As finite element models increase in size and fidelity, HPC environments are being used to improve simulation throughput. This enables parallel processing to go beyond the multi-core processing on a single computer by distributing the analysis across multiple machines or nodes of a compute cluster. In this way, the largest computational problems can be broken down and solved many times quicker, but hardware is not the only requirement. The computational software needs an architecture that can take advantage of this scalability by distributing the analysis tasks and required information between machines.
These demands can be addressed by running nCode DesignLife in HPC environments or other multi-machine configurations using its Distributed Processing option. Turnaround time for the largest jobs can be improved by an order of magnitude with Distributed Processing, opening up the possibility of more robust up-front design through simulation.
How can you perform quicker CAE fatigue prediction?
Use more computing power!
nCode DesignLife scales nearly linerarly with:
- Increasing number of processing threads on a single computer
- Increasing number of computers in a cluster
Intel® MPI (Message Passing Interface) and Microsoft® MPI are supported on Windows® platforms. Intel MPI and IBM® MPI are supported on Linux.
Learn more by watching a webinar on-demand, "Top 10 Ways to Perform a Faster Fatigue Analysis"