Prof. John Sartori Wins NSF CAREER Award for “Application-specific Power Management”

 

Prof. John Sartori has been awarded the CAREER award from the National Science Foundation’s Faculty Early Career Development (CAREER) Program. This is one of the most prestigious awards instituted by the NSF to recognize and support faculty early in their careers who show the potential to “serve as academic role models in research and education and to lead advances in the mission of their department or organization.” The CAREER award ranges from $400,000 to $500,000 (depending on the research area) and is disbursed over a 5-year period.

John’s research will focus on the development of novel techniques for application-specific power management that will reduce the power consumed by general purpose processors (GPPs) without reducing performance. Low-power GPPs are used extensively in several key current and emerging applications such as smart sensors, health monitors, wearable electronics, and the internet of things. The energy efficiency of these systems determines essential characteristics such as size, weight, cost, reliability, and lifespan. Prof. Sartori’s research has the potential to significantly reduce power and energy requirements for emerging low-power systems, which could result in devices that weigh and cost less, but have greater reliability and a longer lifespan. His research also provides a non-intrusive way to improve the energy efficiency of existing systems without re-designing system hardware or software.

Prof. Sartori will use the NSF award to continue to integrate his research goals with his educational activities, and provide research experience for undergraduate and graduate students. Educational activities will center around project-based learning experiences that will allow students to directly engage and contribute to knowledge creation in the project, while also learning key engineering and research skills. Sartori has consistently refreshed courses that he leads with substantial project-based learning components that are essential to helping students not only learn critical engineering skills and concepts, but also put them into practice. He has included a project symposium or design show in each class that he teaches that afford students a platform to deploy what they have learned, and also develop soft skills such as presenting their work to a public audience. EE 1301 is an example of this endeavor. Required of all students aspiring to enter the electrical and computer engineering majors, John helped to redesign the course, and it now introduces students to the emerging field of internet-of-things (IoT) devices that are quickly becoming the most abundant type of computer part manufactured and deployed today. Under the new course design, students create IoT devices, under the broad requirements that these devices connect to the Internet, have the ability to sense the environment, and change the environment as appropriate. The course culminates in a design show which provides a means of community outreach to engage K-12 students as well as other members of the university and the community at large. The NSF CAREER award will also allow Prof. Sartori to engage in global outreach by introducing education on IoT technology to a Kenyan  educational initiative on best practices in farming.

John earned his doctoral degree in 2012 from the University of Illinois at Urbana-Champaign, and he joined the University of Minnesota soon after. He is the recipient of several best paper awards, and the author of three invited conference papers, besides authoring over 30 regular conference papers. His research contributions include the creation of automated techniques for approximate hardware and error-resilient software design, study of peak power management for many-core processors, exploration of CAD and architecture methodologies for energy-efficient multi-modal processing, and the feasibility and benefits of computing with programmable stochastic processors. His leading research interests include extreme energy efficiency, stochastic computing, and exploiting parallelism and scalability.