Xian-He Sun, chair and professor of the Department of Computer Science, was an invited keynote speaker at two conferences recently. The first conference, the Workshop on Large-Scale Parallel Processing held at the Institute of Electrical and Electronics Engineers (IEEE) International Parallel and Distributed Processing Symposium in Shanghai, China, took place May 21–25. He was one of two keynote speakers at this workshop and his seminar topic was “Pursuing Large-Scale Systems from the Data-Centric Point-of-View.” The next conference was the 7th IEEE International Conference on Networking Architecture and Storage, which was held from June 28–30 in Xiamen, China. Here he gave the talk, “Memory System for Extreme-Scale Computing.”
Additionally over the summer, Sun and his team were awarded a three-year collaborative grant from the National Science Foundation for “Decoupled Execution Paradigm for Data-Intensive High-End Computing.” The research is on designing a high-performance computing paradigm to handle and manage large scale of data. It has three technical objectives: understanding the execution paradigm requirement from the data-centric point of view, studying the feasibility of the proposed decoupled execution paradigm, and providing a partially implemented prototyping of DEP and its associated system design to support the first two objectives.