Reference Guide

ANSYS CFX Performance
15 Dell EMC Ready Solutions for HPC Digital Manufacturing with AMD EPYC™ Processors—ANSYS®
Performance
Figure 5 ANSYS CFX Parallel Scaling—AMD EPYC 7452
Figure 5 presents the parallel scalability when running CFX with up to eight CBB nodes configured with AMD
EPYC 7452 processors and ConnectX-6 HDR100 InfiniBand. All processor cores in each server were used
when running these benchmarks. The performance at each node count is presented relative to the
performance of a single node.
The parallel scalability for these benchmarks is as expected, with the largest model demonstrating super-
linear scaling. Super-linear scaling is typically observed when more of the active data fits into the processor
cache at larger node counts. For the Airfoil_100M case, super linear scaling is observed starting at four
nodes.
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
1 2 4 6 8
Performance Relative to One Node
Number of Nodes
ANSYS CFX Parallel Scaling—AMD EPYC 7452
Airfoil_10M Airfoil_50M Airfoil_100M LeMans Pump