This section shows benchmark results using several Amazon EC2 instances for the standard Fluent F1 racing car, that is part of the ANSYS benchmark collection. They have all been run using AWS ParallelCluster in the same way as described in this workshop.
This case has been used since the early versions of Fluent to assess HPC performance. It consists of a race car geometry in a single region with a mesh of approximately 140m cells. The actual case file can be obtained from ANSYS directly and these particular numbers were taken using Fluent’s automatic benchmark mode. This time has been then recalculated as the number of iterations per minute to aid visualization.
All three graphs show the excellent scaling compared to an on-premise Cray XC50 supercomputer, whose datapoints is taken from the ANSYS website. This is thanks to the network optimized instance; c5n.18xlarge which includes the Elastic Fabric Adapter (EFA).
These graphs are shown per node, per core and cells per core to aid visualization of the results.