Congratulations to SRP participant, Dr. Hang Liu, his team of students, and Berkeley Lab scientists Dr. Xiaoye Sherry Li, and Dr. Aydin Buluc for being presented with the 2019 DARPA/MIT/Amazon Graph Challenge Award. The MIT/Amazon/IEEE Graph Challenge is an annual competition that provides a venue for stimulating research and highlighting innovations on graph analytic performance. Participants are encouraged to look for new solutions for measuring and comparing the capabilities of graph processing systems.
“It was super exciting to participate in such an extremely popular and competitive challenge!”, Dr. Hang Liu, his students, Santosh Pandey, Anil Gaihre, and Zhenlin Wu, and Dr. Xiaoye Sherry Li and Dr. Aydin Buluc from Berkeley Lab. shared their excitement with us.
Dr. Liu’s team entry, H-INDEX: Hash-Indexing for Parallel Triangle Counting on GPU’s focused on using Graphics Processing Units (GPUs) to accelerate triangle counting to measure better the performance and energy efficiency of some of the most powerful supercomputers in the world. Dr. Liu explains how the team gained the experience necessary to win the Graph Challenge through the training and support they received from their SRP mentors at Berkeley’s Lab during the summer. “When we were selected to the SRP fellowship, we acquired tremendous experience in graph computing, particularly in triangle counting, which is a benchmark to the Graph Challenge.”
This award is the result of the collaboration between Dr. Liu’s team and Berkeley Lab staff, a partnership established through the SRP Program. The program’s main objective is to connect scientists from underrepresented groups with scientists from Berkeley Lab to encourage research collaborations such as the successful case of Dr. Liu and his team.“I would like to thank our host and host organizations – Dr. Xiaoye Sherry Li and Dr. Aydin Buluc, as well as Berkeley Lab for their support but most importantly, I would like to thank Sustainable Horizons Institute! It is you folks that connected us with our hosts!”, exclaimed Dr. Liu.
Dr. Hang Liu is an assistant professor in the Department of Electrical and Computer Engineering at the University of Massachusetts, Lowell. His research interests include exploiting emerging hardware to build high-performance systems for graph computing, machine learning, and data compression. In the future, Dr. Liu plans to continue working with Dr. Xiaoye Sherry Li and Dr. Aydin Buluc to address other research topics such as how to use a collection of GPUs to accelerate graph learning further.