Advanced Computing and Scientific Discovery

Advanced computing research supporting scientific discovery, spanning high-performance computing systems, AI and large-scale data analysis, scientific visualization, immersive computing, and emerging computing architectures.

Research at the intersection of high-performance computing, artificial intelligence, scientific visualization, and large-scale data analysis. This work explores both how advanced computing systems can be operated more effectively and how their capabilities can be applied to increasingly complex scientific problems. Research spans the transition from petascale to exascale computing, emerging heterogeneous HPC and AI architectures, and the growing challenges associated with managing, moving, analyzing, and understanding scientific data at scale.

Using leadership-class computing resources, including systems at the Argonne Leadership Computing Facility (ALCF), this effort investigates workload behavior, scheduling, resource management, performance, energy efficiency, data movement, and communication across large-scale computing environments. In parallel, it develops scalable AI and data-analysis workflows for scientific applications and new approaches for making large scientific datasets accessible through interactive visualization, immersive environments, and web-based analysis tools. Together, these activities connect research in computing systems with the scientific applications and user environments that ultimately drive the use of advanced computing facilities.

Research activities include:

  • HPC systems and resource management: Characterization of workloads and user behavior on leadership-class systems, along with development and evaluation of scheduling, performance-modeling, energy-aware, data-movement, communication, and resource-management techniques for heterogeneous HPC and AI platforms.

  • AI and large-scale scientific data analysis: Development of scalable machine-learning workflows for scientific applications, including biomedical imaging, connectomics, and large-scale image segmentation, using multi-GPU and leadership-class computing resources to process datasets ranging from terabytes to petabytes.

  • Scientific visualization and interactive data exploration: Development of techniques for visualizing and interacting with large simulation and experimental datasets, including browser-based visualization, virtual and extended reality, and interactive environments for exploring complex scientific data.

  • Integrated environments for scientific discovery: Exploration of emerging architectures, data portals, and software environments that bring together HPC, AI, visualization, and data-intensive workflows while making advanced computing and large-scale scientific datasets more accessible to researchers.

The Team:

  • Students
    • Elizabeth Cardoso Undergraduate Research Assistant
    • Michael Cortez Undergraduate Research Assistant
    • DongJune Park Undergraduate Research Assistant
    • Om Patel Undergraduate Research Assistant
    • Noah Pyrzanowski Undergraduate Research Assistant
    • Idunnuoluwa Adeniji [PhD Student]
    • Niccolò Brembilla [PhD Student]
    • Amy Byrnes [PhD Student]
    • Chris Grams [PhD Student]
    • Jyotsna Rajaraman [PhD Student]
  • Faculty/Researchers

This research was supported in part by the Argonne Leadership Computing Facility, which is a U.S. Department of Energy Office of Science User Facility operated under contract DE-AC02-06CH11357.