Workshop Series: Parallel Computing and GPU Computing with MATLAB on the Lichtenberg II Cluster
13.02. + 27.02.: Speed up your MATLAB programs
2024/02/06
The workshop series will take place on 13 th and probably 27 th February 2024. This workshop is for researchers at TU Darmstadt who would like to use parallel and GPU (Graphics Processing Unit) computing in MATLAB, and to scale their computations to the Lichtenberg II Cluster.
Update
Due to the HPC cluster's current unavailability , the second date of this workshop has to be shifted to its third date, and for the latter, a new date has yet to be announced.
+++
This workshop series is organized by Dr. Robert Feldmann (SAM) and MathWorks in close cooperation with the TU Darmstadt HRZ/HPC group. The workshops in this series will introduce you to parallel and high performance computing with MATLAB on the Lichtenberg II Cluster.
All workshops are free of charge and will be held online and in English. Registration is required (see below).
Workshop content:
February 13th 2024, 14:00 – 17:00: Parallel Computing with MATLAB on the desktop
This hands-on session will focus on speeding up MATLAB on your local machine with Parallel Computing Toolbox. You will learn about simple and advanced parallel programming constructs in MATLAB and review best coding practices for stable and fast code.
February 20th 2024, 14:00 – 17:00: Scaling Parallel Computing with MATLAB to the Lichtenberg II Cluster
This hands-on session will focus on running MATLAB parallel code on the Lichtenberg II Cluster. You will be granted access to the Lichtenberg II system and learn how to start and manage MATLAB jobs on the cluster efficiently. Special attention is given to remote submission, i.e. running a local MATLAB client on your desktop PC, and conveniently submitting batch jobs to the cluster without much manual or command line work.
February 27th 2024, 14:00 – 17:00: Training Deep Learning Networks on the Lichtenberg II Cluster
This workshop will focus on how you can use the GPUs available on the Lichtenberg II Cluster for training of deep neuronal networks. We will discuss best practices to speed up deep learning training and you will learn where and how to store your data for best performance.
Please note:
When you register for “Scaling Parallel Computing with MATLAB to the Lichtenberg II Cluster”, it is recommended that you also attend “Parallel Computing with MATLAB on the desktop”.
When you register for “Training Deep Learning Networks on the Lichtenberg II Cluster”, it is mandatory that you also attend “Scaling Parallel Computing with MATLAB to the Lichtenberg II Cluster”.
Would you like to optimise your MATLAB performance with HPC in the future?