Parallel Computing Toolbox 4.3

Scaling Up to Clusters, Grids, and Clouds Using MATLAB Distributed Computing Server

Parallel Computing Toolbox provides the ability to use up to eight local workers on a multicore or multiprocessor computer using a single toolbox license. When used with MATLAB Distributed Computing Server, you can run an application using any number of workers on large-scale resources such as computer clusters or grid and cloud computing services. The server also supports both interactive and batch workflows.

Using MATLAB Compiler, you can build MATLAB applications that use toolbox functions into standalone executables or shared software components and distribute them royalty-free. These executables and libraries can connect to MATLAB Distributed Computing Server workers and perform MATLAB computations on large computing resources.

Running a gene regulation model on a cluster using MATLAB Distributed Computing Server. The server enables applications developed using Parallel Computing Toolbox to harness computer clusters for large problems. Compiled parallel MATLAB applications can also use the server.

Integrating Products into Your Computing Environment

MATLAB Distributed Computing Server provides a job manager and directly supports several third-party schedulers such as Platform LSF^®, Microsoft^® Windows^® HPC Server, Altair PBS Pro^®, and TORQUE. Using the Configurations Manager in the toolbox, you can maintain named settings such as scheduler type, path settings, and cluster usage policies. Switching between clusters or schedulers typically requires changing the configuration name only.

MATLAB Distributed Computing Server dynamically enables the required product licenses when an application runs on the cluster, based on the user’s profile. As a result, administrators need to manage only the server license on the cluster rather than separate toolbox and blockset licenses for every cluster user.