USQ Logo
The current and official versions of the course specifications are available on the web at https://www.usq.edu.au/course/specification/current.
Please consult the web for updates that may occur during the year.

MAT2409 High Performance Numerical Computing

Units : 1
Faculty or Section : Faculty of Health, Engineering and Sciences
School or Department : School of Mathematics, Physics & Computing
Student contribution band : Band 1
Grading basis : Graded
Version produced : 23 May 2022

Requisites

Pre-requisite: (CSC2410 or CSC1401) and (MAT1102 or ENM1600) or Students must be enrolled in one of the following Programs: MPIT or MCOT or MCTE

Overview

Many areas of computing in engineering, science, technology and games require programmers to have insight and skills in the implementation of common numerical computations. Programming high-performance computers to rapidly perform large scale tasks requires considerable skill. Modern vector and super-scalar computers are very fast - but to achieve anything remotely like the peak speed requires special programming styles sympathetic to the computer architecture. Using fundamental algorithmic tasks of science, this course develops the ability to design good algorithms for modern computer architectures.

This course develops skills in programming modern high-performance computers. It examines some of the typical hardware architectures and how they affect performance and programming. Algorithms to illustrate the principles are chosen from a range of scientific tasks. The course includes the study of numerical solutions of linear and non-linear equations, numerical interpolation and curve fitting, the numerical solution of ordinary differential equations, and Monte Carlo simulation. Interaction utilising modern graphics is exploited.

Course offers

Semester Mode Campus
Semester 1, 2022 On-campus Toowoomba
Semester 1, 2022 Online
Date printed 23 May 2022