Online adaptive systems in both software and hardware, datacenter and warehouse-scale computer architecture, and software / hardware co-design
As we enter the next era of computing, much of the world’s computation is moving into the cloud. As we start to understand the emerging applications that reside in this domain, we are faced with rampant inefficiency. This inefficiency manifests itself as low machine utilization, missed performance opportunities, and unnecessarily high costs of construction and operation for the “warehouse scale computers” (WSCs) that underlie the cloud. As the design of WSCs is in its relative infancy, and due to wide spread societal and environmental implications, this problem space has compelled Mars' research efforts. Specifically, Mars' research interests span the areas of system software, compiler/runtime technologies, computer architecture and the interaction of all of the above as it relates to the unique emerging challenges of today and tomorrow.
Jason Mars joined the UC San Diego faculty in 2012, after receiving his Ph.D. the same year in computer science from the University of Virginia. He received his M.S. in computer science from the University of Virginia in 2008, and a B.S. degree in CS from the University of Pittsburgh in 2005. While at UVA, Mars' research was supported by the prestigious Google Ph.D. Fellowship for Compiler Technology and Ford Foundation Fellowship.