John Burkardt (ARC/ICAM) & Gene Cliﬀ (AOE/ICAM) Virginia Tech .......... FDI Fall Short Course: Introduction to Parallel MATLAB at Virginia Tech http://people.sc.fsu.edu/∼burkardt/presentations/fdi matlab 2009.pdf .......... ARC: Advanced Research Computing AOE: Department of Aerospace and Ocean Engineering ICAM: Interdisciplinary Center for Applied Mathematics

09 September 2009

Burkardt MATLAB Parallel Computing

MATLAB Parallel Computing: Some Announcements

While we have your attention...

Burkardt MATLAB Parallel Computing

MATLAB Parallel Computing: Some Announcements

ITHACA is an IBM iDataPlex cluster recently installed by Virginia Tech’s Advanced Research Computing facility. It is intended to gradually take over the high performance computing load from System X. ITHACA supports OpenMP, MPI and Parallel MATLAB programming. Right Now: Friendly users accepted for Ithaca (talk to John Burkardt if you are interested,); Mid October: Ithaca opened to general users (accounts given out through online application.).

Burkardt

MATLAB Parallel Computing

MATLAB Parallel Computing: Some Announcements

MATLAB Training is available this fall. These classes will be presented by the MathWorks. Some of these classes are tentative. Check the FDI website for details. 2 October: Simulink (daylong); 3 October: SimMechanics (daylong); 8 October: MATLAB Programming Techniques. 8 October: Parallel Computing with MATLAB. 29 October: Parallel Computing with MATLAB (daylong). 19 November: Real-time Data Acquisition and Control. 19 November: Statistical Methods in MATLAB.

Burkardt

MATLAB Parallel Computing

MATLAB Parallel Computing: Things Change

“Why There Isn’t Parallel MATLAB”

”There actually have been a few experimental versions of MATLAB for parallel computers... We have learned enough from these experiences to make us skeptical about the viability of a fully functional MATLAB running on today’s parallel machines.” Cleve Moler, 1995.

Burkardt

MATLAB Parallel Computing

MATLAB Parallel Computing: Things Change

(Let There Be) “Parallel MATLAB”

”We now have parallel MATLAB.” Cleve Moler, 2007.

Burkardt

MATLAB Parallel Computing

MATLAB Parallel Computing

Introduction Local Parallel Computing Parallelism with PARFOR The PRIME NUMBER Example Behind the Magic Bumps in the Road The MD Example Parallelism with SPMD Parallel Computing with fmincon Codistributed Arrays A 2D Heat Equation Parallel MATLAB at Virginia Tech Conclusion Burkardt MATLAB Parallel Computing

Introduction: MATLAB

MATLAB is a computing environment that is halfway between a programming language (where a user must do everything) and a menu-driven application (where the user only makes high level decisions). Users always have the ability to lay out the precise details of an algorithm themselves. They rely on MATLAB commands to access intelligent, ﬂexible, and optimized versions of standard algorithms.

Burkardt

MATLAB Parallel Computing

Introduction: MATLAB Adds Parallelism

The MathWorks has recognized that parallel computing is necessary for scientiﬁc computation. The underlying MATLAB core and algorithms are being extended to work with parallelism. An explicit set of commands has been added to allow the user to request parallel execution or to control distributed memory. New protocols and servers allow multiple copies of MATLAB to carry out the user’s requests, to transfer data and to communicate. MATLAB’s parallelism can be enjoyed by novices and exploited by experts.

Burkardt

MATLAB Parallel Computing

Introduction: Local Parallelism

MATLAB has developed a Parallel Computing Toolbox which is required for all parallel applications. The Toolbox allows a user to run a job in parallel on a desktop machine, using up to 8 ”workers” (additional copies of MATLAB) to assist the main copy. If the desktop machine has multiple processors, the workers will activate them, and...