MATLAB Workshop at Illinois
Overview
Parallel Computing with MATLAB
This hands-on workshop seminar will introduce you to parallel computing with MATLAB®, so that you can solve computationally and data-intensive problems using multicore processors, GPUs, and computer clusters. Examples will be presented on NCSA Delta cluster which participants can follow during the session in their own MATLAB instances. By working through common scenarios to parallelize MATLAB algorithms, you will gain an understanding of parallel computing with MATLAB and learn about best practices.
Examples will vary in difficulty from simple parallel usage concepts to more advanced techniques.
- Speeding up MATLAB applications with parallel computing
- Running multiple Simulink simulations in parallel
- GPU computing
- Working with large data sets
Scaling your MATLAB Code to NCSA Delta Cluster
During this hands-on workshop, attendees will learn how to offload their MATLAB code by submitting MATLAB jobs to NCSA Delta Cluster. At the conclusion, attendees will learn the following:
- How to configure MATLAB to submit jobs to NCSA Delta Cluster
- The job submission workflow
- Ways to tune job submissions, including accessing GPUs on NCSA Delta Cluster
- How to optimize job submissions
- Troubleshooting job submission techniques
- Best practices for rehosting code
Who Should Attend
Faculty, staff, researchers, and students are welcome! Attendees should have some MATLAB knowledge. Optional free introductory MATLAB Onramp Training here.
Additional Information for Workshop Registrants
- Please bring your laptop with MATLAB 2024b and Parallel Computing Toolbox installed. MATLAB Campus-Wide License Access and Support for everyone at UIUC here
- You must have an ACCESS User Account to participate in the hands-on training activities on Delta. If you do not have one, please visit this page to create one: ACCESS USER REGISTRATION
- Please set up your account before registering for this event and enter it into the registration form. Your ACCESS User Account (ACCESS ID) will be added to the training allocation for the hands-on sessions. You will receive a separate email with instructions on setting up your NCSA identity and NCSA Duo, which are necessary for accessing Delta. Please set these up before the training session. If you do not receive this information or encounter any difficulties, please let us know by replying to this email.
- Note: If you do not provide us with your ACCESS user account by Monday, February 10th, your training account will most likely not be available for the hands-on activities.
About the Presenter
Gen Sasaki is a Principal Customer Success Engineer at MathWorks. In this role, he partners closely with educators and students across universities, ensuring they leverage MATLAB to its fullest potential. Gen holds a Bachelor's and Master's degree in Mechanical Engineering, with a specialization in control systems. His professional journey spans nearly three decades, during which he has made significant contributions to the automotive and aerospace sectors, focusing on powertrain systems and a variety of embedded controls.
Damian Pietrus is a Parallel Computing Application Engineer with a BA from Harvard University who has been with MathWorks since 2017. His focus is on integrating MATLAB Parallel Server with various HPC and cloud environments and helping users to leverage these expanded computing resources effectively.
Agenda
| Time | Title |
| 9:00 AM - 12:00 PM CST | Parallel Computing with MATLAB |
| 12:30 PM - 3:30 PM CST | Scaling your MATLAB Code to NCSA Delta Cluster |
