Professor of microelectronics developing options for improving memory technologies for storage and computing

Research at the Rochester Institute of Technology on new energy-efficient materials for computers could improve the bottleneck that often occurs when retrieving large amounts of data, hampering processing capacity and energy efficiency.

Kai Nicassistant professor of electrical engineering in RITs Kate Gleason College of Engineering, received funding to advance alternative materials and processes toward improving the way electronic devices store, retrieve, and process big data. The funding is part of the Defense Advanced Research Projects Agency’s (DARPA) Young Faculty Award program, which is given to aspiring faculty researchers who research national security technologies.

Ni is an expert in semiconductor and novel computer memory technologies† His work is in the field of ferroelectricity, an emerging field where energy efficiency and increased computing power are being developed through integration of new materials, the physical modeling of devices and more effective application of the developed devices in accelerator system architectures – the overall computer processing systems.

To calculate or extract data, users need to move data from the memory to the applications of the computing device. Once functions are completed, the data is saved again. That movement, especially of large data sets, is becoming a bottleneck for data transfer, Ni explained.

“The challenge is memory. In short, it takes more energy and time to get the data from the memory to the CPU, so moving the data from the storage takes the most energy; that is a bottleneck that we are trying to solve,” says Ni, director of the Group of Nanoelectronic Devices and Systems in RITs Department of Electrical and Microelectronic Engineering† “That’s really the goal we’re trying to achieve, and it’s not easy.”

Ni was awarded nearly $1 million for this project. Trends in this area of ​​research show promising energy-efficient processing, and work is aimed at confirming the use of ferroelectric materials to accelerate applications such as processing huge amounts of big data and to improve accelerators for artificial intelligence and neuromorphic computing Technological development. These technologies are sought after for improvements to data and signal processes, secure encryption and resource management in defense applications, but they are also being developed in a variety of sectors, from healthcare to consumer devices.

“Research to advance the use of new materials in practical and scalable microelectronic devices is critical to achieving the performance required for the next generation of computers,” says Doreen Edwards, dean of RIT’s Kate Gleason College of Engineering. “Not only will Dr. Ni’s research help develop new technologies, it will also help educate the next generation of microelectronics and microsystems engineers at RIT.”

Ferroelectrics, such as: perovskite, a semiconductor material, have only recently become a major focus of research into optoelectronic devices and technologies. They can be used to make transducers or capacitors; However, some materials are not easy to integrate without special processing, especially the ability to scale smaller, especially with the increased power requirements in today’s devices. The early findings of Ni indicate that using new materials such as hafnium oxide, an electrical insulator, which turns out to be ferroelectric, can improve processing and lead to the necessary integration of memory and logic functions of a computer.

While there have been breakthroughs in the field for using it, Ni said there are challenges in refining its integration into semiconductor devices. While it has the electrical properties necessary to improve application processing, the scaling process in a small area of ​​a transistor is key to developing and improving memory. It is a unique quality of the ferroelectric technologies and is extensively tested in AI chips for the acceleration of neural networks.

“Conventional ways of packing more transistors and packing more computing resources are no longer tenable, so how can we keep making computers even more powerful? We’re working on that,” says Ni. “It’s the right time to move into electronics, because there are so many opportunities to explore. We want to mimic the brain – how can we do this for more efficient devices? How can we move forward than the 2D chip to create 3D and stack multiple chips on top of each other? And how can you build a quantum computer – all these options are exciting. I like trying new things and I would like to see our group have a contribution in this area.”

Leave a Comment

Your email address will not be published.