The advent of the era of artificial intelligence (AI) has brought new challenges and new opportunities to the microelectronics industry.
“Energy is the biggest deciding factor when it comes to advances in AI,” says Salma Vrdula. “Artificial intelligence computations require incredible amounts of electrical energy. In order for us to continue to progress, and for overall sustainability, we must find ways to reduce power consumption. No.”
Vrudhula is a professor of computer science and engineering in the Department of Computing and Augmented Intelligence, part of the Ira A. Fulton College of Engineering at Arizona State University. He and his team received a $2 million grant from the U.S. National Science Foundation (NSF) to design new microchips and systems that will power the next phase of the AI revolution.
Mr. Vrudula is joined by Marwan Klunz, Regents Professor and Kenneth von Behlen Endowed Professor of Electrical and Computer Engineering and professor of computer science at the University of Arizona, and James C. Wyant Assistant Professor of Computer and Data Science at the University of Arizona. We plan to conduct joint research with Mr. Sanmukh Kupannagari. Case Western Reserve University. Their work is part of the Future of Semiconductors (FuSe2), an NSF initiative aligned with the CHIPS and Science Act of 2022 to address the global A groundbreaking collaboration between U.S. government agencies and industry partners to address the microchip shortage.
New ideas driving new technology
For many years, microchips have been designed by hand. Thanks to the work of operations research and design automation experts such as Vrudhula, software has been developed to take over this plan. Engineers can now input their design and performance expectations into sophisticated software programs and receive an optimized design that meets their needs.
Vrudhula’s new methodology uses complementary metal oxide semiconductor technology. He predicts that the resulting design will be 100 times more energy efficient than current generation technology, in part because AI calculations can be performed on the chip itself. Vrudhula’s design ideas have been awarded 18 patents, and the new chip will be used in devices on 5G and 6G cellular networks, such as phones and tablets.
As Vrudhula and his team complete the initial design process, they are preparing to build a number of prototypes while exploring manufacturing options with the Southwest Advanced Prototyping Hub (SWAP Hub) and industry partners. Masu.
Software designs chips, computer scientists design software
When you build a new semiconductor manufacturing facility, you need trained people to work there. Therefore, one of the key goals of the FuSe2 program is to address the severe labor shortage in the semiconductor industry.
To fill these gaps, Vrdula knew where to look.
“We’re going to work hard to get computer science students interested in hardware design,” he says.
The School of Computing and Augmented Intelligence has one of the largest computer science programs in the country. The school enrolled more than 12,000 students this fall and broke graduation records at the Fulton School’s spring convocation ceremony, awarding more than 1,800 degrees.
Vrudhula is creating a new course as part of a grant that provides a link between computer science curriculum and understanding chip design. Students have the opportunity to participate in paid internships on campus and meetings with key figures in the semiconductor industry.
Vrudula, who is also director of the Center for Intelligent Distributed Embedded Applications and Systems, hopes to demonstrate that computer science students looking for opportunities to create AI technologies, particularly the development of deep neural networks, can have an exciting and rewarding career. I’m thinking. in microelectronics.
“Salma Vrdulla is not only designing great microelectronics, she’s teaching her students to do the same,” said Ros Maciejewski, director of the Department of Computing and Augmented Intelligence. spoke. “This CHIPS-funded project honors FuSe’s mission to create new technologies and train the next generation of engineers.”
