New Highly-Efficient Microprocessor Developed by Researchers with Superconductors
Researchers based at Yokohama National University in Japan have developed a microprocessor that is 80 times more efficient than the most advanced microprocessors available today.
Computing power is one of the largest driving factors for technological advancement today. As the need for computing power grows indefinitely, there's a finite amount of energy that can be consumed for these purposes. This brings up one of the core problems around increasing connectivity, the energy demands of our growing computer infrastructure.
For example, many modern data centers have to be geographically based near water simply to be able to supply enough cooling to the computers. This limits the operability of advanced computing infrastructure. Decreasing the energy demand for computing is the fastest way to solve this problem – the new microprocessor might do just that.
The team's research was published in the IEEE Journal of Solid-State Circuits and it offers up the details of exactly how the team achieved such an efficient microprocessor utilizing superconductors.
The biggest constraint around the new microprocessor is the conditions that are necessary to meet for it to operate efficiently, something the researchers note they will be tackling next.
Utilizing superconductors for the digital electronic structure allowed the team to optimize the microprocessors. The process that they used is called adiabatic quantum-flux-parametron, or AQFP. This process serves as a stepping stone for low-power and high-performance microprocessors.
Traditionally, superconductors require a great deal of cooling to operate in efficient means, making them unfit for use in efficient microprocessor designs after the power needs for the superconductor is taken into account. However, the team found that their superconductor-based microprocessors was still 80 times more energy-efficient, even after taking into account the required superconductor energy for the AQFP device
After this initial proof of concept demonstrated in the paper and research, the team is now working on making improvements to the device, focusing in on scalability so it can operate in a wider range of use cases.
You can read the entire study in IEEE here.