Current Research Projects

  • SHF: Small: Revamping I/O Architectures Using Machine Learning Techniques on Big Compute Machines. 
  • NSF PPoSS Planning: Data Centric Computing for Scalable Heterogeneous Memory and Storage Systems Architecture. 

Recent and Past Research Projects

  • SHF: Small: Developing a Highly Efficient and Accurate Approximation System for Warehouse-Scale Computers with the Sub-dataset Distribution Aware Approach. 
  • DURIP:  WINNERS OF THE FY 2017 COMPETITION UNDER THE DEFENSE UNIVERSITY RESEARCH INSTRUMENTATION PROGRAM, Next-generation all flash big data parallel processing engine.
  • NSF XPS: Collaborative Research: A Scalable and Distributed System Framework for Compute-Intensive and Data-Parallel Applications. 2013-2017. PI at UCF.
  • NASA: The Global-scale Observations of the Limb and Disk (GOLD), co-PI with PI Richard Eastes, co-PIs Hassan Forooshi, Linwood Jones, etc. 2013-2016.
  • NSF CAREER: Data-Intensive HPC Analytics: A Systems Approach through Extended Interfaces, Data Restructuring and Data-centric Scheduling, sole PI. 2010-2016.
  • NSF CSR: DSA-Cloud: Data Semantics Aware Clouds for High Performance Analytics, sole PI. 2011-2014.
  • NSF: Fast Recovery Using Optimal and Near-Optimal Parallelism in Data-Intensive Computing, sole PI. 2008-2011.
  • A Lightweight, High-performance I/O Management Package for Data-intensive Computing (Department Of Energy-DOE Early Career Principal Investigator Award) sole PI. 2005/07-2011.
  • Collaborative Research: Energy conservation in storage systems using coding techniques (NSF) w/ Prof. Varman from Rice. 2006-2008.
  • NSF: Energy-efficient, high performance storage array systems, sole PI. 2005-2008.
  • An NERSC Allocation Year 2009 award, AY 2009 runs from January 13, 2009 through January 11, 2010.
             The MPP award is:       50,000 Cray XT4-equivalent MPP hours for repository m827
             The HPSS award is:         500 SRUs for repository m827
  • An NERSC Allocation Year 2008 Startup award,  AY 2008 runs from January 8, 2008 through January 12, 2009, enable our group to experiment with supercomputers such as Franklin (19,320-processor Opteron MPP/Lustre), Jacquard (712-CPU Opteron Linux cluster/GPFS) (DOE).
  • UCFS — A User-space, High Performance, Custom File System for Web Proxy Servers
  • WOLF — A Novel Reorder Write Buffer for log-structured File System
  • PROFS — Data Placement Scheduling for Log-structured File System on Multi-Zone Disks


    • A 16-node Dell Power-edge 1950 cluster, per node Dual Dual-core Intel Xeon Processor, 4 GB memory, 2 500/144 GB SATA/SAS hard drives
    • A 48-node Sun Solaris cluster
    • A data archive center for SEECS Data warehouse project, including two Dell storage servers and four PowerVault MD1000 Storage Trays (20 TB capacity)
    • Seven Dell precision 690 workstations
  • An 8-SCSI-Disk RAID System
  • A HP/Agilent 34970A System
  • An OpenSSD Platform