Strategic Applications Program

Goals and Overview

Through the Strategic Applications Program (SAP), NCSA staff collaborate with application scientists to find solutions to the problems researchers face as they use the center's cyber-resources. These collaborations involve significant two-way interactions between NCSA staff and the applications group, working together to find solutions that enable scientists to effectively utilize NCSA's resources. The results of these projects are disseminated to the broader NCSA user community in the form of new methodologies, software, and algorithms.

SAP projects have an average lifespan of between three and six months with active sets of projects evolving over time: new projects are added as earlier ones are completed. The selection procedures for taking on new projects contain many of the following attributes:

• High-quality research as evidenced by publications, grants, honors
• A clear need for powerful resources, for computation, visualization, and processing large data sets
• A well-defined and attainable set of goals for the intended time frame
• A clear potential for the expected results of the project to be useful to a wider community
• A willingness on the part of the PIs to actively collaborate with NCSA staff, with a genuine two-way interaction

NCSA Staff Commitment

The Strategic Applications Program involves NCSA staff at various levels:

• NCSA senior management has ultimate authority on the selections of researchers for the Strategic Applications Program.
• The NCSA Strategic Applications Coordinator manages proposals for projects from the user community and works with the NCSA senior management, Allocations board members, and a variety of researchers to identify potential candidates. The Coordinator works with the participating NCSA staff to set up projects, manages all aspects of the program, and reports all results to the NCSA senior management.
• NCSA staff members, who have expertise in a wide range of areas, including HPC architectures, compilers, memory hierarchy issues, message passing and shared memory paradigms, performance analysis tools, scientific libraries, and parallel debuggers; the program also draws on NCSA experts in scientific visualization, large data sets, and grid computing.