Research Computing with Octave at SHARCNET

Description: High level interpreted languages like Octave (and Matlab) are valuable in diverse research settings from engineering, to economics, social sciences and humanities because of their diverse function packages (from instrument control to image processing) and they allow for effective interaction with data and fast prototyping of computational procedures. In the context of large scale computations on High Performance Computing (HPC) resources these interpreted languages lack in the low level control that affords the performance that is possible with a compiled language. This course explores the various ways that Octave can be used at SHARCNET, from data exploration on a visualization systems, to analysis prototyping on a development node, to serial farming Matlab scripts via the systems' schedulers. Finally this course covers strategies for profiling Octave execution at SHARCNET and explores options for increasing the computational performance in the HPC environment.

Instructor: James Desjardins, SHARCNET, Brock University.

Prerequisites: Although no previous experience with Octave is necessary to take part in this course, some previous experience with research computing is helpful. That being said, any attendee familiar with the Software Carpentry Octave/Matlab course material will be able to follow all topics in this course. Attendees will require their own laptops with Octave 4.0.0 (or later) installed

Course materials: