LSTC LS-DYNA® Performance with AMD EPYC™ 7002 Series Processors

LSTC LS-DYNA® Performance with AMD EPYC™ 7002 Series Processors


This article was written by Joshua Weage of the Dell EMC HPC & AI Innovation Lab in October 2019.

Introduction

This blog discusses the performance of LSTC LS-DYNA® on the Dell EMC Ready Solutions for HPC Digital Manufacturing with AMD EPYC™ 7002 series processors. This Dell EMC Ready Solutions for HPC was designed and configured specifically for digital manufacturing workloads, where computer-aided engineering (CAE) applications are critical for virtual product development. The Dell EMC Ready Solutions for HPC Digital Manufacturing uses a flexible building block approach to HPC system design, where individual building blocks can be combined to build HPC systems which are optimized for customer-specific workloads and use cases.

The Dell EMC Ready Solutions for HPC Digital Manufacturing is one of many solutions in the Dell EMC HPC solution portfolio. Please visit www.dellemc.com/hpc for a comprehensive overview of the HPC solutions offered by Dell EMC.




Benchmark System Configuration

Performance benchmarking was performed using dual-socket Dell EMC PowerEdge servers with 7001 and 7002 series AMD EPYC processors. All servers were populated with two processors and one DIMM per channel memory configuration. The system configurations used for the performance benchmarking are shown in Table 1 and Table 2. The BIOS configuration used for the benchmarking systems is shown in Table 3.


Table 1 7001 Series AMD EPYC System Configuration
Server Dell EMC PowerEdge R7425
Processor 2x AMD EPYC 7601 32-core Processors
Memory 16x16GB 2400 MTps RDIMMs
BIOS Version 1.10.6
Operating System Red Hat Enterprise Linux Server release 7.5
Kernel Version 3.10.0-862.el7.x86_64


Table 2 7002 Series AMD EPYC System Configuration
Server Dell EMC PowerEdge C6525
Processor 2x AMD EPYC 7702 64-Core Processors
2x AMD EPYC 7502 32-Core Processors
2x AMD EPYC 7452 32-Core Processors
2x AMD EPYC 7402 24-Core Processors
Memory 16x16GB 3200 MTps RDIMMs
BIOS Version 1.0.1
Operating System Red Hat Enterprise Linux Server release 7.6
Kernel Version 3.10.0-862.el7.x86_64

Table 3 BIOS Configuration
System Profile Performance Optimized
Logical Processor Disabled
Virtualization Technology Disabled
NUMA Nodes Per Socket 4 (PowerEdge C6525 only


Software and Benchmark Versions

Application software and benchmark versions are as described in Table 4


Table 4 Software and Benchmark Versions
LSTC LS-DYNA ls-dyna_mpp_s_R11_1_0_x64_centos65_ifort160_avx2_platformmpi
MPI Platform MPI 9.1.4.3
car2car Benchmark car2car-ver10.tar.gz from www.topcrunch.org
ODB-10M Benchmark odb10m-ver16.k.gz from www.topcrunch.org



LSTC LS-DYNA Performance

LSTC LS-DYNA is a multi-physics finite element analysis (FEA) software commonly used in multiple engineering disciplines. Depending on the specific solution type, FEA codes may or may not scale well across multiple processor cores and servers. The two benchmark cases presented here use the LS-DYNA explicit FEA solver, which typically scales more efficiently than the implicit FEA solver.
The car2car benchmark is a simulation of a two-vehicle collision. This benchmark model contains 2.4 million elements, which is relatively modest compared to current automotive industry usage. The ODB-10M benchmark is a simulation of a vehicle colliding into an offset deformable barrier. This benchmark model contains 10.6 million elements, which is representative of current automotive industry usage. For these benchmarks, the simulation end time was set to 0.02 sec. The benchmark results reported here are single-server performance results, with the benchmark run using all processor cores available in the server. The performance for LS-DYNA is measured using the Elapsed Time value. This value is the total elapsed time in seconds as reported by LS-DYNA, so a smaller elapsed time is better performance.


Figure 1 Single Server Performance

LSTC LS-DYNA Performance

The results in Figure 1 are plotted relative to the performance of a single server configured with AMD EPYC 7601 processors. Larger values indicate better overall performance. These results show the performance advantage available with 7002 series AMD EPYC processors. The 32-core AMD EPYC 7452 and 7502 processors provide very good performance for these benchmarks. Per-server, the 64-core AMD EPYC 7702 provides a significant performance advantage over the 32-core processors, particularly for the larger ODB-10M benchmark case.

Conclusion

The results presented in this blog show that 7002 series AMD EPYC processors offer a significant performance improvement for LSTC LS-DYNA relative to 7001 series AMD EPYC processors.




Article ID: SLN319179

Last Date Modified: 10/30/2019 09:55 AM


Rate this article

Accurate
Useful
Easy to understand
Was this article helpful?
Yes No
Send us feedback
Comments cannot contain these special characters: <>()\
Sorry, our feedback system is currently down. Please try again later.

Thank you for your feedback.