Matthew R. Norman
Matt Norman is a Computational Climate Scientist with the Scientific Computing Group in the National Center for Computational Sciences and the Climate Change Science Institute. Attending North Carolina State University for both undergraduate and graduate work, Matt received a BS in Computer Science, a BS in Meteorology, and a minor in Mathematics, working on projects ranging from stability analysis of semi-implicit semi-Lagrangian schemes to data mining and analysis. For the MS in Atmospheric Sciences, the thesis title was "Investigation of Higher-Order Accuracy for a Conservative Semi-Lagrangian Discretization of the Atmospheric Dynamical Equations". For the PhD in Atmospheric Sciences, the dissertation title was "Characteristics-Based Methods for Efficient Parallel Integration of the Atmospheric Dynamical Equations".
Most recently, Matt’s focus has been on striving for GPU (graphics processing unit) efficiency in the atmospheric dynamical core from the level of the algorithm choices more so than extensive memory optimization. Matt has also been working to gain efficiency with minimal coding effort by ensuring efficient use of the per-multiprocessor L1 cache. He is working with the Oak Ridge Leadership Computing Facility 3 (OLCF3) readiness for Community Earth System Model/Community Atmosphere Model/HOMME (High-Order Method Modeling Environment) along with user and project assistance in NCCS. In addition, he will continue work in dynamical core algorithm development thereby complimenting the GPU porting work.