Newsletters November 2008

November 2008

Welcome to our final Newsletter of 2008. It seems that 2008 has flown by, and we are certainly living in interesting times, with the advent of a global financial crisis, the election of America’s first black president, and closer to home, the dawn of a new political party.

This newsletter provides a focus on High Performance Computing (HPC), describing HPC applications, and showing how HPC can be accomplished using The MathWorks Parallel Computing Toolbox and MATLAB Distributed Computing Server products.

As always, you will find information on our upcoming public seminars, public training courses, and webinars available to you at your desktop. With broadband access becoming accessible to more South Africans, webinars are an effective way to learn about technology from The MathWorks without having to leave your office. Of course, if you would like us to visit you, you can always contact us on 011 325 6238, or e-mail info@optinum.co.za.

Please note that our offices will close on 19 December 2008 and re-open on 5 January 2009. We will also be closed between 8 and 10 December 2008 for a company meeting. You can contact us by e-mailing info@optinum.co.za for any urgent queries.

Wishing you a refreshing December break.
The OPTI-NUM solutions team. 

Release 2008b from The MathWorks

Release 2008b is the second release of The MathWorks products in 2008. This release features a number of enhancements for MATLAB products, including an enhanced development environment, including a function browser, automatic help for functions, deployment of Parallel Computing Toolbox applications, a new notebook interface in Symbolic Math Toolbox (plus access to the MuPAD symbolic engine) and the release of Econometrics Toolbox, which incorporates the functionality of GARCH Toolbox.

New features for Simulink users include the introduction of a MATLAB-based language in Simscape for authoring physical modelling domains and environments, support for embedding Simulink function-call subsystems in Stateflow charts, and fixed-point data types up to 128 bits for accelerated simulation, automatic code generation, Embedded MATLAB code, and Simulink Fixed Point.

For more information on R2008b, see the latest features page on The MathWorks web site.
Those customers who are current on Software Maintenance can expect to receive your DVDs in the next two weeks.

Technical Content: Focus on High Performance Computing

In this newsletter, we showcase The MathWorks solutions for High Performance Computing.

The Need for High Performance Computing

What is high performance computing (HPC) and why should I care? Do I need this stuff anyway? Isn’t it difficult to use HPC resources? This section describes the types of applications suitable for HPC treatment, and describes how The MathWorks tools can address your HPC needs. The article describes the three levels of user-interaction with HPC resources that MATLAB provides. Two of them are described in more detail in subsequent sections.
For the full technical article, click here.

Zero-Effort HPC

What if you could utilise HPC resources without changing any of your code, or the way you work? This section describes some of the tools available from The MathWorks  that automatically make use of your HPC cluster, if you have one available. An example shows how you can reduce testing time, enabling you to focus on choosing the correct parameter sweeps rather than worrying about how long that choice will take.
For the full technical article, click here.

HPC Made Easy: Using language constructs to scale your problem

From zero-effort HPC, you can “graduate” to using MATLAB constructs to seamlessly parallelise your tasks or data, without worrying about the communications required to do that. Parallel Computing Toolbox provides parfor and spmd constructs that effortlessly “collapse” into serial code when you are not using a cluster, but make utilising that cluster as easy as programming with MATLAB.

For the full technical article, click here.

Achieving HPC Scalability Using MathWorks Tools

What good is seamless scalability if you cannot access HPC resources of appropriate size. This section describes how Parallel Computing Toolbox provides access to local, departmental, corporate, grid and cloud resources, using configurations.

For the full technical article, click here.

Featured User Story

Max Planck Institute of Biochemistry Uses MathWorks Tools in its Quest to Cure Cancer

When researchers at Max Planck Institute of biochemistry needed to study the relationship between the structure and activity of macromolecular protein complexes involved in protein degradation in cells, they turned to MathWorks tools to develop streamlined procedures for their data-intensive applications. Using MATLAB Distributed Computing Server and Parallel Computing Toolbox, they were able to speed up their processing by 20 to 30 times.

Read the full user story.

View featured user stories.

Events

Our seminars and training courses in 2008 were well attended. By the time you read this, almost all of our events for 2008 will be completed. Visit us early in January for a list of seminars and training courses to be held in 2009.

Upcoming Seminars

06 October 2010

Johannesburg

Algorithm Development with MATLAB for C and C++ Programmers

Upcoming Webinars

07 September 2010	Speeding Up MATLAB Applications
09 September 2010	Image Acquisition and Processing Using GigE Vision Cameras with MATLAB
14 September 2010	Analyze, Model and Simulate Energy Risk with MATLAB – a SAP Integration at RWE

View all upcoming events.