SPEChpc™ 2021 Medium Result

Copyright 2021-2023 Standard Performance Evaluation Corporation

Intel

Endeavour: Intel Server D50DNP2MFALACB (Intel
Xeon Platinum 8480+)

SPEChpc 2021_med_base = 5.15

SPEChpc 2021_med_peak = 5.42

hpc2021 License: 13 Test Date: Jul-2023
Test Sponsor: Intel Hardware Availability: Jan-2023
Tested by: Intel Software Availability: Apr-2023

Benchmark result graphs are available in the PDF report.

Results Table

Benchmark Base Peak
Model Ranks Thrds/Rnk Seconds Ratio Seconds Ratio Seconds Ratio Model Ranks Thrds/Rnk Seconds Ratio Seconds Ratio Seconds Ratio
SPEChpc 2021_med_base 5.15
SPEChpc 2021_med_peak 5.42
Results appear in the order in which they were run. Bold underlined text indicates a median measurement.
705.lbm_m OMP 512 14 170 7.19 164 7.49 164 7.49 OMP 512 14 170 7.19 164 7.49 164 7.49
718.tealeaf_m OMP 512 14 410 3.29 421 3.21 417 3.24 OMP 512 14 410 3.29 421 3.21 417 3.24
719.clvleaf_m OMP 512 14 477 3.88 486 3.81 492 3.76 OMP 1024 7 419 4.41 446 4.15 424 4.37
728.pot3d_m OMP 512 14 542 3.41 552 3.35 552 3.35 OMP 1024 7 508 3.64 515 3.59 501 3.69
734.hpgmgfv_m OMP 512 14 268 3.73 277 3.62 272 3.67 OMP 1024 7 274 3.65 266 3.76 263 3.81
735.weather_m OMP 512 14 143 16.80 157 15.30 146 16.40 OMP 1024 7 136 17.60 139 17.20 137 17.50
Hardware Summary
Type of System: Homogenous Cluster
Compute Node: Intel Server D50DNP2MFALACB (Xeon 8480+)
Interconnect: Mellanox HDR
Compute Nodes Used: 64
Total Chips: 128
Total Cores: 7168
Total Threads: 14336
Total Memory: 32 TB
Max. Peak Threads: 14
Software Summary
Compiler: Intel oneAPI Compiler 2023.1.0
MPI Library: Intel MPI Library 2021.9 for Linux OS
Other MPI Info: None
Other Software: None
Base Parallel Model: OMP
Base Ranks Run: 512
Base Threads Run: 14
Peak Parallel Models: OMP
Minimum Peak Ranks: 512
Maximum Peak Ranks: 1024
Max. Peak Threads: 14
Min. Peak Threads: 7

Node Description: Intel Server D50DNP2MFALACB (Xeon 8480+)

Hardware
Number of nodes: 64
Uses of the node: Compute
Vendor: Intel
Model: Intel Server D50DNP2MFALACB (Xeon 8480+)
CPU Name: Intel Xeon Platinum 8480+
CPU(s) orderable: 1, 2 chips
Chips enabled: 2
Cores enabled: 112
Cores per chip: 56
Threads per core: 2
CPU Characteristics: Turbo Boost Technology up to 3.8 GHz
CPU MHz: 2000
Primary Cache: 32 KB I + 48 KB D on chip per core
Secondary Cache: 2 MB I+D on chip per core
L3 Cache: 105 MB I+D on chip per chip
Other Cache: None
Memory: 512 GB (16 x 32 GB 2Rx8 PC5-4800B)
Disk Subsystem: 1 x 1 TB NVMe U.2 2.5" SSD
Other Hardware: None
Accel Count: None
Accel Model: None
Accel Vendor: None
Accel Type: None
Accel Connection: None
Accel ECC enabled: None
Accel Description: None
Adapter: Mellanox ConnectX-6 HDR
Number of Adapters: 1
Slot Type: PCI-Express 4.0 x16
Data Rate: 200Gbit/s
Ports Used: 1
Interconnect Type: Mellanox HDR
Software
Accelerator Driver: None
Adapter: Mellanox ConnectX-6 HDR
Adapter Driver: 5.9-0.5.5
Adapter Firmware: 20.36.1010
Operating System: Rocky Linux 8.7 (Green Obsidian)
4.18.0-372.32.1.el8_6.crt3.x86_64
Local File System: NFS
Shared File System: PANASAS FS
System State: Multi-user
Other Software: None

Interconnect Description: Mellanox HDR

Hardware
Vendor: Mellanox
Model: Mellanox HDR
Switch Model: Mellanox MQM8790-HS2F Quantum HDR
InfiniBand Switch
Number of Switches: 18
Number of Ports: 40
Data Rate: 200 Gbit/s
Firmware: 20.36.1010
Topology: Fat-tree
Primary Use: MPI Traffic
Software

Submit Notes

The config file option 'submit' was used.

General Notes

The PANASAS filesystem as described on this result page was formerly
generally available.  At the time of this publication, it may
not be shipping, and/or may not be supported, and/or may fail
to meet other tests of General Availability described in the
SPEC HPG Policy document, http://www.spec.org/hpg/policy.html

Compiler Version Notes

==============================================================================
 CC  705.lbm_m(base, peak) 718.tealeaf_m(base, peak) 734.hpgmgfv_m(base,
      peak)
------------------------------------------------------------------------------
Intel(R) oneAPI DPC++/C++ Compiler 2023.1.0 (2023.1.0.20230320)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir:
  /global/panfs01/admin/opt/intel/oneAPI/2023.1.0.46401/compiler/2023.1.0/linux/bin-llvm
Configuration file:
  /global/panfs01/admin/opt/intel/oneAPI/2023.1.0.46401/compiler/2023.1.0/linux/bin-llvm/../bin/icx.cfg
------------------------------------------------------------------------------

==============================================================================
 FC  719.clvleaf_m(peak) 728.pot3d_m(peak)
------------------------------------------------------------------------------
ifort (IFORT) 2021.9.0 20230302
Copyright (C) 1985-2023 Intel Corporation.  All rights reserved.
------------------------------------------------------------------------------

==============================================================================
 FC  719.clvleaf_m(base) 728.pot3d_m(base) 735.weather_m(base, peak)
------------------------------------------------------------------------------
ifx (IFX) 2023.1.0 20230320
Copyright (C) 1985-2023 Intel Corporation. All rights reserved.
------------------------------------------------------------------------------

Base Compiler Invocation

C benchmarks:

 mpiicc -cc=icx 

Fortran benchmarks:

 mpiifort -fc=ifx 

Base Portability Flags

705.lbm_m:  -lstdc++   -std=c++14 
718.tealeaf_m:  -lstdc++   -std=c++14 
734.hpgmgfv_m:  -lstdc++   -std=c++14 

Base Optimization Flags

C benchmarks:

 -O3   -Ofast   -xCORE-AVX512   -mprefer-vector-width=512   -qopt-multiple-gather-scatter-by-shuffles   -fiopenmp   -ffast-math   -flto   -funroll-loops 

Fortran benchmarks:

 -O3   -Ofast   -xCORE-AVX512   -mprefer-vector-width=512   -qopt-multiple-gather-scatter-by-shuffles   -fiopenmp   -ffast-math   -flto   -funroll-loops   -nostandard-realloc-lhs   -align array64byte 

Base Other Flags

C benchmarks:

 -Wno-incompatible-function-pointer-types 

Peak Compiler Invocation

C benchmarks:

 mpiicc -cc=icx 

Fortran benchmarks (except as noted below):

 mpiifort 
735.weather_m:  mpiifort -fc=ifx 

Peak Portability Flags

705.lbm_m:  -lstdc++   -std=c++14 
718.tealeaf_m:  -lstdc++   -std=c++14 
734.hpgmgfv_m:  -lstdc++   -std=c++14 

Peak Optimization Flags

C benchmarks:

705.lbm_m:  basepeak = yes 
718.tealeaf_m:  basepeak = yes 
734.hpgmgfv_m:  -O3   -Ofast   -xCORE-AVX512   -mprefer-vector-width=512   -qopt-multiple-gather-scatter-by-shuffles   -fiopenmp   -ffast-math   -flto   -funroll-loops 

Fortran benchmarks:

719.clvleaf_m:  -O3   -Ofast   -xCORE-AVX512   -ansi-alias   -qopenmp   -ipo   -qopt-zmm-usage=high   -qopt-multiple-gather-scatter-by-shuffles   -nostandard-realloc-lhs   -align array64byte 
728.pot3d_m:  Same as 719.clvleaf_m 
735.weather_m:  -O3   -Ofast   -xCORE-AVX512   -mprefer-vector-width=512   -qopt-multiple-gather-scatter-by-shuffles   -fiopenmp   -ffast-math   -flto   -funroll-loops   -nostandard-realloc-lhs   -align array64byte 

Peak Other Flags

C benchmarks:

 -Wno-incompatible-function-pointer-types 

The flags file that was used to format this result can be browsed at
http://www.spec.org/hpc2021/flags/Intel_compiler_flags.2023-08-16.html.

You can also download the XML flags source by saving the following link:
http://www.spec.org/hpc2021/flags/Intel_compiler_flags.2023-08-16.xml.