I'm a senior software enginer in Couchbase. I received my Ph.D. in Computer Science at University of California, Irvine, advised by Prof. Michael J. Carey. Before that, I got a B.Sc. from Nanjing University, China, an M.Phil. from The Chinese University of Hong Kong, and also fulltimely worked in Microsoft SQL Server group .
- 2016/04/05 Our tutorial "Big Graph Analytics Systems" has been accepted to SIGMOD'16!
- 2016/02/05 Our paper about scaling Google Fusion Tables has been accepted to ICDE'16!
- 2015/08/20 I defended my thesis and joined Couchbase. (a related news:-) )
- 2015/06/22 After 5 years's R&D of Algebricks, our paper about the framework has been accepted to SOCC'15!
- 2015/03/01 AsterixDB is accepted to be an Apache incubator project!
- 2014/11/11 Our paper about the Facade compiler and runtime for Big Data applciations has been accepted to ASPLOS'15!
- 2014/08/05 After 3 years's R&D of Pregelix, our paper about the system has been accepted to VLDB'15!
- 2014/08/04 After 5 years's R&D of AsterixDB, our paper about the system has been accepted to VLDB'15!
- 2014/07/06 Our paper about graph connectivity analysis on Pregel has been accepted to VLDB'15!
- 2014/06/23 Start a returned internship in Google Research, working with Chris Olston and Peter Hawkins!
- 2014/05/09 New Pregelix website is online now! Checkout our juicy perf. numbers!
- 2013/08/13 AsterixDB team are visiting Couchbase for the Couchbase/AsterixDB workshop!
- 2013/06/11 I'm awarded the 2013-2015 Google Fellowship in Structured Data!
- Archived news
My primary area of research interest is in building and evaluating Big Data management systems.
- Pregelix. Pregelix is an open-source implementation of Google's Pregel programming model. We architect the Pregel programming model on top of a general-purpose data-parallel execution engine, which leads to better scaling properties, out-of-core support, physical flexibility, and software simplicity.
- AsterixDB. We are working towards an open source data-intensive computing platform, with new technologies for ingesting, storing, managing, indexing, querying, analyzing, and subscribing intensive semi-structured data.
- HaLoop. In HaLoop, we designed and implemented a modified version of the Hadoop MapReduce framework for efficiently support data-intensive iterative data analysis.
Publications (dlbp entry) (google scholar)
(Start Apache incubation in March 2015) (In the News) (Press Release)
(Open-source systems using our design paradigm: AsterixDB, Hyracks, Pregelix )
(Best of VLDB 2010 )
System Demos and Posters
Honors and Awards
- 2013-2015 Google Fellowship in Structured Data
- 2013-2014 Facebook Fellowship Finalist
- 2010 Yahoo! Key Scientific Challenage Award