Module 2 - Parallel Computing with MATLAB
on Thursday, 04/11/2024, 10:00 am - 5:00 pm
Speaker: Mihaela Jarema, MathWorks, Raymond Norris, Principal Application Engineer, MathWorks
In this tutorial, you will be introduced to parallel and distributed computing in MATLAB for
speeding up your application and offloading work. By working through common scenarios and workflows, you will gain an understanding of the parallel constructs in MATLAB, their capabilities, and some of the typical issues that arise when using them.
This module is divided into two sessions:
Tutorial Part I - Morning Session:
Parallel Computing with MATLAB @ TU Dresden
10 am - 1 pm CET
by Mihaela Jarema, MathWorks
Tutorial Part II - Afternoon Session:
Scaling Parallel Computing with MATLAB to HPC @ TU Dresden
2 pm - 5 pm CET
by Raymond Norris, Principal Application Engineer, MathWorks
Overall Agenda for Module 2
-
During this hands-on tutorial, we will show users how to best submit MATLAB jobs to an HPC cluster. Users will learn:
-
How to configure MATLAB to submit remote jobs to the HPC cluster
-
The job submission workflow
-
Ways to tune job submissions to the HPC cluster
-
How to optimize job submissions
-
Troubleshooting job submission techniques
-
Best practices for rehosting code onto the HPC cluster
-
Speeding up programs with parallel computing
-
GPU computing
-
Working with large datasets
-
You can find a Getting Started Guide with MATLAB here.
Prerequisites
- Basic MATLAB knowledge
Who is this for
This introductory tutorial is designed for PhD students, researchers, and early
career scientists who want to speed up their code on their compute cores and GPUs with minimal changes to the original code.
Learning Objectives
Participants will be able to speed up their calculations by optimizing their code and applying various levels of parallelization in their applications.
Registration
Link: https://event.zih.tu-dresden.de/nhr/matlab-2
Registration is closing on 04/10/2024. The NHR tutorial is limited to 60 participants.
You will receive the access data shortly before the event by email to your registered email address.