POTSDAM — A research team at Clarkson University is developing a new tool to combat the “intense heat” that semiconductor chips generate while operating smartphones, gaming consoles, and artificial intelligence (AI). “When temperatures rise too high, performance suffers, energy is wasted, and hardware can fail,” according to Clarkson’s Aug. 22 announcement. Thermal Analysis of Semiconductor […]
POTSDAM — A research team at Clarkson University is developing a new tool to combat the “intense heat” that semiconductor chips generate while operating smartphones, gaming consoles, and artificial intelligence (AI).
“When temperatures rise too high, performance suffers, energy is wasted, and hardware can fail,” according to Clarkson’s Aug. 22 announcement.
Thermal Analysis of Semiconductor Chips, known as TASChips, is an open-source, high-performance simulation tool for thermal analysis of modern microprocessors. The software can quickly pinpoint heat buildup inside a chip, enabling engineers to design systems that “run faster, last longer, and consume less energy,” the school said.
TASChips operates in real time and can simulate even the largest and most complex processors — including those used in data centers and advanced AI systems. The software will be released as open source on GitHub, complete with documentation and case studies for researchers, educators, and students.
Yu Liu, an associate professor, and professor Ming-Cheng Cheng of Clarkson’s department of electrical and computer engineering are leading the project. Their effort is funded by a three-year, $597,316 grant from the National Science Foundation, an independent federal agency that supports science and engineering in all 50 states and U.S. territories.
The project also includes strong educational and outreach components, including integration with STEM (science, technology, engineering, mathematics) education programs. The components also include support for undergraduate research and collaborations with experts in AI, real-time scheduling, power systems, and sustainable computing from Syracuse University, Clemson University, University of Minnesota Morris, and CISTER Research Centre, ISEP in Portugal.
These efforts aim to broaden participation in computing and provide a foundation for scalable thermal-management solutions in AI and HPC (high performance computing) environments, Clarkson said.
