Autodyn has been designed for parallel processing on a variety of systems ranging from a Massively Parallel Processor (MPP) using shared memory to heterogeneous distributed networks of computers.
When using Autodyn for parallel processing, data must be exchanged between cooperating tasks, and some message passing protocol has to be used to achieve this. We currently use MPI (Message Passing Interface) to allow a heterogeneous collection of computers networked together to be viewed by Autodyn as a single parallel computer. The component computers can be single or multiple processor machines, including MPPs (Massively Parallel Processors).
The following MPI parallel solutions are available for the standalone Autodyn solver and Explicit Dynamics systems. Parallel solutions are not supported for the Autodyn component system available from the Workbench framework.
| Windows | Linux | |||||
|---|---|---|---|---|---|---|
| Local Parallel | Distributed Parallel | Windows HPC Jobscheduler | Local Parallel | Distributed Parallel | Parallel Jobschedulers | |
| Explicit STR (Mechanical) | Intel MPI, Microsoft MPI | Intel MPI | N/A | Intel MPI, OpenMPI | Intel MPI, OpenMPI | N/A |
| Autodyn | Intel MPI, Microsoft MPI | Intel MPI | N/A | Intel MPI, OpenMPI | Intel MPI, OpenMPI | N/A |
Local Parallel means that the job is run on one machine with multiple cores.
Distributed Parallel means that the job is run on multiple machines each with multiple cores.
Instructions for setting up the various supported MPI Software packages are found in the remaining sections of this chapter.
If you wish to use the parallel processing options on multiple computers, you must install an implementation of the MPI message passing protocol from one of the supported providers on all the machines you intend to use. If the parallel processing options are used on a single machine, this installation is not necessary.
Intel MPI and OpenMPI are part of the Ansys unified installation package.
Whether you intend to run a calculation in serial mode or parallel mode, Autodyn always runs using the autodyn process.
If a serial calculation is being performed, this is the only process that is run. It performs all aspects of the problem (for example, setup, execution, I/O, post processing).
If a parallel calculation is being performed, this process is considered to be the master process and additional worker processes are started on each CPU used by the parallel system (including the CPU running the master process). These worker processes are spawned automatically by the MPI process, which issues the required system commands to start the worker process once on each CPU. This parallel processing environment is shown schematically in the figure above.
The setup instructions described here assume that the configuration
being implemented consists of n hosts, each having
m CPU's, and that the hostname of the nth host is hostn. Host1 is special, in that it is
the head node which runs the master process (autodyn) and is
usually the system on which the program files are stored.
When running Autodyn in parallel on a system of networked computers, the program executables and libraries must be accessible on each of the computers. They can either be installed on all computers in the system or installed on one computer (usually the one running the master process) and shared with the others using NFS mounting or Windows Sharing. We recommend sharing as it simplifies the handling of the files. Setup is further simplified if the shared mount points/folders are the same on each host (including the master).