SPEC® MPIM2007 Result

Copyright 2006-2010 Standard Performance Evaluation Corporation

Supermicro

Hyper A+ Server AS -2126HS-TN (AMD EPYC 9755)

MPI2007 license: 6569 Test date: Sep-2024
Test sponsor: Supermicro Hardware Availability: Oct-2024
Tested by: Supermicro Software Availability: Apr-2024
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 256 18.5 84.6 18.4 85.1 18.3 85.6 256 18.5 84.6 18.4 85.1 18.3 85.6
107.leslie3d 256 61.4 85.0 61.5 84.9 61.9 84.4 256 61.4 85.0 61.5 84.9 61.9 84.4
113.GemsFDTD 256 104   60.7 104   60.6 104   60.5 256 104   60.7 104   60.6 104   60.5
115.fds4 256 13.9 141   13.9 140   13.9 140   256 13.9 141   13.9 140   13.9 140  
121.pop2 256 61.5 67.2 61.6 67.0 61.3 67.4 256 61.5 67.2 61.6 67.0 61.3 67.4
122.tachyon 256 30.2 92.6 29.4 95.0 29.4 95.1 256 30.2 92.6 29.4 95.0 29.4 95.1
126.lammps 256 64.2 45.4 64.2 45.4 64.1 45.5 256 64.2 45.4 64.2 45.4 64.1 45.5
127.wrf2 256 42.1 185   42.5 183   42.1 185   256 42.1 185   42.5 183   42.1 185  
128.GAPgeofem 256 14.1 147   14.0 147   14.0 147   256 14.1 147   14.0 147   14.0 147  
129.tera_tf 256 24.4 114   24.4 114   24.4 114   256 24.4 114   24.4 114   24.4 114  
130.socorro 256 37.6 102   37.6 101   38.5 99.1 256 37.6 102   37.6 101   38.5 99.1
132.zeusmp2 256 33.0 93.9 33.0 94.0 33.0 94.0 256 33.0 93.9 33.0 94.0 33.0 94.0
137.lu 256 30.4 121   30.4 121   30.4 121   256 30.4 121   30.4 121   30.4 121  
Hardware Summary
Type of System: Homogeneous
Compute Node: Hyper A+ Server AS -2126HS-TN
Total Compute Nodes: 1
Total Chips: 2
Total Cores: 256
Total Threads: 256
Total Memory: 1536 GB
Base Ranks Run: 256
Minimum Peak Ranks: 256
Maximum Peak Ranks: 256
Software Summary
C Compiler: Intel oneAPI DPC++/C++ Compiler 2024.2.1
C++ Compiler: Intel oneAPI DPC++/C++ Compiler 2024.2.1
Fortran Compiler: Intel oneAPI DPC++/C++ Compiler 2024.2.1
Base Pointers: 64-bit
Peak Pointers: 64-bit
MPI Library: Intel MPI Version 2021.13
Other MPI Info: None
Pre-processors: No
Other Software: Jemalloc-5.3.0

Node Description: Hyper A+ Server AS -2126HS-TN

Hardware
Number of nodes: 1
Uses of the node: compute
Vendor: Supermicro
Model: Hyper A+ Server AS -2126HS-TN
CPU Name: AMD EPYC 9755
CPU(s) orderable: 1,2 chips
Chips enabled: 2
Cores enabled: 256
Cores per chip: 128
Threads per core: 1
CPU Characteristics: Max. Boost Clock upto 4.1GHz
CPU MHz: 2700
Primary Cache: 32 KB I + 48 KB D on chip per core
Secondary Cache: 1 MB I+D on chip per core
L3 Cache: 512 MB I+D on chip per chip,
32 MB shared / 8 cores
Other Cache: None
Memory: 1536 GB (24 x 64 GB 2Rx4 PC5-6400B-R,
running at 6000)
Disk Subsystem: 1 x 3.5 TB NVMe SSD
Other Hardware: None
Adapter: None
Number of Adapters: 1
Slot Type: None
Data Rate: None
Ports Used: 0
Interconnect Type: None
Software
Adapter: None
Adapter Driver: None
Adapter Firmware: None
Operating System: Ubuntu 24.04 LTS
6.8.0-44-generic
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.
mpiexec.hydra -bootstrap ssh -hosts localhost -genv I_MPI_COMPATIBILITY=3 -np $ranks -ppn $ranks $command

General Notes

 MPI startup command:
   mpiexec.hydra command was used to start MPI jobs.
 RAM configuration:
   Compute nodes have 1 x 64 GB RDIMM on each memory channel.
 BIOS settings:
   SMT = Disabled
   NUMA nodes per socket = NPS4
   ACPI SRAT L3 Cache as NUMA Domain = Enabled
   Determinism Control = Manual
   Determinism Enable = Power
   xGMI Link Configuration = 4 xGMI Links
   4 Link xGMI max speed = 32Gbps
   TDP Control = Manual
   TDP = 500
   Package Power Limit Control = Manual
   Package Power Limit = 500
 NA: 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:

 mpiicc -cc=icx 

C++ benchmarks:

126.lammps:  mpiicpc -cxx=icpx 

Fortran benchmarks:

 mpiifort -fc=ifx 

Benchmarks using both Fortran and C:

 mpiicc -cc=icx   mpiifort -fc=ifx 

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 
126.lammps:  -DMPICH_IGNORE_CXX_SEEK 
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   -ipo   -march=skylake-avx512   -mtune=skylake-avx512   -ansi-alias 

C++ benchmarks:

126.lammps:  -Ofast   -ipo   -march=skylake-avx512   -mtune=skylake-avx512   -ansi-alias 

Fortran benchmarks:

 -Ofast   -ipo   -march=skylake-avx512   -mtune=skylake-avx512   -nostandard-realloc-lhs   -align array64byte 

Benchmarks using both Fortran and C:

 -Ofast   -ipo   -march=skylake-avx512   -mtune=skylake-avx512   -ansi-alias   -nostandard-realloc-lhs   -align array64byte 

Base Other Flags

C benchmarks:

104.milc:  -Wno-implicit-function-declaration -Wno-implicit-int   -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 
122.tachyon:  -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 

C++ benchmarks:

126.lammps:  -Wno-register   -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 

Fortran benchmarks:

 -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 

Benchmarks using both Fortran and C (except as noted below):

 -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 
127.wrf2:  -Wno-implicit-int -Wno-implicit-function-declaration -Wno-int-conversion   -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 
128.GAPgeofem:  -Wno-implicit-function-declaration -Wno-implicit-int   -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 
132.zeusmp2:  -Wno-implicit-function-declaration   -limf   -Wl,--rpath=/usr/local/lib -ljemalloc 

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/Intel_compiler_flags_hpc.2024.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/mpi2007/flags/Intel_compiler_flags_hpc.2024.xml.

SPEC and SPEC MPI are registered trademarks of the Standard Performance Evaluation Corporation. All other brand and product names appearing in this result are trademarks or registered trademarks of their respective holders.

For questions about this result, please contact the tester.
For other inquiries, please contact webmaster@spec.org
Copyright 2006-2010 Standard Performance Evaluation Corporation
Tested with SPEC MPI2007 v2.0.1.
Report generated on Thu Oct 10 12:36:16 2024 by SPEC MPI2007 HTML formatter v1463.
Originally published on 10 October 2024.