advanced semiconductor electronic devices found in high performance computer chips available today. ”

Now that the team has proven the concept of this superconducting chip architecture, they plan to optimize the chip and determine the scalability and post-optimization speed of the chip.

“We are now working to improve the technology, including the development of more compact AQFP devices, increasing operating speed and further increasing energy efficiency through reversible computing,” Ayala said. “We are also adapting our design approach so that we can integrate as many devices as possible into a single chip and have them all operate reliably at high clock rates.”

In addition to building standard microprocessors, the team is also interested in examining how AQFPs could help in other computing applications such as neuromorphic computing hardware for artificial intelligence as well as quantum computing applications.


Yokohama National University (YNU or Yokokoku) is a Japanese national university founded in 1949. YNU provides students with hands-on training utilizing the vast expertise of its faculty and facilitates engagement with the global community. YNU’s strength in academic research of the practical sciences leads to high impact publications and contributes to international scientific research and global society. For more information, please see:

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of any press releases posted on EurekAlert! by contributing institutions or for the use of any information via the EurekAlert system.

Source link

Comments are closed.