Octave is an open-source clone of MATLAB, and is a high-level language for numerical simulations in a wide range of research areas.
Do you frequently solve numerical problems so large and complex that they take hours or even days on a single workstation?
Learn how you can prototype even larger modelling problems in parallel or using distributed memory with Octave in a High Performance Computing environment. Without the cost restriction of MATLAB licenses, it is possible to scale some problems up to hundreds of processors in some circumstances.
Despite slower execution times, interpreted languages such as Octave, Matlab and Python greatly benefit from optimization tips, parallelisation or distributed memory implementations. Their key advantages are a fast parallel implementation with reduced code development time, access to a variety of built-in functions and a much simpler syntax than lower level languages such as Fortan, C, C++ and MPI.
This workshop intends to provide hands-on experience and an overview on using Octave in an HPC environment, to introduce techniques to optimize and speed-up Octave with mex files. It will then introduce the open source pMaltab and MatlabMPI libraries developed by MIT for distributed memory and parallel implementations with a diverse range of example problems. Participants will also be able to submit simulations to the New Zealand eScience Infrastructure BlueFern HPC systems.
Who should attend?
All researchers and students familiar with MATLAB/Octave.