Justin Wozniak joined the Mathematics and Computer Science Division at Argonne National Laboratory in spring 2008 and the staff of the Computation Institute at the University of Chicago in Winter 2009.  He became a Fellow at the University of Chicago in 2014, and transferred to the new Data Science and Learning Division at ANL in 2018.

He has designed and implemented novel workflows systems for scientific applications and deep learning workloads, combining techniques from high-performance computing, grid computing, and Internet computing. To gain insight into next-generation computing systems, he has developed simulators to study the effects of policy changes in large-scale system software such as schedulers and storage management algorithms. Additionally, he collaborates with the computing community to support the rapid development of scalable applications and provide portable, efficient access to the largest computing installations.

Research Interests

• High-level programming models for concurrency at large scale
• Distributed computing, fault tolerance, and recovery
• Workflow systems for simulation, experiment, and learning
• Simulation of computer systems
• Control theoretic applications in computing systems

Education

• Ph.D., Computer Science and Engineering, University of Notre Dame, 2008.
• MMath, University of Waterloo, ON, 2003.
• B.S., Mathematics and Computer Science, University of Illinois at Urbana Champaign, 2000,
  minors in Chemistry and Latin.

Awards

• Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research Finalist, 2020
• R&D 100 Winner, Swift/T, 2018.
• Best poster, University of Chicago MindBytes Posters 2017.
• Best paper, Cancer Workshop at SC 2016.
• Best paper, ScienceCloud 2014.
• Arthur J. Schmitt Leadership Fellowship, 2004-2008.

Activities

• Co-Chair, XLOOP: Workshop on Large-scale Experiment-in-the-Loop Computing @ SC
• Co-Organizer, WoWoHa: DOE Workflow Workshop and Hackathon

Key Publications

• High-bypass learning: Automated detection of tumor cells that significantly impact drug response
  Justin M. Wozniak, Hyunseung Yoo, Jamaludin Mohd-Yusof, Bogdan Nicolae, Nicholson Collier, Jonathan Ozik, Thomas Brettin, and Rick Stevens
  Proc. Machine Learning in HPC Environments @ SC 2020.
• MPI jobs within MPI jobs: A practical way of enabling task-level fault-tolerance in HPC workflows
  Justin M. Wozniak, Matthieu Dorier, Robert Ross, Tong Shu, Tahsin Kurc, Li Tang, Norbert Podhorszki, and Matthew Wolf.
  Future Generation Computing Systems 101, 2019.
• CANDLE/Supervisor: A workflow framework for machine learning applied to cancer research
  Justin M. Wozniak, Rajeev Jain, Prasanna Balaprakash, Jonathan Ozik, Nicholson Collier, John Bauer, Fangfang Xia, Thomas Brettin, Rick Stevens, Jamaludin Mohd-Yusof, Cristina Garcia Cardona, Brian Van Essen, and Matthew Baughman. BMC Bioinformatics, 2018.
• The relation of local order to material properties in relaxor ferroelectrics
  Matthew J. Krogstad, Peter M. Gehring, Stephan Rosenkranz, Raymond Osborn, Feng Ye, Yaohua Liu, Jacob P. C. Ruff, W. Chen, Justin M. Wozniak, Haosu Luo, Omar Chmaissem, Zuo-Guang Ye, and Daniel Phelan.  Nature Materials 17, 2018.
• Turbine: A distributed-memory dataflow engine for high performance many-task applications
  Justin M. Wozniak, Timothy G. Armstrong, Ketan Maheshwari, Ewing L. Lusk, Daniel S. Katz, Michael Wilde, and Ian T. Foster.  Fundamenta Informaticae 28(3), 2013.

A more complete list is maintained here: https://www.mcs.anl.gov/~wozniak/papers.html