Scheduler Support in Parallel Computing Toolbox and MATLAB Distributed Computing Server
Schedulers manage, monitor, and distribute workload, and they administer resources across distributed computing systems comprising disparate hardware and software resources.
The scheduler interface provided by MathWorks parallel computing products is at the software level, providing engineers and scientists an interface to submit jobs to computation resources without having to be concerned with differences in operating systems, environments, and schedulers. The infrastructure allows administrators to integrate MathWorks parallel computing products with existing scheduler environments at the application layer.
MATLAB Distributed Computing Server™ includes a MATLAB® job scheduler and support for other schedulers. See the following table for details.
|Scheduler||Turn-Key Integration Provided||Customizable Integration*||Sample Custom Integration Scripts Provided**||Additional Comments|
|MATLAB job scheduler||Included with MATLAB Distributed Computing Server|
|IBM® Platform LSF|
|Microsoft® Windows® HPC Server||Customizable via job templates|
|PBS family||Support for PBS Professional® and TORQUE|
|Grid Engine family||Support for Oracle® Grid Engine and Sun® N1 Grid Engine|
|SLURM||Integration scripts available on MathWorks File Exchange|
|Other schedulers||Integration via the generic scheduler interface, using the sample custom integration scripts as a basis|
* Customizable integration uses the generic scheduler interface.
** Integration scripts are included with Parallel Computing Toolbox (in matlabroot\toolbox\distcomp\examples\integration).