The first part of this workshop focuses on parallel computing with MATLAB on the desktop. The host introduces the concept of parallel computing and briefly covers optimizing code before diving into parallelism in MATLAB and common programming constructs such as parfor — including optimization and performance considerations — and parfeval. Next, scaling options are introduced with discussion about GPUs and clusters/clouds and the workflow of submitting to these resources. Throughout this part of the workshop, users will complete hands-on exercises alongside the instructor as concepts are introduced.
The second part of the workshop focuses on showing users how to submit jobs on the ARCC cluster. With MATLAB installed on their own machine, users will be guided through the process of configuring a cluster profile, setting jobs submission arguments, accessing GPU resources, and submitting both single-node and multi-node jobs that run remotely on ARCC.
This workshop is aimed at new users of parallel computing in MATLAB, and previous MATLAB experience is beneficial but not required.
PREREQUISITES:
If you have any questions with the above, please contact Research IT at ResearchIT@ucf.edu.
This event is not being recorded. Any materials provided by the presenter will be sent to all registrants by the end of the second business day after the conclusion of the event. Please email ResearchITEvents@ucf.edu to request access to the materials two business days after the completion of the event if you did not register.
Presented by Damian Pietrus, Evan Cosgrove, Ph.D.
Read MoreRegister for this event.