SPEC® MPIM2007 Result

Copyright 2006-2010 Standard Performance Evaluation Corporation

Lenovo

ThinkSystem SR665 V3 (AMD EPYC 9654)

MPI2007 license: 28 Test date: Jan-2023
Test sponsor: Lenovo Hardware Availability: Feb-2023
Tested by: Lenovo Software Availability: Feb-2023
Benchmark results graph

Results Table

Benchmark Base Peak
Ranks Seconds Ratio Seconds Ratio Seconds Ratio Ranks Seconds Ratio Seconds Ratio Seconds Ratio
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
104.milc 192 28.3 55.2 28.3 55.3 28.2 55.6 192 28.3 55.2 28.3 55.3 28.2 55.6
107.leslie3d 192 86.1 60.6 86.0 60.7 86.0 60.7 192 86.1 60.6 86.0 60.7 86.0 60.7
113.GemsFDTD 192 129   48.9 128   49.1 129   49.0 192 129   48.9 128   49.1 129   49.0
115.fds4 192 37.6 51.8 37.2 52.4 38.6 50.5 192 37.6 51.8 37.2 52.4 38.6 50.5
121.pop2 192 77.8 53.1 77.7 53.1 77.5 53.3 192 77.8 53.1 77.7 53.1 77.5 53.3
122.tachyon 192 44.5 62.9 43.9 63.7 44.9 62.2 192 44.5 62.9 43.9 63.7 44.9 62.2
126.lammps 192 71.2 40.9 65.5 44.5 65.8 44.3 192 71.2 40.9 65.5 44.5 65.8 44.3
127.wrf2 192 62.9 124   63.6 123   63.5 123   192 62.9 124   63.6 123   63.5 123  
128.GAPgeofem 192 20.2 102   21.1 97.7 20.3 102   192 20.2 102   21.1 97.7 20.3 102  
129.tera_tf 192 73.6 37.6 73.6 37.6 74.0 37.4 192 73.6 37.6 73.6 37.6 74.0 37.4
130.socorro 192 37.6 101   37.7 101   38.3 99.7 192 37.6 101   37.7 101   38.3 99.7
132.zeusmp2 192 43.1 72.0 43.1 72.0 43.0 72.2 192 43.1 72.0 43.1 72.0 43.0 72.2
137.lu 192 40.2 91.6 40.0 92.0 39.5 93.0 192 40.2 91.6 40.0 92.0 39.5 93.0
Hardware Summary
Type of System: Homogeneous
Compute Node: ThinkSystem SR665 V3
Total Compute Nodes: 1
Total Chips: 2
Total Cores: 96
Total Threads: 192
Total Memory: 768 GB
Base Ranks Run: 192
Minimum Peak Ranks: 192
Maximum Peak Ranks: 192
Software Summary
C Compiler: AMD Optimizing C/C++ and Fortran Compilers (AOCC)
Version 4.0.0 Build 389 for Linux
C++ Compiler: AMD Optimizing C/C++ and Fortran Compilers (AOCC)
Version 4.0.0 Build 389 for Linux
Fortran Compiler: AMD Optimizing C/C++ and Fortran Compilers (AOCC)
Version 4.0.0 Build 389 for Linux
Base Pointers: 64-bit
Peak Pointers: 64-bit
MPI Library: Open MPI Library for Linux
Version 4.1.1
Other MPI Info: None
Pre-processors: No
Other Software: None

Node Description: ThinkSystem SR665 V3

Hardware
Number of nodes: 1
Uses of the node: compute
Vendor: Lenovo
Model: ThinkSystem SR665 V3
CPU Name: AMD EPYC 9654
CPU(s) orderable: 1, 2 chip
Chips enabled: 2
Cores enabled: 96
Cores per chip: 96
Threads per core: 2
CPU Characteristics: Max. Boost Clock upto 3.7 GHz
CPU MHz: 2400
Primary Cache: 32 KB I + 32 KB D on chip per core
Secondary Cache: 1 MB I+D on chip per core
L3 Cache: 384 MB I+D on chip per chip
32 MB shared / 8 cores
Other Cache: None
Memory: 768 GB (24 x 32 GB 2Rx8 PC5-4800B-R)
Disk Subsystem: 1x ThinkSystem 2.5" 5300 480GB SSD
Other Hardware: None
Adapter: Mellanox ConnectX-6 HDR
Number of Adapters: 1
Slot Type: PCIe Gen5 x16
Data Rate: 200Gb
Ports Used: 1
Interconnect Type: Mellanox ConnectX-6 HDR
Software
Adapter: Mellanox ConnectX-6 HDR
Adapter Driver: Mellanox
Adapter Firmware: 20.28.1002
Operating System: Red Hat Enterprise Linux Server release 8.6,
Kernel 4.18.0-372.9.1.el8.x86_64
Local File System: ext4
Shared File System: None
System State: Multi-user, run level 3
Other Software: None

Submit Notes

The config file option 'submit' was used.

General Notes

 MPI startup command:
   mpiexec command was used to start MPI jobs.
 Yes: The test sponsor attests, as of date of publication,
 that CVE-2017-5754 (Meltdown) is mitigated in the system as tested and documented.
 Yes: The test sponsor attests, as of date of publication,
 that CVE-2017-5753 (Spectre variant 1) is mitigated in the system as tested and documented.
 Yes: The test sponsor attests, as of date of publication,
 that CVE-2017-5715 (Spectre variant 2) is mitigated in the system as tested and documented.

Base Compiler Invocation

C benchmarks:

 mpicc 

C++ benchmarks:

126.lammps:  mpic++ 

Fortran benchmarks:

 mpif90 

Benchmarks using both Fortran and C:

 mpicc   mpif90 

Base Portability Flags

104.milc:  -DSPEC_MPI_LP64 
115.fds4:  -DSPEC_MPI_LP64 
121.pop2:  -DSPEC_MPI_CASE_FLAG   -DSPEC_MPI_LP64 
122.tachyon:  -DSPEC_MPI_LP64 
127.wrf2:  -DSPEC_MPI_CASE_FLAG   -DSPEC_MPI_LINUX   -DSPEC_MPI_LP64 
128.GAPgeofem:  -DSPEC_MPI_LP64 
130.socorro:  -DSPEC_MPI_LP64 
132.zeusmp2:  -DSPEC_MPI_LP64 

Base Optimization Flags

C benchmarks:

 -Ofast   -flto   -ffast-math   -march=znver4   -lamdlibm 

C++ benchmarks:

126.lammps:  -Ofast   -flto   -ffast-math   -march=znver4   -DMPICH_IGNORE_CXX_SEEK(*) 

Fortran benchmarks:

 -Ofast   -flto   -ffast-math   -march=znver4   -funroll-loops 

Benchmarks using both Fortran and C:

115.fds4:  -Ofast   -flto   -ffast-math   -march=znver4   -funroll-loops 
121.pop2:  Same as 115.fds4 
127.wrf2:  Same as 115.fds4 
128.GAPgeofem:  -Ofast   -flto   -ffast-math   -march=znver4   -funroll-loops   -lamdlibm 
130.socorro:  Same as 115.fds4 
132.zeusmp2:  Same as 115.fds4 

(*) Indicates an optimization flag that was found in a portability variable.

Base Other Flags

Benchmarks using both Fortran and C:

127.wrf2:  -Wno-return-type 

Peak Optimization Flags

C benchmarks:

104.milc:  basepeak = yes 
122.tachyon:  basepeak = yes 

C++ benchmarks:

126.lammps:  basepeak = yes 

Fortran benchmarks:

107.leslie3d:  basepeak = yes 
113.GemsFDTD:  basepeak = yes 
129.tera_tf:  basepeak = yes 
137.lu:  basepeak = yes 

Benchmarks using both Fortran and C:

115.fds4:  basepeak = yes 
121.pop2:  basepeak = yes 
127.wrf2:  basepeak = yes 
128.GAPgeofem:  basepeak = yes 
130.socorro:  basepeak = yes 
132.zeusmp2:  basepeak = yes 

The flags file that was used to format this result can be browsed at
http://www.spec.org/mpi2007/flags/amd2021_flags.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/mpi2007/flags/amd2021_flags.xml.