Increase computing power | Arizona State University news

October 19, 2022

When Matthew Marinella Leaving a research role in Sandia National Laboratories After 11 years to become an associate professor of electrical engineering at Arizona State University, he didn’t leave all of his Sandia connections behind.

Since joining college Faculty of Electrical, Computer and Energy Engineeringpart of IRA A Colleges. Fulton EngineeringMarinella continued to collaborate with Sandia on his research. He’s the principal investigator on a project to increase computing performance by increasing the energy efficiency of radiation-hardened electronic devices as part of Sandia’s Big challenge Scientific Research Series. Work at Arizona State University builds on previous electronics research conducted by Marinella at Sandia.
A satellite docked at the International Space Station
Matthew Marinella, assistant professor of electrical engineering in the IRA A. Fulton Schools of Engineering at Arizona State University, and his collaborators at Sandia National Laboratories are looking at how to make radiation-enhanced electronics more energy-efficient. Semiconductor chips, along with the experimental analog computing devices integrating them, can improve “radio” computing performance by 100 times for a given onboard power source. Image courtesy of Pexels
Download the full image

Radioactive hardening is a process that increases the durability of electronics used in highly radioactive environments such as outer space. This could keep critical computer components, such as those on a spacecraft, operating where normal electronics would fail under exposure to radiation.

Sandia’s Grand Challenge projects are high-risk, high-reward and funded for a period of three years.

“I just started at Arizona State University as of January 2022, coming from Sandia where we have been working on a very large project like this for a number of years,” Marinella says. “It’s finally funded at Sandia, so of course we hope that this work will lead to bigger projects.”

Collaborators at . also participate in this Grand Challenge research University of California, BerkeleyThe The University of Texas at Austin and the University of Michigan. Rick McCormick, senior project management sponsor and strategic client for research at Sandia, says researchers from these institutions have been brought in for their expertise in emerging analog computing devices, which is an essential part of the project.

To increase the computing efficiency of radiation-enhanced electronics beyond traditional capabilities, the research team is working to develop new analog devices that use RAM Resistanceor ReRAM and Electrochemical RAMor ECRAM.

These analog devices will be combined into arrays placed on top Complementary metal oxide semiconductoror CMOS, computer chips manufactured by Taiwan Semiconductor Manufacturing Corporation.

Once these devices are developed, they will then be used in radiation-rich environments both in space and on Earth.

“If you see Chernobyl or somewhere like that, you have bots trying to go somewhere you don’t want to put people,” Marinella says.

He presented the satellite camera as another example of what increased processing efficiency can do. The capabilities of satellites are limited by parameters such as their size, weight, and battery power. Making onboard computing more efficient, as with the technology the Marinella team is developing, would save power for other tasks, including increasing the resolution of satellite images.

Marinella says the Department of Defense also has an interest in radiation-enhanced computing technology, including for uses such as image processing with edge computing. In edge computing, data is processed in a computer system soon after it has been collected, resulting in fewer large raw data files that need to be transferred and speeding up file exchange.

Although chips are currently intended for use in radiation-hardened applications, Marinella sees efficiency-enhancing memory chip arrays, along with traditional semiconductors, eventually becoming ubiquitous in consumer electronics.

“This will be the chip in your mobile phone, self-driving vehicles, and cloud computing systems,” he says.

according to Saban AgarwalPrincipal Investigator on the Sandia side of this Grand Challenge project, integrating memory devices with conventional chips will push the boundaries far beyond what is possible with conventional chips alone. The processing efficiency of conventional CMOS chips is limited by the size and voltage of their transistors.

Agarwal says that one of the main factors slowing computing with traditional chips is the need to move data between separate memory storage and processing elements. Incorporating the team’s new ReRAM and ECRAM hardware with CMOS chips will enable processing and memory storage in one location.

According to Agarwal, this will result in 100 times greater computing performance per watt than is currently possible.

McCormick notes that the team will test different analog computing devices, which will help them see which works best for their applications.

Marinella is also working on a separate project with Hugh BarnabyProfessor of Electrical Engineering at Arizona State University, to understand the effects of radiation on transistors known as fin field-effect transistorsOr FinFETs, which are more efficient than older standard node transistors. While this project is also a collaboration with and funded by Sandia, it differs from Marinella’s work on the Grand Challenge program, which focuses on integrating CMOS chips with analog hardware arrays.

Although the Grand Challenge project that combines CMOS chips and analog devices will not use FinFETs in research, Marinella’s future goal is to use FinFETs as transistors for CMOS chips and integrated analog arrays if FinFETs prove they can tolerate radiation well enough in environments. A more efficient combination will release more radiation-hardening computing power, making the system more efficient even than the current Grand Challenge CMOS project and the Analog Memory Integration project it’s studying.

McCormick says that with Marinella working at Arizona State University and leaving Sandia with a track record of innovation and training postdoctoral researchers, he has put Sandia in a great position to collaborate with the university for research opportunities.

“Matt has already connected us with other high-impact professors at ASU and has maintained strong collaborative relationships with Sandia,” he says. “We look forward to continuing to work with him and having him train the next generation of Sandia researchers.”

Agarwal agrees that the partnership with Arizona State University is a great opportunity to expand Sandia’s research capabilities.

“The team at Arizona State University is an essential part of our broader research effort, helping to develop key enabling technologies and partnering with us on a broader emerging microelectronics strategy,” he says. “At Arizona State University, Matt is still an important collaborator.”

Leave a Comment