主讲人简介：

Feifei Li is currently an associate professor at  the School of Computing, University of Utah. He obtained his Bachelor's degree  from Nanyang Technological University (transferred from Tsinghua University) in  2001 and PhD from Boston University in 2007. His research focuses on improving  the scalability, efficiency, and effectiveness of database and big data systems.  He also works on data security problems in these systems.  He was a recipient  for a NSF career award in 2011, two HP IRP awards in 2011 and 2012 respectively,  a Google App Engine award in 2013, the IEEE ICDE best paper award in 2004, the  IEEE ICDE 10+ Years Most Influential Paper Award in 2014, a Google Faculty award  in 2015, SIGMOD Best Demonstration Award in SIGMOD 2015, the SIGMOD 2016 Best  Paper Award, and the SIGMOD Research Highlight Award in 2017. He is/was the demo  PC co-chair for SIMGOD 2018, a member of the SIGMOD Jim Gray Dissertation Award  selection committee in 2017, a PC area chair for SIGMOD 2015 and ICDE 2014, the  demo PC co-chair for VLDB 2014, and the general co-chair for SIGMOD 2014.  He  currently serves as an associate editor for ACM TODS and IEEE TKDE.

报告摘要：

Large spatial and spatio-temporal data are  ubiquitous (e.g., sensor readings, mobile app data). Supporting interactive  queries and analytics over such data is a critical requirement in many  data-driven applications. We will present the Simba system that offers scalable  and efficient in-memory spatial query processing and analytics for big spatial  and spatio-temporal data. Simba extends the Spark SQL engine to support rich  spatial queries and analytics through both SQL and the Data Frame API (e.g.,  spatial join, knn join, trajectories), with an effective query optimizer  leveraging its indexing support and novel spatial-aware query optimizations.  Furthermore, Simba has incorporated our latest results from online aggregation  and analysis, and is able to provide online analytics that explores the  accuracy-efficiency tradeoff.  Lastly, we will also present ongoing extensions  to Simba that explores spatio-temporal learning and sentiment analysis over  large data.