Managing Defects in HPC Software DevelopmentSeries: HPC Best Practices Webinars
Software Quality Engineering (SQE) and methods research and scientific investigation are often thought to be incompatible. However, in reality they are not only compatible, but required in order to have confidence in the results of even basic scientific computations. This is especially true for parallel software. In this talk we will look at methods for performing software verification. Software verification is a method for removing defects at code construction time; these techniques can help in both algorithm and method development, as well as increased productivity.
- Tom Evans (Oak Ridge National Laboratory)
Thomas Evans works in the development, implementation, and application of computational radiation transport in nuclear engineering, radiation detection, astrophysics, high energy density physics, and medical applications. His interests include stochastic and deterministic transport methods on massively parallel platforms, nonlinear and time-dependent transport methods, coupled physics including radiation-hydrodynamics and core-reactor physics, acceleration and preconditioning techniques, optimization and performance analysis, and large-scale scientific software design for parallel codes. He has published over 90 refereed journal and conference articles on highly parallel stochastic and deterministic transport methods, coupled physics including radiation-hydrodynamics and core-reactor physics, acceleration and preconditioning techniques, etc. He is the primary developer of the Denovo parallel deterministic and Shift Monte Carlo transport codes at Oak Ridge National Laboratory. He is currently a PI and Focus Area Deputy in Radiation Transport for the Consortium for Advanced Simulation of Light Water Reactors Energy Innovation Hub at Oak Ridge. He is also the PI for the Coupled Monte Carlo Neutronics and Fluid Flow Simulation of Small Modular Reactors (ExaSMR) Exascale Application Project (ECP).