Solver: Parallel Efficiency

CFX-5 - Parallel EfficiencyThe ANSYS CFX Solver has been designed from the ground up to be fully parallel, allowing you to easily distribute the CFD calculation across multiple processors for any set of physics and mesh type. Both CPU scalability and memory scalability are maintained, to make optimal use of your available computing resources.

Parallel operation is available on any combination of single- or multiple-CPU or networked UNIX workstations and Windows NT machines, including mixed UNIX/Windows NT clusters. This provides the most powerful method for solving large-scale CFD problems. The software automatically decomposes the grid into pieces that fit onto the individual processors, and the flow calculation then proceeds in parallel. This allows large problems to be computed faster or cases to be tackled that exceed the capacity of individual computers.

Efficient Scalable Parallel Performance

ANSYS CFX Performance and Scaling on AMD Opteron Processor