Overview:

Graph processing is experiencing a renewed surge of interest as applications have emerged or grown in importance in social networks analysis, recognition and the sciences. Graph algorithms are notoriously difficult to execute efficiently, so there has been a considerable effort in improving the performance of processing large graphs for these applications.

As an architect specializing for a particular workload, I will describe my vertically integrated research on improving graph algorithm performance. This ranges from algorithms with a new approach to breadth-first search to architecture with a detailed graph workload characterization. In between, this includes a graph domain-specific language, a performance model, and a benchmark suite. I will conclude the talk with my intended future directions of making architecture more communication-centric.

Bio:

Scott Beamer is a Computer Architecture Ph.D. candidate at the University of Calf., Berkeley advised by Krste Asanović and David Patterson. He is currently investigating how to accelerate graph algorithms through software optimization and hardware specialization. In the past, he looked into how to best use monolithically integrated silicon photonics to create memory interconnects. He received his bachelor's of science in Electrical Engineering and Computer Science and a master's of science in Computer Science, both from UC Berkeley.