Feng “George” Yu is currently an assistant professor in the Department of Computer Science and Information Systems at Youngstown State University, Youngstown, OH, where he founded a research lab, Data Lab, devoted to the research of data-oriented sciences. Yu’s major research interests include databases, big data, and cloud computing.
His primary research focuses are query processing and optimization for database engines. The central topic of this field is called the join ordering problem, which is the search for the optimal plan for a given join query. He uses statistical sampling methods to develop data structure summaries, which can provide accurate query size estimations to improve the query plan generation of database query optimizers. His methods feature higher computational efficiency and accuracy. In addition, his work can also be applied in another related field, called approximate query processing, for fast query result estimation with controlled accuracy.
Yu’s second research focus lies in NoSQL for big data. Write optimization of column-store databases, which compose a major category of NoSQL, is a well-known challenge in this field. Instead of in-memory scenarios, he focuses on out-of-core scenarios and developed a new data structure, time-stamped binary association tables, and several data cleaning methods to hasten the writing operations on column-store databases.
His research interest in cloud computing lies in a new cloud architecture called heterogeneous cloud. Different from a hybrid cloud, which combines public and private clouds, a heterogeneous cloud features the combination of both cloud computing services and big data services on the same cluster for better resource utilization and cost efficiency. More importantly, big data services are deployed directly on the bare-metal servers in the same cloud cluster with a special infrastructure design.
Besides his work at Youngstown State University, Yu also serves as a campus champion of NSF XSEDE (Extreme Science and Engineering Discovery Environment). He is active in promoting supercomputing resources in XSEDE to regional researchers to assist their research. He is also enthusiastic in cultivating the next-generation specialists of big data and data-centric computing. Collaborating with XSEDE and Pittsburgh Supercomputing Center, Yu has been continuously organizing regional students, staff, and faculty to attend national workshops on high-performance computing since 2014.
Yu received his PhD (2013) in Computer Science at Southern Illinois University, Carbondale, IL. He received his MS (2008) in Pure Mathematics at Shandong University and BS (2005) in Information and Computation Science at Northeastern University, both in P.R. China. His research work led to a number of publications in leading international conferences including SIGMOD, EDBT, DEXA, and IEEE Big Data Congress.
For more information of Yu’s lab and research work, please visit his homepage.