报 告 人：夏应龙教授
报告题目：Graphs in Big Data— Analytics, Storage, and its Application in Education
报告时间：2015年1月18日（周日）上午9:00
报告地点：静远楼1508学术报告厅
主办单位：太阳成集团、科技处
摘 要： Many Big Data analytics essentially explore the relationship among interconnected entities, which are naturally represented as graphs with properties. In this talk, we introduce several real-world solutions, including those in education, where Big Graph has been applied or can be potentially utilized. Due to the irregular data access patterns in graph computations, it remains a fundamental challenge to deliver highly efficient solutions for large scale graph analytics. Such inefficiency significantly restricts the utilization of many graph algorithms in Big Data scenarios. To address the performance issue in large scale graph analytics, we develop a whole-spectrum graph processing system from scratch, called IBM System G, which explores efficient graph data organization techniques on commodity parallel computing architectures. We discuss various aspects of the IBM System G runtime and storage, and explain its data locality during graph traversals, which leads to improved performance in big graph analytics.
夏应龙教授个人简介：
Yinglong Xia is currently a research staff member in the IBM T.J. Watson Research Center. He is a technical leader of a cross-department team on high performance graph runtime and storage for big data analytics, which contribute to the IBM System G and . He also serves as a director on the board of the Linked Data Benchmark Council (LDBC). He received his PhD in Computer Science from the University of Southern California (USC) in 2010. Prior to that, he received his MS from Tsinghua University and BS from University of Electronic Science and Technology of China (UESTC) in 2006 and 2003, respectively. He publishes extensively including 2 book chapters and 40+ papers in refereed journals and conferences/workshops. He is active in professional activities, currently chairing the technical program committee of IEEE CBD’15, the publicity of IEEE IPDPS’15 and HiPC’15, the industry program of IEEE ICME’14, etc. He also serves as track chairs, TPC members, or guest editors in numerous conferences/workshops or journals. He received the IBM Research Division Eminence & Excellence award in 2013, and was a NSF/CRA Computing Innovative Fellow in 2010-2012.