Parallel Computing Toolbox™ 3.3

Product Description

• Parallel Computing Toolbox Key Features
• Programming Task-Parallel Applications
• Programming Data-Parallel Applications
• Working in an Interactive Parallel Environment
• Working in a Batch Environment
• Scaling to a Cluster Using MATLAB® Distributed Computing Server™

Working in an Interactive Parallel Environment

Parallel Computing Toolbox extends the MATLAB interactive environment, letting you use the familiar MATLAB environment for prototyping and developing task- and data-parallel applications. The matlabpool command sets up a task-parallel execution environment in which parfor-loops can be executed interactively from the MATLAB command prompt. This environment helps parfor-loops detect the presence of workers and transfer the necessary workspace data between your MATLAB session and the workers.

The Parallel Command Window sets up a data-parallel execution environment in which you can use distributed arrays, associated parallel functions, and message passing functions. Commands issued at the parallel prompt are executed simultaneously on all participating workers. Results are accessible immediately, enabling you to better track application behavior across your multiprocessor system at each step. Commands can be collected into a MATLAB function and submitted for offline execution. A profiling utility enables you to gauge performance of parallel applications that use distributed arrays and message passing functions.

Prototyping a data-parallel application in the Parallel Command Window. Click on image to see enlarged view.