For clusters with third-party schedulers
Our integration scripts, which are used for clusters with third-party schedulers, currently only support Linux-only clusters. End users can use Windows, Linux, or macOS. Our integration scripts can be found from the link below.
matlab-parallel
For clusters using MATLAB Job Scheduler
It is recommended that all systems involved in the cluster, including the head node and worker nodes, be on the same Operating System.
Parallel processing constructs that work on the infrastructure enabled by parpool—parfor, spmd, distributed arrays, and message passing functions—cannot be used on a heterogeneous cluster configuration. The underlying MPI infrastructure requires that all cluster computers have matching word sizes and processor endianness. A limited set of functions in Parallel Computing Toolbox can work in heterogeneous cluster configurations.
Please note that while it is recommended that all machines in the cluster are on the same Operating System, this requirement does not pertain to the client machines that are submitting the job to the cluster. The client machine can be running any supported Operating System and still submit to the cluster.
NOTE: Starting in R2019a the following name changes occurred:
- MATLAB Distributed Computing Server was renamed to MATLAB Parallel Server
- mdce_def was renamed to mjs_def
- mdce binary was renamed to mjs
