Author: Mullowney, P.J.
Paper Title Page
TUPEA073 Performances of VORPAL-GPU Slab-symmetric DLW 1298
  • F. Lemery, K. Duffin, N. Karonis, D. Mihalcea, P. Piot, J. Winans
    Northern Illinois University, DeKalb, Illinois, USA
  • P.J. Mullowney, P. Stoltz
    Tech-X, Boulder, Colorado, USA
  • P. Piot
    Fermilab, Batavia, USA
  Funding: HDTRA1-10-1-0051, DOE(Grant No will be specified later)
GPU-based computing has gained popularity in recent years due to its growing software support and greater processing capabilities than its CPU counterpart.  GPU computing was recently added in the finite-difference time-domain program VORPAL. In this paper we carry electromagnetic simulations and optimization of a flat beam passing through a slab-symmetric dielectric-lined waveguide (DLW). We use this simulation model to explore the scaling of the GPU version of VORPAL on a new TOP1000-grade hybrid GPU/CPU computer cluster available at Northern Illinois University.