The MathWorks has scheduled a free on-site demonstration for Vanderbilt’s MATLAB community:
Topic: Speeding Up MATLAB Applications, including Parallel Computing
Date/Time: Wednesday, March 3rd, 2-5 p.m.
Location: 134 Featheringill
Agenda:
Presenter: Brett Shoelson, Ph.D., Principal Application Engineer
Brett will demonstrate simple ways you can improve and optimize your code to boost execution speed by orders of magnitude. He will also address common pitfalls in writing MATLAB code and explore the use of the MATLAB Profiler to find bottlenecks.
You will see how to perform parallel and distributed computing in MATLAB to solve computationally and data-intensive problems on multi-core computers and clusters. Brett will introduce you to parallel processing constructs such as parallel for-loops, distributed arrays, parallel numerical algorithms, and message-passing functions that let you implement task-parallel and data-parallel algorithms in MATLAB at a high level without programming for specific hardware and network architectures. In addition, he will show you how to run the same application on a computer cluster without changing the code.
Highlights include:
* Understanding memory usage and vectorization in MATLAB
* Addressing bottlenecks in your programs
* Optimizing file I/O to streamline your code
* Transitioning from serial to parallel MATLAB programs
* Executing applications on a single multi-core or multiprocessor desktop
* Performing parallel computing
Register at www.mathworks.com/seminars/va0310