Helen and John C. Hartmann Department of Electrical and Computer Engineering
New Jersey Institute of Technology, Newark, NJ
Joint Faculty,
Computer Science and Mathematics Division
Oak Ridge National Laboratory, Oak Ridge, TN
Email: qliu at njit.edu, liuq at ornl.gov (not frequently checked)
Office: ECEC 345
✧✧
Bio
Dr. Gary Liu is an Associate Professor in the Department of Electrical and Computer Engineering at NJIT, and he holds Joint Faculty Appointment at ORNL (with Workflow Systems Group under Data and AI Systems Section).
His research interests include high-performance computing, Big Data in data-intensive science, and high-speed networking. In particular, he has done extensive research on scalable data storage and analysis solutions on emerging architectures for HPC applications. His sole research products have been adopted by more than twenty HPC applications in fusion energy, high-energy physics, cancer research, quantum physics, turbine engine design, weather modeling, and etc., for production purposes. He was awarded outstanding graduate of the ECE Department at UNM. He was the distinguished employee of Computing and Computational Science Directorate at ORNL in 2012. He received NSF CAREER award in 2022 and R&D 100 award as a principle investigator in 2013 (with Rutgers University, Georgia Tech, and NCSU) for his contributions to adaptable I/O systems for Big Data applications. He has played leadership roles in several DOE projects. Learn Moreabout Dr. Liu's experience.
Publication
Dr. Liu's areas of research include high-performance computing and Big Data science. He has published on premier HPC conferences and journals, such as SC, HPDC, SIGMETRICS, IPDPS, ICDCS, CLUSTER, TPDS, TC, TOS, and etc.
News
Received NSF CAREER award, 2022.
MGARD+: Optimizing Multilevel Methods for Error-bounded Scientific Data Reduction,” IEEE Transactions on Computers, 2021.
Enhancing Proportional IO Sharing on Containerized Big Data File Systems,” IEEE Transactions on Computers. 2020
The Exascale Framework for High Fidelity coupled Simulations (EFFIS): Enabling whole device modeling in fusion science, The International Journal of High Performance Computing Applications. May 2021.
ADIOS 2: The Adaptable Input Output System. A Framework for High-Performance Data Management, accepted by Elsevier Journal of SoftwareX.
Compression Ratio Modeling and Estimation across Error Bounds for Lossy Compression, IEEE Transactions on Parallel and Distributed Systems, 2019
Identifying Latent Reduced Models to Precondition Lossy Compression, IPDPS'19
Load-aware Elastic Data Reduction and Re-computation for Adaptive Mesh Refinement, The International Conference on Networking, Architecture, and Storage, 2019, best paper
Exploring Transfer Learning to Reduce Training Overhead of HPC Data in Machine Learning, The International Conference on Networking, Architecture, and Storage, 2019
SIRIUS: Enabling Progressive Data Exploration for Extreme-Scale Scientific Data, IEEE Transactions on Multi-scale Computing Systems, 2019
Can I/O Variability be Reduced on QoS-less HPC Storage Systems?, IEEE Transactions on Computers, 2019
High Performance I/O Frameworks 101, tutorial at SC'18
Harnessing Data Movement in Virtual Clusters for In-Situ Execution, IEEE Transactions on Parallel and Distributed Systems, 2019
Workhighlighted by Oak Ridge Leadership Computing Facility
DuoModel: Leveraging Reduced Model for Data Reduction and Re-computation on HPC Storage, IEEE Letters of Computer Society, 2018
Write Energy Reduction for PCM via Pumping Efficiency Improvement, ACM Transactions on Storage, 2018
A View from ORNL: Scientific Data Research Opportunities in the Big Data Age, accepted to ICDCS'18
Canopus+: Intent-driven Data Refactoring for Extreme-Scale Data Analytics, ICNSC'18 (abstract paper)
Understanding and Modeling Lossy Compression Schemes on HPC Scientific Data, IEEE IPDPS'18 (1st rounder, 38 out of 461 submissions, best paper nominee).
Canopus: A Paradigm Shift Towards Elastic Extreme-Scale Data Analytics on HPC Storage, IEEE Cluster'17 (acceptance rate 21%)
TGE: Machine Learning Based Task Graph Embedding for Large-scale Topology Mapping, IEEE Cluster'17 (acceptance rate 21%)
Computing Just What You Need: Online Data Analysis and Reduction at Extreme Scales, (invited paper) accepted to EuroPar'17. (acceptance rate 28%)
SELF: A High Performance and Bandwidth Efficient Approach to Exploiting Die-stacked DRAM as Part of Memory, IEEE MASCOT, Banff, Canada, September, 2017 (acceptance rate 30%)
Canopus: Enabling Extreme-Scale Data Analytics on Big HPC Storage via Progressive Refactoring, USENIX Hotstorage, Santa Clara, July, 2017
Exacution: Enhancing Scientific Data Management for Exascale, ICDCS 2017, Atlanta, GA, June 2017.
StoreRush: An Application-Level Approach to Harvesting Idle Storage in a Best Effort Environment, accepted to ICCS, Zurich, Switzerland, June, 2017. (acceptance rate 28%)
Excited to be part of DOE Co-design Center for Online Data Analysis and Reduction (led by Ian Foster) to work on exascale data reduction techniques, and be part of DOE SIRIUS project (led by Scott Klasky) to work on exascale data storage.
Invited to DOE ASCR Panel Review, 2018, IPDPS (16, 18), CCGrid (17, 18,19), ICPADS'17, SC (14, 15, 17), Cluster (15), SSDBM (17,18), ICDCS'17, and NVMSA'17, Publication co-chair of ICNSC 2018.