Research on Big data analytics, Cloud Computing, Parallel I/O and storage systems.
(1) Develop an efficient and fair cross-layer query scheduling framework on Hadoop/Hive.
(2) Exploit analytics shipping with virtualized MapReduce on HPC backend storage servers.
(3) Leverage parallel I/O interfaces for optimizing I/O performance of NASA GEOS-5 application.
(4) Build up Phase Change Memory based hybrid storage system for fast I/O, and address PCM's endurance issue through a novel Space Filling Curve based wear leveling technique.