Parallel Computing with MATLAB 1: Speeding up your MATLAB Code on local parallel hardware

Contact

Name

Dunja Gath

Contact for Organizational Matters

Telephone

workPhone
+49 241 80 24680

E-Mail

E-Mail

Key Info

Basic Information

Format:
Online
Institution:
IT Center
Participation Fee:
free

Content

Speeding up your MATLAB Code on local parallel hardware.

Live Session: Tuesday May 2nd, 2 PM – 5 PM (3 hours)

Instructor: Dr. Thomas Künzel, The MathWorks Academia Team

This workshop will focus on speeding up MATLAB on local desktop machines with Parallel Computing Toolbox. You will be introduced to parallel and GPU computing language constructs in MATLAB and we will also introduce the language mechanisms which are the prerequisite for scaling your applications to clusters and cloud.

By working through practical code examples, you will be introduced to parallel programming constructs for multi-core desktop machines, clusters, and GPUs. As we will be using MATLAB Online (MATLAB in a browser), a MATLAB installation on your desktop is not required. You will receive the log-in credentials prior to the event. Instead of the MATLAB Online instance, you may use your own installation of MATLAB if it has a multi-core CPU and access to a GPU.

 

Topics

  • Improving serial MATLAB code (best coding practices) and profiling application performance

  • MATLAB parallel computing language constructs

  • GPU computing with MATLAB

  • Options for scaling MATLAB code to clusters and clouds

 

Who is this for?

The training is for students, junior scientists and researchers who have a beginner to intermediate MATLAB knowledge. If you have a growing demand for MATLAB performance or plan to increase the scale of your data analysis or simulation codes, this course is for you!

Appointments

Number:2023-MAT-001
Date:

Tuesday, 2. May, 14:00 - 17:00

Place:

Online

Slots: up to 60
Registration:RWTH-SingleSignOn
External registration