MCSC 6030G - High-Performance Computing

The goal of this course is to introduce students to the tools and methods of high-performance computing (HPC) using state-of-the-art technologies. The course includes an overview of high-performance scientific computing architectures (interconnection networks, processor arrays, multiprocessors, shared and distributed memory, etc.) and examples of applications that require HPC. The emphasis is on giving students practical skills needed to exploit distributed and parallel computing hardware for maximizing efficiency and performance. In a number of in-class projects, students implement numerical algorithms that can be scaled up for large systems of linear or nonlinear equations. Topics may include survey of computer architectures; efficiency guidelines for HPC; parallel algorithm design; programming tools; timing, profiling and benchmarking; and optimizations.
Credit hours: 3
Prerequisite(s): Registration in the Modelling and Computational Science program.