Research Interests

• Parallel and distributed file I/O and storage system design
• Data mining algorithm design and their parallelization
• Data management for large-scale scientific applications
• Computational model design for large-scale applications on parallel and distributed environment
• Parallel netCDF

Research Projects

• The Center for Ultra-scale Computing and Information Security (CUCIS)

• Major Research Projects
  • 2016 -- 2023
    Data Libraries and Services Enabling Exascale Science
    Position: co-principle investigator
    Sponsor: DOE under the ECP program of the Office of Advanced Scientific Computing Research
  • 2018 -- 2021
    PROTEUS: Machine Learning Driven Resilience for Extreme-scale Systems
    Position: co-principle investigator
    Sponsor: DOE under the Office of Advanced Scientific Computing Research
  • 2017 -- 2020
    Machine Learning/Deep Learning and I/O Research, Development, and Deployment for RAPIDS
    Position: principle investigator
    Sponsor: RAPIDS, SciDAC Institute for Computer Science and Data, DOE
  • 2018 -- 2019
    AMASE: Architect and Manage Autonomic Science Ecosystem
    Position: co-principle investigator
    Sponsor: Argonne National Laboratory
  • 2015 -- 2018
    Scalable, In-situ Data Clustering Data Analysis for Extreme Scale Scientific Computing
    Position: co-principle investigator
    Sponsor: DOE under the Office of Advanced Scientific Computing Research
  • 2014 -- 2017
    Dynamical Modeling of Dense Star Clusters with a Parallel Monte Carlo Code
    Position: co-principle investigator
    Sponsor: NASA
  • 2014 -- 2017
    Scalable Algorithms for Spatio-temporal Data Analysis
    Position: co-principle investigator
    Sponsor: NSF under the program of Computing and Communication Foundations (CCF)
  • 2013 -- 2016
    Black Holes in Dense Star Clusters
    Position: co-principle investigator
    Sponsor: NSF under the program of Computational and Data-Enabled Science and Engineering (CDS&E)

  
  
  
  

  Selected Publications:

  1. Wei-keng Liao and Rajeev Thakur. MPI-IO. Book Chapter 13 of High Performance Parallel I/O, Chapman and Hall/CRC, October 2014.
  2. Arifa Nisar, Wei-keng Liao, and Alok Choudhary. Delegation-based I/O Mechanism for High Performance Computing Systems. In the IEEE Transactions on Parallel and Distributed Sytems, vol. 23, no. 2, pages 271-79, February 2012.
  3. Wei-keng Liao, and Alok Choudhary. Dynamically Adapting File Domain Partitioning Methods for Collective I/O Based on Underlying Parallel File System Locking Protocols. In the Proceedings of International Conference for High Performance Computing, Networking, Storage and Analysis, Austin, Texas, November 2008. ---- (This paper proposes MPI-IO optimizations for Lustre and GPFS parallel file systems. An extended version of this work can be found here published in TPDS.)

  Recent Publications:

  1. Rajkumar Kettimuthu, Zhengchun Liu, Ian Foster, Pete Beckman, Alex Sim, Kesheng Wu, Wei-keng Liao, Qiao Kang, Ankit Agrawal, and Alok Choudhary. Towards Autonomic Science Infrastructure: Architecture, Limitations, and Open Issues. In the International Workshop on Autonomous Infrastructure for Science, held in conjunction with the 27th International Symposium on High-Performance Parallel and Distributed Computing, June 2018.
  2. Tatiana Martsinkevich, Balazs Gerofi, Guo-Yuan Lien, Seiya Nishizawa, Wei-keng Liao, Takemasa Miyoshi, Hirofumi Tomita, Yutaka Ishikawa, and Alok Choudhary. DTF: An I/O Arbitration Framework for Multi-Component Data processing Workflows. In the ISC High Performance Conference, June 2018.
  3. Zijiang Yang, Yuksel Yabansu, Reda Al-Bahrani, Wei-keng Liao, Alok Choudhary, Surya Kalidindi, and Ankit Agrawal. Deep Learning Approaches for Mining Structure-property Linkages in High Contrast Composites from Simulation Datasets. Computational Materials Science, 151:278-287, Springer, May 2018.
  4. Arindam Paul, Pinar Acar, Ruoqian Liu, Wei-Keng Liao, Alok Choudhary, Veera Sundararaghavan, and Ankit Agrawal. Data Sampling Schemes for Microstructure Design with Vibrational Tuning Constraints. AIAA Journal, 56(3):1239-1250, American Institute of Aeronautics and Astronautics, March 2018.
  5. Sunwoo Lee, Dipendra Jha, Ankit Agrawal, Alok Choudhary, and Wei-keng Liao. Parallelizing Deep Convolutional Neural Network Training by Exploiting the Overlapping of Computation and Communication (best paper finalist). In Proceedings of the 24th International Conference on High-Performance Computing, Data, and Analytics, December 2017.
  6. Esteban Rangel, Nicholas Frontiere, Salman Habib, Katrin Heitmann, Wei-keng Liao, Ankit Agrawal, and Alok Choudhary. Building Halo Merger Trees from the Q Continuum Simulation. In Proceedings of the 24th International Conference on High-Performance Computing, Data, and Analytics, December 2017.
  7. Reda Al-Bahrani, Margaret Danilovich, Wei-keng Liao, Alok Choudhary, and Ankit Agrawal. Analyzing Informal Caregiving Expression in Social Media. In The Workshop on Sentiment Elicitation from Natural Text for Information Retrieval and Extraction, (SENTIRE), held in conjunction with the IEEE International Conference on Data Mining, pp. 342-349, November 2017.
  8. Qiao Kang, Wei-keng Liao, Ankit Agrawal, and Alok Choudhary. A Hybrid Training Algorithm for Recurrent Neural Network Using Particle Swarm Optimization-based Preprocessing and Temporal Error Aggregation. In The Workshop on Optimization Based Techniques for Emerging Data Mining Problems, (OEDM), held in conjunction with the IEEE International Conference on Data Mining, pp. 812-817, November 2017.
  9. Reda Al-Bahrani, Margaret Danilovich, Wei-keng Liao, Alok Choudhary, and Ankit Agrawal. Towards Identifying Informal Caregivers of Alzheimer's and Dementia Patients in Social Media. In the Fifth IEEE International Conference on Healthcare Informatics (ICHI), August 2017.

  
  
  

  Full list of publications can be found in CUCIS.

  
  
  

  Software packages

  • Parallel implementations for K-means data clustering algorithm -- including MPI, OpenMP, serial versions.
  • Parallel netCDF -- a parallel I/O library for accessing netCDF files
  
  

  Contact Information

  L359 Technological Institute
  2145 Sheridan Road
  Department of Electrical Engineering and Computer Science
  Northwestern University
  Evanston, IL 60208-3118
  

  Phone: (847) 491-2916
  Fax: (847) 491-4455
  wkliao@northwestern.edu