On 23 of April 2013, a two-day course «MLPC / Parallel Computing in MATLAB» successfully ended. The course was conducted by our resource center RC CC along with the company's training center Softline.

MATLAB is a high-level programming language and interactive environment for computing, visualization and programming. It enables you to perform data analysis, development of algorithms, creation of mathematical models and applications. MATLAB  language, supplemented with Mathworks toolkit software and integrated highly optimized mathematical functions, allows solving problems in many ways and getting a solution faster than using traditional programming languages.

Currently, a lot of different problems, such as signal processing problems, rendering images and video, creation of control systems, problems of financial mathematics, computational biology, chemistry, and others, can be solved by using MATLAB .

Of particular interest is a fast-paced package for MATLAB - Parallel Computing Toolbox. Recently in this package have been incorporated many flexible and powerful features such as the use of data and task parallelism, possibility of parallel computing on remote clusters in the clouds, the use of CUDA, including direct use of CUDA-cores.

The main advantage of this functional implementation in MATLAB is the ability to exonerate the programmer from the management of creation and destruction processes, memory allocation and cleaning. At the same time, if necessary, all of this can be performed manually.

In this two-day course with practical and theoretical classes, participants learned techniques for using MATLAB Parallel Toolbox and try out the key features of the package. As a result, the course participants were able to significantly speed up the computation of their scientific problems. Practical classes were carried out on virtual cluster of Resource Center "Computer Center SPbU".

After the completion of the course, all participants were given certificates.