Mars’ research in compiler and runtime technology addresses the problem of contention in multicore processors. The contention problem leads to slower performance when computers are running multiple applications at the same time and processors are competing for the computer’s shared memory resources, such as its on-chip cache and Random Access Memory, or RAM. Ultimately, Mars’ research aims to continue Moore’s Law, which states that computers will double processing speeds about every two years, and also create more energy-efficient Internet data storage and computing centers.

Without improvements in energy efficiency, servers and data centers in the United States are expected to consume more than 100 billion kilowatt-hours, at a cost of $7.4 billion annually, by 2011, according to a 2007 report by the Environmental Protection Agency.

“It’s really satisfying to have the opportunity to attack some of today’s most challenging problems in computing,” Mars said. “To address some of these pressing problems, we must enable both software and its underlying hardware to adapt dynamically. I strive to do truly great work and look forward to the problems on the horizon.”

This most recent fellowship adds to a list of several other Google awards for research projects on which Mars has been a team member. In April 2009, he and his adviser at U.Va. secured an $80,000 Google Research grant to study how software can adapt to hardware to improve performance and efficiency. Earlier this year, he co-authored a research proposal that won a $55,000 Google Research Grant to study the effects of contention in datacenters and develop compiler technology to address this challenge.

Prior to these recent accomplishments, in January 2007 Mars was also awarded a $5,000 Google Scholarship through the United Negro College Fund.

“Jason is a rising star in computer science,” said Mars’ adviser, Mary Lou Soffa, chair of the U.Va. Department of Computer Science. “His work as a graduate student is helping our department advance research that will allow for continued improvements in computer performance efficiency, while at the same time reducing their energy use. We are very proud of Jason and his accomplishment at this early stage of his career. We look forward to his future contributions to the field.”

In addition to Mars’ fellowship for compiler technology, Google awarded fellowships to 30 Ph.D. students in Canada, China, Europe, Israel and the U.S in the fields of machine learning, statistics, computer networking, web application security, machine translation, distributed systems, human computer interaction, language security, neural networks, computer graphics, computer vision, natural language processing, search and information retrieval and speech.