• screen 1

    New computing
    research
    infrastructure

    New computing research infrastructure will accelerate existing research and enable ground-breaking new research that shares the common theme of massive parallel computing at USF.

  • screen 2

    Energy-efficient
    and secure computations

    Energy efficiency and security are two key driving forces for future systems dominated by mobile devices and Internet-of-Things (IoT).

  • screen 3

    Build pattern
    theory based
    representations

Project Summary

This infrastructure is a computer cluster with about 120 manycore co-processors including GPUs and Intel Phi cards. This acquisition will provide the much needed parallel computing capacity that is currently lacking in the existing USF IT infrastructure.

more info

 Our Team

This infrastructure will bring together eight investigators with research projects in several core disciplines of computer science and engineering: big data management, system security, hardware design, data mining, computer vision and pattern recognition.

more info

Project Objective

This infrastructure will enable or enhance research in design, implementation, and evaluation of parallel programs at various levels of the software stack, ranging from cluster-level systems to individual application components and domain-specific languages.

more info

USF Parallel Computing Workgroup

The USF Parallel Computing Workgroup coordinate related research, education, and outreach activities. This ensures the project team act in an integrated way to maximize the impact of the new infrastructure on the entire USF community.

Intellectual Merits:
The infrastructure will bring together researchers with projects that are currently separate but share the same needs of an infrastructure with massive computing capacity. The first source of intellectual merits is those from the individual projects. They include:

  • design and optimization of a novel data stream management system architecture
  • coarse grained molecular simulation approach
  • efficient modeling and design of new hardware systems
  • new system to deploy security policies
  • automated interpretation of activities using pattern theory;
  • fast large scale clustering
  • pattern identification from radiomics data