MATLAB Distributed Computing Server 5.0
Product Description
- MATLAB Distributed Computing Server Introduction & Key Features
- Using MATLAB Distributed Computing Server
- Licensing
- Requirements and Installation
- Administering Clusters
Requirements and Installation
Hardware and Software Support
MATLAB Distributed Computing Server can be installed on any hardware platforms and operating systems that MATLAB and Simulink support. The server can execute any MATLAB and Simulink code that uses toolboxes and blocksets for which the user is licensed, with the exception of those on the list of ineligible products (see Supported Products section). It can also execute MATLAB computations received from executables and software components that were built from parallel MATLAB programs using MATLAB Compiler.
Multiple MATLAB Distributed Computing Server workers can be launched on a single computer, including on a user’s workstation. However, the benefits will accrue only with sufficient availability of RAM and enough processing cores on that computer. The recommendation is to run one worker per processing core.
Supported Schedulers
MATLAB Distributed Computing Server can be integrated with the MathWorks job manager or any third-party scheduler. The MathWorks job manager is intended for personal or small workgroup clusters that run MATLAB jobs exclusively.
Third-party schedulers are supported at two levels. For directly supported schedulers—Platform LSF, Windows Compute Cluster Server, PBS Pro, and TORQUE—the necessary integration scripts are available with the product. Other schedulers are indirectly supported and can be integrated using the product’s generic scheduler interface API (some examples are available in the product). For both directly and indirectly supported schedulers, the server workers are launched as any other programs that run on the cluster.
Learn more about scheduler support and integration
Installation and Configuration
Detailed instructions for configuring the installation, including customizations for operating systems and integration with various schedulers, are available online.
Store