Many technical computing algorithms are computationally intensive, or make use of very large data sets. These algorithms take long periods of time to execute, or are unable to execute due to memory constraints. The implementation of tasks and data in parallel presents a simple and effective solution to this problem. This workshop will provide a hands-on demonstration of the different levels of parallelisation that can be achieved for an algorithm in MATLAB with different levels of code changes, ranging from inherent parallel processing to intricate parallel task management.

Topics to be covered:

• Introduction to parallel computing
• Speeding up algorithms without code changes
• Developing task parallel applications
• Developing data parallel applications
• Scheduling your programs on a computer cluster
• Tips, Wrap-up, Questions and Answers

Should you be interested in having OPTI-NUM solutions host this Free Workshop at your University, please let us know by sending an e-mail to info@chpc.ac.za.