Workshop: Parallel Computing with MATLAB on the Lichtenberg Cluster
Overview
During this hands-on workshop, you will be introduced to parallel and GPU computing in MATLAB for speeding up your applications and offloading computations. By working through common scenarios and workflows, you will gain an understanding of the parallel constructs in MATLAB, their capabilities, and some of the issues that may arise when using them.
Agenda
This workshop will be held in two parts:
Part I will focus on speeding up MATLAB on the desktop with Parallel Computing Toolbox. You will learn about simple and advanced parallel programming constructs in MATLAB and review best coding practices for robust and fast code. After a short lecture, you will work on self-paced exercises on your machine or in the cloud.
Part II will focus on running MATLAB parallel code on the Lichtenberg HPC cluster.
Highlights
- Speeding up programs with parallel computing
- Working with large data sets
- GPU computing
- Scaling to your HPC cluster
Who Should Attend
PhD students, Post Docs and Professors at TU Darmstadt that want to (Part I) use parallel and GPU computing in MATLAB, and (Part II) scale their computations to the Lichtenberg HPC Cluster.
Requirements for Part I
- Basic working knowledge of MATLAB
- A MathWorks Account. If you do not already have a MathWorks Account, please go to https://www.mathworks.com/login to create one.
Requirements for Part II
- A Lichtenberg user account. If you do not already have a Lichtenberg user account, please request one as described on the HRZ website. You do not need to specify a Project-ID. If you need technical support, contact the HPC Help Desk: hhlr@hrz.tu-darmstadt.de. Please put your request in as soon as possible, at the latest on October 20th. You are required to send in the signed application by post. However, you can advance the process by sending a digital version of the signed application beforehand.
- Local installation of MATLAB R2020b or R2021a including MATLAB Parallel Computing Toolbox. You can download the software via the TU Darmstadt MATLAB Portal.
