Xeon & EPYC Tests

Various Server CPU benchmarks

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2307093-VE-2105245VE78
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

AV1 2 Tests
Bioinformatics 2 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 16 Tests
Compression Tests 2 Tests
CPU Massive 27 Tests
Creator Workloads 14 Tests
Cryptography 3 Tests
Encoding 3 Tests
Fortran Tests 4 Tests
Game Development 3 Tests
HPC - High Performance Computing 13 Tests
Java 2 Tests
Machine Learning 4 Tests
Molecular Dynamics 3 Tests
MPI Benchmarks 5 Tests
Multi-Core 28 Tests
NVIDIA GPU Compute 4 Tests
Intel oneAPI 4 Tests
OpenMPI Tests 6 Tests
Programmer / Developer System Benchmarks 9 Tests
Python 5 Tests
Renderers 7 Tests
Scientific Computing 6 Tests
Server 2 Tests
Server CPU Tests 22 Tests
Single-Threaded 6 Tests
Video Encoding 3 Tests
Common Workstation Benchmarks 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable
Show Perf Per RAM Channel Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 2 7R32
August 22 2020
  7 Hours, 24 Minutes
Xeon W-3275M
April 19 2020
  9 Hours, 52 Minutes
Xeon Platinum 8259CL
October 03 2020
  10 Hours, 8 Minutes
2 x Xeon Platinum 8259CL NUMA off
September 17 2020
  9 Hours, 36 Minutes
2 x Xeon Platinum 8259CL
September 16 2020
  9 Hours, 25 Minutes
Xeon Platinum 8272CL
October 19 2020
  9 Hours, 25 Minutes
2 x Xeon Platinum 8272CL NUMA off
October 18 2020
  9 Hours, 58 Minutes
2 x Xeon Platinum 8272CL
October 17 2020
  9 Hours, 33 Minutes
Xeon Platinum 8280
March 28 2020
  8 Hours, 39 Minutes
2 x Xeon Platinum 8280
March 28 2020
  8 Hours, 49 Minutes
Xeon Platinum 8375C
May 22 2021
  8 Hours, 39 Minutes
2 x Xeon Platinum 8375C NUMA off
May 23 2021
  9 Hours, 17 Minutes
2 x Xeon Platinum 8375C
May 21 2021
  10 Hours, 27 Minutes
Xeon Platinum 8480+
July 08 2023
  9 Hours, 6 Minutes
Invert Hiding All Results Option
  9 Hours, 18 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


Xeon & EPYC TestsProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkAudioOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionDisplay DriverEPYC 2 7R32Xeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+AMD EPYC 7R32 48-Core @ 2.80GHz (48 Cores / 96 Threads)Supermicro H11SSL-i v2.00 (2.1 BIOS)AMD Starship/Matisse8 x 16384 MB DDR4-3200MT/s HMA82GR7CJR8N-XN32GB Mass Storage + 256GB TOSHIBA THNSNJ25 + 600GB INTEL SSDSC2BB60 + 6001GB HGST HUS726060ALllvmpipe 126GBCrystal View2 x Intel I210Ubuntu 19.105.3.0-64-generic (x86_64)GNOME Shell 3.34.1X Server3.3 Mesa 19.2.8 (LLVM 9.0 128 bits)GCC 9.2.1 20191008ext4800x600Intel Xeon W-3275M @ 4.60GHz (28 Cores / 56 Threads)Supermicro X11SPA-T v1.01 (3.1 BIOS)Intel Sky Lake-E DMI3 Registers188GB2 x 1024GB Samsung SSD 970 PRO 1TBeVGA NVIDIA GeForce RTX 2080 Ti 11GB (1350/7000MHz)Realtek ALC888-VDBenQ SW271Intel I210 + Aquantia AQC107 NBase-T/IEEE5.3.0-46-generic (x86_64)GNOME Shell 3.34.3X Server 1.20.5NVIDIA 430.504.6.0GCC 9.2.1 20191008 + CUDA 10.13840x2160Intel Xeon Platinum 8259L @ 3.50GHz (24 Cores / 48 Threads)ASRockRack EP2C621D16-4LP (P2.20 BIOS)6 x 16384 MB DDR4-3200MT/s Hynix HMA82GR7CJR8N-XN256GB TOSHIBA THNSNJ25llvmpipe 93GBNVIDIA Device 10f8DELL UP2516D4 x Intel X722 for 1GbE5.3.0-64-generic (x86_64)GNOME Shell 3.34.1X Server3.3 Mesa 19.2.8 (LLVM 9.0 256 bits)GCC 9.2.1 201910082560x14402 x Intel Xeon Platinum 8259L @ 3.50GHz (48 Cores / 96 Threads)12 x 16384 MB DDR4-3200MT/s Hynix HMA82GR7CJR8N-XNllvmpipe 188GBIntel Xeon Platinum 8272CL @ 3.70GHz (26 Cores / 52 Threads)6 x 16384 MB DDR4-3200MT/s Hynix HMA82GR7CJR8N-XNllvmpipe 93GB2 x Intel Xeon Platinum 8272CL @ 3.70GHz (52 Cores / 104 Threads)12 x 16384 MB DDR4-3200MT/s Hynix HMA82GR7CJR8N-XN600GB INTEL SSDSC2BB60 + 10001GB Seagate ST10000NM0016-1T + 256GB TOSHIBA THNSNJ25llvmpipe 188GB256GB TOSHIBA THNSNJ25Intel Xeon Platinum 8280 @ 4.00GHz (28 Cores / 56 Threads)GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS)6 x 32 GB DDR4-2933MT/s HMA84GR7CJR4N-WM280GB INTEL SSDPED1D280GAASPEEDVE2282 x Intel X722 for 1GbE + 2 x QLogic FastLinQ QL41000 10/25/40/50GbEUbuntu 20.045.4.0-18-generic (x86_64)GNOME Shell 3.36.0X Server 1.20.7modesetting 1.20.7GCC 9.3.01024x7682 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads)12 x 32 GB DDR4-2933MT/s HMA84GR7CJR4N-WMIntel Xeon Platinum 8375C @ 3.50GHz (32 Cores / 64 Threads)Supermicro X12DPi-N6 v1.00 (1.0a BIOS)Intel Device 09988 x 16384 MB DDR4-3200MT/s Samsung M393A2K43CB2-CWE600GB INTEL SSDSC2BB60 + 10001GB Seagate ST10000NM0016-1T + 256GB TOSHIBA THNSNJ25llvmpipe 126GBCrystal View2 x Intel I350Ubuntu 19.105.3.0-64-generic (x86_64)GNOME Shell 3.34.1X Server3.3 Mesa 19.2.8 (LLVM 9.0 256 bits)GCC 9.2.1 20191008800x6002 x Intel Xeon Platinum 8375C @ 3.50GHz (64 Cores / 128 Threads)16 x 16384 MB DDR4-3200MT/s Samsung M393A2K43CB2-CWEllvmpipe 251GBllvmpipe 252GBIntel Xeon Platinum 8480+ @ 2.00GHz (56 Cores / 112 Threads)Supermicro X13DEI v1.01 (1.1 BIOS)Intel Device 1bce8 x 32 GB 4800MT/s MTC20F2085S1RC48BA19256GB TOSHIBA THNSNJ25llvmpipe 8GBNVIDIA Device 10f82 x Broadcom NetXtreme BCM5720 2-port PCIe5.3.0-18-generic (x86_64)X Server 1.20.5modesetting 1.20.53.3 Mesa 19.2.1 (LLVM 9.0 256 bits)OpenBenchmarking.orgCompiler Details- EPYC 2 7R32: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- Xeon W-3275M: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- Xeon Platinum 8259CL: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8259CL NUMA off: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8259CL: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- Xeon Platinum 8272CL: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8272CL NUMA off: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8272CL: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- Xeon Platinum 8280: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8280: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- Xeon Platinum 8375C: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8375C NUMA off: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- 2 x Xeon Platinum 8375C: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v- Xeon Platinum 8480+: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Details- EPYC 2 7R32: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301034- Xeon W-3275M: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- Xeon Platinum 8259CL: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x4002f01- 2 x Xeon Platinum 8259CL NUMA off: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x4002f01- 2 x Xeon Platinum 8259CL: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x4002f01- Xeon Platinum 8272CL: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01- 2 x Xeon Platinum 8272CL NUMA off: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01- 2 x Xeon Platinum 8272CL: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01- Xeon Platinum 8280: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- 2 x Xeon Platinum 8280: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- Xeon Platinum 8375C: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0xd0001b2- 2 x Xeon Platinum 8375C NUMA off: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0xd0001b2- 2 x Xeon Platinum 8375C: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0xd0001b2- Xeon Platinum 8480+: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x2b000161Java Details- EPYC 2 7R32: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- Xeon W-3275M: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-1ubuntu119.10.1)- Xeon Platinum 8259CL: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8259CL NUMA off: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8259CL: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- Xeon Platinum 8272CL: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8272CL NUMA off: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8272CL: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- Xeon Platinum 8280: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- 2 x Xeon Platinum 8280: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- Xeon Platinum 8375C: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8375C NUMA off: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8375C: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- Xeon Platinum 8480+: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)Python Details- EPYC 2 7R32: Python 2.7.17 + Python 3.7.5- Xeon W-3275M: Python 2.7.17 + Python 3.7.5- Xeon Platinum 8259CL: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8259CL NUMA off: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8259CL: Python 2.7.17 + Python 3.7.5- Xeon Platinum 8272CL: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8272CL NUMA off: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8272CL: Python 2.7.17 + Python 3.7.5- Xeon Platinum 8280: + Python 3.8.2- 2 x Xeon Platinum 8280: + Python 3.8.2- Xeon Platinum 8375C: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8375C NUMA off: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8375C: Python 2.7.17 + Python 3.7.5- Xeon Platinum 8480+: Python 2.7.17 + Python 3.7.5Security Details- EPYC 2 7R32: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected- Xeon W-3275M: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + tsx_async_abort: Mitigation of TSX disabled- Xeon Platinum 8259CL: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 2 x Xeon Platinum 8259CL NUMA off: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 2 x Xeon Platinum 8259CL: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- Xeon Platinum 8272CL: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 2 x Xeon Platinum 8272CL NUMA off: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- 2 x Xeon Platinum 8272CL: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled- Xeon Platinum 8280: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + tsx_async_abort: Mitigation of TSX disabled- 2 x Xeon Platinum 8280: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + tsx_async_abort: Mitigation of TSX disabled- Xeon Platinum 8375C: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected- 2 x Xeon Platinum 8375C NUMA off: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected- 2 x Xeon Platinum 8375C: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected- Xeon Platinum 8480+: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB fillingOpenCL Details- Xeon W-3275M: GPU Compute Cores: 4352

EPYC 2 7R32Xeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+Logarithmic Result OverviewPhoronix Test SuiteArrayFireMKL-DNN DNNLOSPrayPostgreSQL pgbenchNAMDCloverLeafNAS Parallel BenchmarksLuxCoreRenderJohn The RipperEmbreeChaos Group V-RAYIntel MPI BenchmarksStockfishasmFishBlenderHimeno BenchmarkLAMMPS Molecular Dynamics SimulatorIntel Open Image DenoiseTimed LLVM Compilationdav1dDeepSpeechTimed Linux Kernel CompilationScikit-LearnPyPerformanceRenaissanceTimed MrBayes AnalysisPyBenchBasis UniversalPHPBenchZstd CompressionRadiance BenchmarkTungsten RendererMlpack BenchmarkSVT-AV1Timed GDB GNU Debugger CompilationNumpy BenchmarkAppleseedNettleSVT-VP9Build2lzbenchTimed GCC CompilationDaCapo Benchmark

Xeon & EPYC Testsbuild-gcc: Time To Compilemkl-dnn: Convolution Batch conv_all - bf16bf16bf16mkl-dnn: Convolution Batch conv_all - u8s8f32mkl-dnn: Convolution Batch conv_all - f32pgbench: Buffer Test - Normal Load - Read Writemkl-dnn: Deconvolution Batch deconv_all - bf16bf16bf16mkl-dnn: Deconvolution Batch deconv_all - f32pgbench: Buffer Test - Normal Load - Read Onlyblender: Barbershop - CPU-Onlyradiance: Serialbuild-llvm: Time To Compilenumpy: renaissance: Savina Reactors.IOhpcg: mlpack: scikit_linearridgeregressionblender: Pabellon Barcelona - CPU-Onlymkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8f32mlpack: scikit_qdamkl-dnn: Convolution Batch conv_googlenet_v3 - f32asmfish: 1024 Hash Memory, 26 Depthmrbayes: Primate Phylogeny Analysisbuild-gdb: Time To Compileblender: Classroom - CPU-Onlydav1d: Chimera 1080p 10-bitmkl-dnn: Convolution Batch conv_googlenet_v3 - bf16bf16bf16luxcorerender: Rainbow Colors and Prismappleseed: Emilyospray: San Miguel - Path Tracerbuild-linux-kernel: Time To Compiledeepspeech: CPUospray: XFrog Forest - Path Tracerappleseed: Material Testerpyperformance: python_startupospray: San Miguel - SciVismlpack: scikit_icablender: Fishy Cat - CPU-Onlyospray: XFrog Forest - SciVisrenaissance: Apache Spark PageRankradiance: SMP Parallelv-ray: CPUpyperformance: floatbuild2: Time To Compilerenaissance: Apache Spark Bayesluxcorerender: DLSCstockfish: Total Timejohn-the-ripper: MD5appleseed: Disney Materialpyperformance: regex_compilebasis: ETC1Sblender: BMW27 - CPU-Onlymkl-dnn: Convolution Batch conv_3d - u8s8f32himeno: Poisson Pressure Solvernamd: ATPase Simulation - 327,506 Atomsmkl-dnn: IP Batch All - u8s8f32mkl-dnn: IP Batch All - f32npb: EP.Dospray: NASA Streamlines - Path Tracernpb: BT.Csvt-av1: Enc Mode 4 - 1080ppyperformance: django_templatemkl-dnn: Convolution Batch conv_3d - bf16bf16bf16embree: Pathtracer - Asian Dragon Objjohn-the-ripper: Blowfishembree: Pathtracer ISPC - Asian Dragon Objtungsten: Water Causticmkl-dnn: IP Batch 1D - f32compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19lzbench: Crush 0 - Decompressionlzbench: Crush 0 - Compressionintel-mpi: IMB-P2P PingPongrenaissance: Scala Dottynpb: LU.Cbasis: UASTC Level 3lzbench: Zstd 8 - Decompressionlzbench: Zstd 8 - Compressionpybench: Total For Average Test Timeslzbench: Brotli 2 - Decompressionlzbench: Brotli 2 - Compressionlzbench: Zstd 1 - Decompressionlzbench: Zstd 1 - Compressionmlpack: scikit_svmmkl-dnn: Deconvolution Batch deconv_1d - u8s8f32mkl-dnn: Deconvolution Batch deconv_1d - bf16bf16bf16ospray: NASA Streamlines - SciVismkl-dnn: Deconvolution Batch deconv_1d - f32mkl-dnn: Convolution Batch conv_alexnet - bf16bf16bf16arrayfire: BLAS CPUsvt-av1: Enc Mode 8 - 1080pembree: Pathtracer - Crownmkl-dnn: Convolution Batch conv_alexnet - u8s8f32dav1d: Summer Nature 4Kembree: Pathtracer ISPC - Crownmkl-dnn: Convolution Batch conv_alexnet - f32npb: SP.Bembree: Pathtracer - Asian Dragonphpbench: PHP Benchmark Suitemkl-dnn: IP Batch 1D - bf16bf16bf16dav1d: Summer Nature 1080pmkl-dnn: Deconvolution Batch deconv_3d - u8s8f32embree: Pathtracer ISPC - Asian Dragonscikit-learn: svt-vp9: Visual Quality Optimized - Bosphorus 1080parrayfire: Conjugate Gradient CPUospray: Magnetic Reconnection - SciVissmhasher: wyhashsmhasher: wyhashtungsten: Hairdacapobench: Jythonlammps: Rhodopsin Proteinoidn: Memorialnpb: MG.Cmkl-dnn: Deconvolution Batch deconv_3d - bf16bf16bf16npb: EP.Cnettle: aes256nettle: sha512mkl-dnn: Deconvolution Batch deconv_3d - f32cloverleaf: Lagrangian-Eulerian Hydrodynamicsnettle: chachanettle: poly1305-aesEPYC 2 7R32Xeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+849.19211728.8710.1406105.7289571485.47891303.121943177.79730.7200.039323.5819303.56117.06081.79132.04730.30639.5442.541311327897792.361108.675101.34140.148.12147.6858564.7324.00572.815995.85175.19370816.358.8249.4765.6711.0321641.954219.6526289759.9563057.0597.28116765028397200062.43104721354.31041.763076.763769.7236640.48614149.3398.883903570.4616.60106415.535.71466.447.58066647545.433022.32471.675609.2303808120348261.0289856577.34399821.8919.1931266831226564165114841815.161131.1376.922.22540990.79854.47247.64641301.36321.1745.186996.688151225.1852.9532500524676.161899.2452.354110.419291.1735.13407.67455524620.82128.7149330.153485.994665.33478.092.204710.36750.3572046.35994.5833880.442967.92878.39011704.2318043293.941292.62567696.505912304.79665.7281.678367.5625186.66614.16231.99265.4113.450432.6549.874466225641143.348134.824213.3681.60188.6113.79196.9297863.2536.367123.852523.17124.14730117.734.4854.91118.095.7221015.543219.8363301494.566.4152743.6233.53611706513654000106.64339215747.65374.966183.113866.0909880.814343.592187.544393210.528.6969347.586.38949.916.590627.19524032630.345821.55172.867158.38050910920246604.6231886970.90469606.8430.634129884951721196129950815.780.4584227.6159545.451.74515702.0653128.3953.81924.550331.9893256.6026.4010108.56831426.5330.10577343395.56304537.843852.3735.14959.661282.184.4234026298.7718.50812.7104432114.56227.3535376.118.462733025.117070.60550.932.275261.141172.1533207.981067.6414364.003402.07874.2015941.3181413368.211029.24512031.419231350.47842.099317.610321.1326711.17313.64252.00305.2012.705931.9848.064059350794138.846132.228246.8167.33207.9113.30219.31973.0041.137117.356922.96127.99629611.432.9761.20134.225.3822226.845273.1252906112276.1473017.3253.04544371173384000120.16934620057.39986.897095.623152.5014370.869253.432367.890712677.648.0951975.355.11264.117.321723.71923559928.292825.35552.699779.6724049116759832.260877200.75265522.4635.9831089711250587160106141018.670.4156147.9829243.481.63928796.7532950.1646.87321.382930.8800205.6324.7267107.80424288.3426.06385697935.69900353.784426.0632.908812.088256.244.67737.0420221.2524.18714.8046501213.04027.9933490.039.371072641.325517.02436.732.200531.11920.5132492.481073.7622198.181734.20461.9353168.7320811904.45743.695790827.572464211.37841.323244.940293.5328251.82816.51712.42157.997.6365141.2925.0672105070606141.898146.435124.6788.66106.7145.52163.7072325.5429.276135.051005.71197.29377422.661.5857.6979.2210.3225636.694255.5885305112369.1953813.7465.86104106676660833373.21622420058.28048.003548.603130.8587130.454222.781936.658425284.3415.5484766.145.99664.19.1875142.49467095550.864824.77921.5372910.2943968632444647.4347837566.34294642.9923.147966711253543159103240719.620.3614144.4652376.921.22589399.3085650.7152.31438.103319.3723212.9946.418455.356538493.5746.90845743484.46440400.242252.6759.050212.954229.537.60462.520231.5824.1549.09073516121.88733.1043490.154.856895059.025523.31438.231.263950.63917.5052516.871012.0352197.811793.75559.4684009.4053311904.28739.163795715.311355211.33841.647232.053310.4028107.03627.10602.30158.138.7479442.8825.7722119912357139.750129.689124.5989.55106.8936.40161.8463145.5827.581117.726135.71185.22973411.362.558.4879.0010.4225275.159258.3435180312366.6673495.6125.88103513853655466773.58191320057.82647.633551.383147.9075210.445513.048669.976695310.1615.625104107.995.95564.29.2066542.76137088850.976924.54371.576879.9493948831988149.2463777256.920120961.9123.0881011691252551157102739918.760.3762224.4903976.921.25795399.2985319.5853.04038.122222.3830215.6446.175756.646553978.9046.79205734834.48918431.762238.0559.056312.459229.918.43062.520225.1824.1519.26217517222.53533.8963393.354.871945052.635513.08437.621.266490.78917.7502497.861107.3663639.273089.34821.9674087.3028972751.001123.68546028.338296310.51796.578301.531340.7223866.7401.91271.6812.028831.5145.127164230868133.483142.501215.9163.01173.3813.92211.9412513.3638.734117.866443.33129.030611.637.0458.57119.366.0221895.643257.2323394774.2652739.0023.63581721073727667117.15179854.19475.866437.603318.9791660.777243.344277.407043033.289.0969011.355.65360.614.074727.66853960431.567524.21272.386068.3074279719096342.6521746900.19332.1991145751184619168112243017.830.3782296.5977847.621.47084662.7553281.5851.59125.121827.9779208.5027.784498.666429919.0130.56536048575.59680315.513985.1036.752411.212256.024.1964020556.9623.67613.8818478614.01430.2234708.608.316772974.155830.30462.782.165811.05970.3832661.611110.2041842.971579.11436.8604921.0770961833.27877.207873066.757012195.61797.156230.854313.3528365.56816.52822.13142.816.8254040.3523.6729107388344134.883159.498109.3482.6590.26366.79161.3768146.2128.129140.439496.41195.11148625.266.6755.9070.2211.6324909.655241.0636217511765.7723859.3796.95109753377710266772.84842619055.45641.223243.003315.2954710.401733.105306.386525973.1817.44102113.056.45560.87.6556149.10617842056.459523.82071.6310610.0884129336872935.1449287347.07921.1201022741182571168107843019.050.3364263.9506790.911.30395332.3196330.9353.96143.989418.3929211.0550.946051.442946295.7954.19646064362.85310371.482063.2365.545212.055227.908.94771.4320554.9223.6718.34744503722.66633.5745188.234.494645667.765852.82465.401.252940.61975.0812657.681068.4721844.781640.45534.3282150.9176521838.60881.087857060.935930195.96796.433216.886324.3927891.30126.70252.16143.317.7970840.6424.1147122933930134.997143.645109.9182.8790.39827.44162.4956156.2526.583123.814726.44186.30603311.866.6754.8470.1211.6323916.230246.1596137811768.5583402.0466.99112362723715000073.50382818955.27841.333271.403331.3774030.388403.296159.713556088.0817.54131794.236.39960.67.6751749.28157830156.506723.90771.606419.6534159336807805.9275366956.90121.4041058731182580166108842317.940.3517193.9839390.911.34411337.8833972.3254.13543.823820.0909216.0051.198252.997966865.4554.20416065482.89086391.252064.0265.367911.763227.897.46771.4320553.0523.6458.66791501323.43435.5266470.564.494055763.375852.22465.181.251270.71974.4582654.231063.7053497.472846.12765.74553139.0577772935.991113.54560887.766696287.80728.361308.741330.3321975.12214.90971.67246.3111.290030.4542.075868762075128.848144.643198.2565.40166.9844.11197.6495053.5240.898118.093773.46123.97910311.238.4667.36110.676.2520621.2292353471010974.1182674.1753.79628913124090000103.83309916451.44669.805919.943510.2051590.707093.231216.795093283.389.4971842.346.03046.214.078029.20614286433.049022.01672.394518.20344010116653864.7681166664.25474255.6730.6891224771076673181121245717.190.3710366.55374501.46232637.3743432.1651.91726.280026.8391199.0828.773094.193732682.1432.24746491644.83170323.203716.9838.31009.127265.404.11542.8722.67812.6863454715.31431.1037089.057.696053194.186313.16501.981.962790.961051.1112867.481005.5301766.351517.44496.81624466.9112271779.15881.783909916.490086187.33728.246221.515321.8226764.94128.85421.82129.017.4605038.3523.1772133839896129.986142.695100.4785.1486.53917.15154.9807526.5327.309122.224686.67193.4424111.269.5263.3566.3912.0523588.397226.0766433110962.7763228.6337.19119715795788000068.61827216452.69438.793005.483514.6041060.362263.110029.524666284.6518.18136046.537.03345.97.9291251.78538408458.694521.58961.639869.1374289631821690.5507256890.606140090.3219.8851124751073623176116844917.190.3944554.0684790.911.24029320.0603430.2856.98245.612119.8143221.3153.230449.843770893.4157.48706496933.65964409.881893.7268.42419.392233.177.48171.4322.6777.82706471124.78834.4969175.124.358176104.946320.00500.981.136080.691051.9362868.691006.6352712.072125.83598.1873870.4330162063.42724.602751908.533381218.42654.183215.837374.9021743.32220.16872.54177.577.3503826.5130.82799389689999.688139.362136.5596.93129.5956.02156.2159355.1629.357103.080175.1890.81975219.854.3945.5081.649.3819555.057201.2244476013460.3263492.9915.5885313242571433378.5794918648.22752.284454.224214.9440790.469481.845154.595504730.2614.2296192.054.09865.410.750840.18465632646.143723.42061.085537.28342211124658155.9565228999.76897414.0726.597141289978624184132146417.050.3010074.9788871.431.24001500.1204875.4741.84237.633917.7241247.8642.597583.360946438.5844.30557121402.41906444.302530.9653.57209.566266.292.81462.59.21365511215.90136.5650840.766.071174003.828052.90396.691.340960.51822.1081909.601024.0841381.171107.34327.2764148.6882941166.69458.494735555.970405158.72654.575178.751316.7529523.64527.20722.7599.214.7157429.5317.3496148521449103.112149.86374.20118.8867.968810.91213.5059629.5222.410113.287729.88179.24090620.499.2444.0353.6917.8622022.805198.5018793713456.0145550.05510.091531134921071433364.97065419148.92131.092382.114198.5146910.256321.498653.730319334.3127.03164568.745.20465.35.7196370.054311184782.316423.64320.7652798.42241610949151696.0579719170.215152798.5617.341129188979617183127146117.540.3063092.91495136.901.03115251.8264631.3867.08469.139210.9680246.9678.553542.910386029.3278.53797044531.74213478.051295.4895.98579.472243.244.036111.116.11790540027.34532.2982390.243.777696744.578007.34396.530.8510070.33815.1791909.43970.9161421.841166.32399.2403301.6799581211.21469.4281063447.923986161.71654.978167.672370.9631368.93539.15622.60100.515.3542629.9221.0065164553317101.872140.61075.73118.3771.006010.90214.9546879.5221.399108.161589.88185.45497719.810044.6853.9617.8622456.612200.88738813356.1554402.82310.151529068601049733364.75075418349.40331.102540.744211.2108420.249411.446136.002649339.0427.03188359.364.83763.25.8124370.388310997881.900923.49410.7317538.17941310650148418.0000009329.431177387.1816.915132086980600178125845317.080.2994733.07797140.481.01890252.4064520.1865.29569.264812.3317251.6378.634243.8561112350.4778.76727085711.73950500.831320.6095.27069.332249.767.355108.336.05668533825.98232.96111939.623.783196352.448044.39456.390.8525180.35821.1771963.03994.918685.1341227.39350.6214605.890292993.580611.295694754.044806189.21530.307163.209427.4636700.70226.51662.26122.995.3038423.0020.045912744252478.872155.56184.0797.5334.12479.98239.25277333.3323.929110.84881129.14724823.450043.1961.4924744.428161.3977273615357.0996203.1588.78130983602913300077.32905614639.58135.272550.216769.2731880.291961.292402.914526288.82116616.824.50278.22.9810651.59269340962.053828.21030.7703066.15046012027196959.43478311351.879114346.4016.269152910069371121414005277.990.2811592.157420.998629129.0325996.4040.64645.466011.6822211.2153.210343.659595294.9756.56778931241.47256403.301445.0771.62396.830205.612.1496.66529525619.76056.0182633.641.740053863.009126.49616.591.021990.46968.5161439.35OpenBenchmarking.org

Timed GCC Compilation

This test times how long it takes to build the GNU Compiler Collection (GCC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 9.3.0Time To CompileEPYC 2 7R322 x Xeon Platinum 8375CXeon W-3275MXeon Platinum 8480+2 x Xeon Platinum 8280Xeon Platinum 8375C2 x Xeon Platinum 8259CL2 x Xeon Platinum 8375C NUMA offXeon Platinum 8280Xeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2004006008001000SE +/- 0.63, N = 3SE +/- 0.28, N = 3SE +/- 0.44, N = 3SE +/- 4.55, N = 3SE +/- 0.24, N = 3SE +/- 1.19, N = 3SE +/- 0.68, N = 3SE +/- 0.07, N = 3SE +/- 0.29, N = 3SE +/- 0.67, N = 3SE +/- 2.75, N = 3SE +/- 1.08, N = 3SE +/- 0.80, N = 3SE +/- 0.66, N = 3849.19970.92994.58994.921005.531006.641012.041024.081063.711067.641068.471073.761107.371110.20

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL9001800270036004500SE +/- 0.65, N = 3SE +/- 1.02, N = 3SE +/- 4.43, N = 3SE +/- 1.13, N = 3SE +/- 1.52, N = 3SE +/- 5.10, N = 3SE +/- 0.11, N = 3SE +/- 0.30, N = 3SE +/- 0.18, N = 3SE +/- 0.26, N = 3SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3685.131381.171421.841766.351842.971844.782197.812198.182712.073497.473639.273880.444364.00MIN: 1361.11MIN: 1364.34MIN: 1757.1MIN: 1827.15MIN: 1827.01MIN: 2191.04MIN: 2191.82MIN: 2707.4MIN: 3490.41MIN: 3627.59MIN: 3767MIN: 4358.261. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: u8s8f322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CLEPYC 2 7R323K6K9K12K15KSE +/- 0.16, N = 3SE +/- 11.98, N = 3SE +/- 0.75, N = 3SE +/- 3.12, N = 3SE +/- 1.89, N = 3SE +/- 0.32, N = 3SE +/- 0.82, N = 3SE +/- 1.51, N = 3SE +/- 1.48, N = 3SE +/- 1.24, N = 3SE +/- 15.10, N = 3SE +/- 9.96, N = 3SE +/- 4.13, N = 3SE +/- 19.01, N = 31107.341166.321227.391517.441579.111640.451734.201793.752125.832846.122967.923089.343402.0711728.80MIN: 2813.78MIN: 2893.12MIN: 3035MIN: 3393.59MIN: 11565.11. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_all - Data Type: f322 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275M2004006008001000SE +/- 0.26, N = 3SE +/- 3.63, N = 3SE +/- 5.03, N = 3SE +/- 0.30, N = 3SE +/- 0.29, N = 3SE +/- 0.78, N = 3SE +/- 0.50, N = 3SE +/- 0.29, N = 3SE +/- 0.29, N = 3SE +/- 2.05, N = 3SE +/- 0.39, N = 3SE +/- 0.28, N = 3SE +/- 0.32, N = 3SE +/- 0.68, N = 3327.28350.62399.24436.86461.94496.82534.33559.47598.19710.14765.75821.97874.20878.39MIN: 316.03MIN: 342.71MIN: 367.6MIN: 423.96MIN: 457.45MIN: 488.36MIN: 520.78MIN: 553.8MIN: 588.29MIN: 700.52MIN: 755.35MIN: 794.87MIN: 868.41MIN: 819.941. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

PostgreSQL pgbench

This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteXeon Platinum 82802 x Xeon Platinum 8280Xeon W-3275MEPYC 2 7R32Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8480+2 x Xeon Platinum 8375C NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL11K22K33K44K55KSE +/- 370.72, N = 3SE +/- 19.82, N = 3SE +/- 85.02, N = 3SE +/- 319.58, N = 12SE +/- 245.15, N = 12SE +/- 204.93, N = 12SE +/- 207.67, N = 9SE +/- 312.08, N = 9SE +/- 151.64, N = 9SE +/- 56.30, N = 3SE +/- 156.69, N = 9SE +/- 61.49, N = 9SE +/- 408.27, N = 6SE +/- 72.69, N = 653139.0624466.9111704.236105.735941.324921.084605.894148.694087.304009.413870.433301.683168.732150.921. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_all - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8272CLXeon Platinum 8280Xeon W-3275MXeon Platinum 8259CL7001400210028003500SE +/- 0.72, N = 3SE +/- 2.42, N = 3SE +/- 1.11, N = 3SE +/- 0.87, N = 3SE +/- 3.61, N = 3SE +/- 1.93, N = 3SE +/- 0.13, N = 3SE +/- 0.44, N = 3SE +/- 7.07, N = 3SE +/- 1.11, N = 3SE +/- 5.43, N = 3SE +/- 4.61, N = 3SE +/- 0.16, N = 3993.581166.691211.211779.151833.271838.601904.281904.452063.422751.002935.993293.943368.21MIN: 979.7MIN: 1113.61MIN: 1117.5MIN: 1772.45MIN: 1806.74MIN: 1808.61MIN: 1898.87MIN: 1899.24MIN: 2042.76MIN: 2720.09MIN: 2926MIN: 3137.48MIN: 3365.011. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_all - Data Type: f322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8280Xeon Platinum 8259CLXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MEPYC 2 7R3230060090012001500SE +/- 0.19, N = 3SE +/- 1.02, N = 3SE +/- 0.39, N = 3SE +/- 0.11, N = 3SE +/- 0.85, N = 3SE +/- 0.88, N = 3SE +/- 0.90, N = 3SE +/- 10.37, N = 3SE +/- 0.39, N = 3SE +/- 0.97, N = 3SE +/- 0.33, N = 3SE +/- 0.87, N = 3SE +/- 16.26, N = 5SE +/- 1.29, N = 3458.49469.43611.30724.60739.16743.70877.21881.09881.781029.241113.541123.681292.621485.47MIN: 452.45MIN: 455.81MIN: 603.09MIN: 719.66MIN: 731.99MIN: 737.27MIN: 867.81MIN: 860.45MIN: 874.01MIN: 1023.45MIN: 1108.65MIN: 1108.94MIN: 1177.9MIN: 1462.151. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

PostgreSQL pgbench

This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read Only2 x Xeon Platinum 8375C2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+Xeon W-3275MXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CL200K400K600K800K1000KSE +/- 83725.25, N = 9SE +/- 10457.35, N = 3SE +/- 953.22, N = 3SE +/- 12996.71, N = 4SE +/- 2758.93, N = 3SE +/- 2473.02, N = 3SE +/- 1194.42, N = 3SE +/- 3720.11, N = 3SE +/- 3241.20, N = 3SE +/- 10308.26, N = 4SE +/- 3571.60, N = 3SE +/- 1625.61, N = 3SE +/- 2217.17, N = 3SE +/- 194.45, N = 31063447.92909916.49891303.12873066.76857060.94795715.31790827.57751908.53735555.97694754.04567696.51560887.77546028.34512031.421. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: Barbershop - Compute: CPU-Only2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CEPYC 2 7R322 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL80160240320400SE +/- 0.37, N = 3SE +/- 0.83, N = 3SE +/- 0.49, N = 3SE +/- 0.40, N = 3SE +/- 0.19, N = 3SE +/- 0.14, N = 3SE +/- 0.56, N = 3SE +/- 0.16, N = 3SE +/- 0.10, N = 3SE +/- 0.41, N = 3SE +/- 0.20, N = 3SE +/- 0.22, N = 3SE +/- 0.21, N = 3SE +/- 0.10, N = 3158.72161.71177.79187.33189.21195.61195.96211.33211.37218.42287.80304.79310.51350.47

Radiance Benchmark

This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRadiance Benchmark 5.0Test: SerialXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CLXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8259CL2004006008001000530.31654.18654.58654.98665.70728.25728.36730.70796.43796.58797.16841.32841.65842.10

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offEPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon W-3275MXeon Platinum 8272CLXeon Platinum 8280Xeon Platinum 8259CL70140210280350SE +/- 0.88, N = 3SE +/- 1.19, N = 3SE +/- 0.59, N = 3SE +/- 0.72, N = 3SE +/- 2.77, N = 3SE +/- 0.90, N = 3SE +/- 1.01, N = 3SE +/- 0.97, N = 3SE +/- 1.13, N = 3SE +/- 1.40, N = 3SE +/- 2.68, N = 3SE +/- 1.00, N = 3SE +/- 4.79, N = 3SE +/- 5.07, N = 3163.21167.67178.75200.04215.84216.89221.52230.85232.05244.94281.68301.53308.74317.61

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375CXeon W-3275MXeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 8272CLEPYC 2 7R322 x Xeon Platinum 8280Xeon Platinum 8259CL2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off90180270360450SE +/- 3.64, N = 3SE +/- 1.61, N = 3SE +/- 0.82, N = 3SE +/- 2.73, N = 3SE +/- 0.38, N = 3SE +/- 2.42, N = 3SE +/- 1.12, N = 3SE +/- 1.14, N = 3SE +/- 3.00, N = 3SE +/- 0.75, N = 3SE +/- 4.12, N = 3SE +/- 4.33, N = 3SE +/- 0.76, N = 3SE +/- 3.32, N = 3427.46374.90370.96367.56340.72330.33324.39323.58321.82321.13316.75313.35310.40293.53

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.10.0Test: Savina Reactors.IOEPYC 2 7R32Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+8K16K24K32K40KSE +/- 365.25, N = 20SE +/- 414.26, N = 25SE +/- 593.14, N = 25SE +/- 642.38, N = 25SE +/- 599.32, N = 25SE +/- 428.64, N = 25SE +/- 474.27, N = 25SE +/- 484.84, N = 25SE +/- 482.43, N = 25SE +/- 502.20, N = 25SE +/- 618.01, N = 25SE +/- 472.37, N = 25SE +/- 695.27, N = 20SE +/- 383.97, N = 719303.5621743.3221975.1223866.7425186.6726711.1726764.9427891.3028107.0428251.8328365.5729523.6531368.9436700.70

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.12 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CLXeon Platinum 8480+Xeon Platinum 8375CEPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon W-3275MXeon Platinum 8259CL918273645SE +/- 0.19, N = 3SE +/- 0.25, N = 3SE +/- 0.01, N = 3SE +/- 0.13, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 3SE +/- 0.03, N = 339.1628.8527.2127.1126.7026.5220.1717.0616.5316.5214.9114.1613.641. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionXeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8480+2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off0.61881.23761.85642.47523.094SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 15SE +/- 0.02, N = 9SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 15SE +/- 0.02, N = 12SE +/- 0.01, N = 3SE +/- 0.02, N = 10SE +/- 0.01, N = 31.671.791.821.911.992.002.132.162.262.302.422.542.602.75

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: Pabellon Barcelona - Compute: CPU-Only2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL70140210280350SE +/- 0.80, N = 3SE +/- 0.71, N = 3SE +/- 0.27, N = 3SE +/- 0.26, N = 3SE +/- 0.51, N = 3SE +/- 0.25, N = 3SE +/- 0.50, N = 3SE +/- 0.13, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.73, N = 3SE +/- 0.31, N = 3SE +/- 0.07, N = 399.21100.51122.99129.01132.04142.81143.31157.99158.13177.57246.31265.41271.68305.20

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f322 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275MEPYC 2 7R32160320480640800SE +/- 0.02282, N = 3SE +/- 0.00176, N = 3SE +/- 0.04742, N = 12SE +/- 0.00507, N = 3SE +/- 0.00678, N = 3SE +/- 0.03597, N = 3SE +/- 0.01740, N = 3SE +/- 0.04094, N = 3SE +/- 0.02555, N = 3SE +/- 0.01386, N = 3SE +/- 0.02875, N = 3SE +/- 0.00834, N = 3SE +/- 0.01662, N = 3SE +/- 7.24186, N = 34.715745.303845.354266.825407.350387.460507.636517.797088.7479411.2900012.0288012.7059013.45040730.30600MIN: 698.321. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275M2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL1020304050SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.40, N = 3SE +/- 0.31, N = 8SE +/- 0.29, N = 3SE +/- 0.20, N = 3SE +/- 0.12, N = 3SE +/- 0.18, N = 3SE +/- 0.34, N = 3SE +/- 0.28, N = 3SE +/- 0.60, N = 3SE +/- 0.50, N = 5SE +/- 0.19, N = 3SE +/- 0.54, N = 1223.0026.5129.5329.9230.4531.5131.9832.6538.3539.5440.3540.6441.2942.88

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: f322 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280EPYC 2 7R32Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275M1122334455SE +/- 0.08, N = 3SE +/- 0.01, N = 3SE +/- 0.18, N = 12SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 317.3520.0521.0123.1823.6724.1125.0725.7730.8342.0842.5445.1348.0649.87MIN: 16.52MIN: 19.6MIN: 16.59MIN: 22MIN: 22.38MIN: 22.8MIN: 24.48MIN: 24.46MIN: 29.83MIN: 41.31MIN: 40.72MIN: 43.29MIN: 47.62MIN: 44.781. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Depth2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLEPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL40M80M120M160M200MSE +/- 721194.49, N = 3SE +/- 441971.10, N = 3SE +/- 491864.96, N = 3SE +/- 197662.62, N = 3SE +/- 499401.71, N = 3SE +/- 1707658.59, N = 3SE +/- 617926.69, N = 3SE +/- 526881.33, N = 3SE +/- 1538634.72, N = 3SE +/- 597319.54, N = 3SE +/- 450047.19, N = 3SE +/- 845752.34, N = 4SE +/- 950786.26, N = 3SE +/- 95127.60, N = 31645533171485214491338398961274425241229339301199123571132789771073883441050706069389689968762075662256416423086859350794

Timed MrBayes Analysis

This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisXeon Platinum 8480+EPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon W-3275M306090120150SE +/- 0.52, N = 3SE +/- 0.68, N = 3SE +/- 0.43, N = 3SE +/- 0.68, N = 3SE +/- 0.96, N = 12SE +/- 0.01, N = 3SE +/- 0.37, N = 3SE +/- 0.09, N = 3SE +/- 0.45, N = 3SE +/- 0.99, N = 3SE +/- 0.18, N = 3SE +/- 0.32, N = 3SE +/- 1.87, N = 3SE +/- 0.13, N = 378.8792.3699.69101.87103.11128.85129.99133.48134.88135.00138.85139.75141.90143.35-msha -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi-msse4a -msha-msha -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi-msha -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi-msha -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx-mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mmpx -lreadline1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm

Timed GDB GNU Debugger Compilation

This test times how long it takes to build the GNU Debugger (GDB) in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileEPYC 2 7R322 x Xeon Platinum 8259CLXeon Platinum 8259CLXeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375CXeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 8272CL NUMA off4080120160200SE +/- 0.08, N = 3SE +/- 0.45, N = 3SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.22, N = 3SE +/- 0.21, N = 3SE +/- 0.11, N = 3SE +/- 0.17, N = 3SE +/- 0.52, N = 3SE +/- 0.17, N = 3SE +/- 0.27, N = 3SE +/- 0.09, N = 3SE +/- 0.36, N = 3SE +/- 0.18, N = 3108.68129.69132.23134.82139.36140.61142.50142.70143.65144.64146.44149.86155.56159.50

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: Classroom - Compute: CPU-Only2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL50100150200250SE +/- 0.90, N = 3SE +/- 0.37, N = 3SE +/- 0.06, N = 3SE +/- 0.26, N = 3SE +/- 0.40, N = 3SE +/- 0.33, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 3SE +/- 0.36, N = 3SE +/- 0.32, N = 3SE +/- 0.37, N = 3SE +/- 0.14, N = 374.2075.7384.07100.47101.34109.34109.91124.59124.67136.55198.25213.36215.91246.81

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.6.0Video Input: Chimera 1080p 10-bitEPYC 2 7R322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon W-3275MXeon Platinum 8259CLXeon Platinum 8280Xeon Platinum 8272CL306090120150SE +/- 0.60, N = 3SE +/- 0.57, N = 3SE +/- 0.19, N = 3SE +/- 0.06, N = 3SE +/- 0.62, N = 3SE +/- 0.04, N = 3SE +/- 0.56, N = 3SE +/- 0.07, N = 3SE +/- 0.45, N = 3SE +/- 0.27, N = 3SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.18, N = 3140.14118.88118.3797.5396.9389.5588.6685.1482.8782.6581.6067.3365.4063.01MIN: 101.2 / MAX: 212.94MIN: 87.23 / MAX: 179.18MIN: 86.83 / MAX: 179.15MIN: 72.28 / MAX: 140.3MIN: 68.44 / MAX: 150.56MIN: 69.22 / MAX: 130.8MIN: 69.45 / MAX: 129.53MIN: 66.74 / MAX: 126.85MIN: 65.6 / MAX: 121.56MIN: 65.81 / MAX: 118.98MIN: 55.07 / MAX: 147.52MIN: 49.41 / MAX: 122.02MIN: 49.2 / MAX: 111MIN: 47.23 / MAX: 110.251. (CC) gcc options: -pthread

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_googlenet_v3 - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL50100150200250SE +/- 0.03, N = 3SE +/- 0.29, N = 3SE +/- 0.25, N = 3SE +/- 0.00, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.20, N = 3SE +/- 0.02, N = 334.1267.9771.0186.5490.2690.40106.71106.89129.60166.98173.38188.61207.91MIN: 33.66MIN: 65.18MIN: 65.42MIN: 85.56MIN: 88.8MIN: 88.8MIN: 105.9MIN: 105.9MIN: 129.23MIN: 166.21MIN: 171.91MIN: 178.56MIN: 207.271. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

LuxCoreRender

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on the CPU as opposed to the OpenCL version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: Rainbow Colors and Prism2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL3691215SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.40, N = 15SE +/- 0.33, N = 12SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.37, N = 15SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 15SE +/- 0.00, N = 310.9110.909.988.127.447.156.796.406.025.524.113.923.793.30MIN: 8.72 / MAX: 11.1MIN: 9.55 / MAX: 11.14MIN: 6.87 / MAX: 10.41MIN: 7.87 / MAX: 8.3MIN: 6.75 / MAX: 7.63MIN: 2.92 / MAX: 8.11MIN: 3.95 / MAX: 7.68MIN: 5.65 / MAX: 6.54MIN: 5.57 / MAX: 6.15MIN: 2.11 / MAX: 6.54MIN: 3.96 / MAX: 4.16MIN: 3.78 / MAX: 3.98MIN: 3.45 / MAX: 3.89MIN: 3.24 / MAX: 3.33

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: EmilyEPYC 2 7R322 x Xeon Platinum 8280Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon W-3275MXeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8259CLXeon Platinum 8480+50100150200250147.69154.98156.22161.38161.85162.50163.71196.93197.65211.94213.51214.95219.32239.25

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL70140210280350SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.01, N = 5SE +/- 0.00, N = 3SE +/- 0.00, N = 12SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 5SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 12SE +/- 0.00, N = 3333.339.529.526.536.256.215.585.545.164.733.523.363.253.00MIN: 250 / MAX: 500

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To Compile2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon W-3275MXeon Platinum 8272CLXeon Platinum 8280Xeon Platinum 8259CL918273645SE +/- 0.30, N = 13SE +/- 0.32, N = 13SE +/- 0.55, N = 15SE +/- 0.19, N = 13SE +/- 0.29, N = 13SE +/- 0.25, N = 14SE +/- 0.23, N = 14SE +/- 0.28, N = 14SE +/- 0.37, N = 14SE +/- 0.35, N = 13SE +/- 0.35, N = 9SE +/- 0.40, N = 13SE +/- 0.42, N = 8SE +/- 0.44, N = 721.4022.4123.9324.0126.5827.3127.5828.1329.2829.3636.3738.7340.9041.14

DeepSpeech

Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDeepSpeech 0.6Acceleration: CPUEPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8375C NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CLXeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 82802 x Xeon Platinum 8272CLXeon W-3275M2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL NUMA off306090120150SE +/- 0.08, N = 3SE +/- 0.33, N = 3SE +/- 0.95, N = 3SE +/- 1.55, N = 4SE +/- 0.33, N = 3SE +/- 1.18, N = 8SE +/- 0.65, N = 3SE +/- 1.58, N = 4SE +/- 1.60, N = 3SE +/- 1.16, N = 3SE +/- 0.58, N = 3SE +/- 0.15, N = 3SE +/- 1.04, N = 3SE +/- 1.87, N = 372.82103.08108.16110.85113.29117.36117.73117.87118.09122.22123.81123.85135.05140.44

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path Tracer2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL3691215SE +/- 0.02, N = 5SE +/- 0.02, N = 4SE +/- 0.00, N = 12SE +/- 0.01, N = 3SE +/- 0.00, N = 15SE +/- 0.02, N = 3SE +/- 0.00, N = 12SE +/- 0.01, N = 6SE +/- 0.00, N = 12SE +/- 0.00, N = 4SE +/- 0.00, N = 5SE +/- 0.00, N = 7SE +/- 0.00, N = 69.889.886.676.446.415.855.715.715.183.463.333.172.96MIN: 7.81 / MAX: 10MIN: 9.71 / MAX: 9.9MIN: 5.85 / MAX: 6.76MIN: 5.92 / MAX: 6.49MIN: 6.21 / MAX: 6.45MIN: 5.65 / MAX: 5.99MIN: 5.08 / MAX: 5.78MIN: 5.59 / MAX: 5.75MIN: 5.13 / MAX: 5.21MIN: 3.41 / MAX: 3.47MIN: 3.3 / MAX: 3.36MIN: 3.09 / MAX: 3.19MIN: 2.92 / MAX: 2.99

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Material TesterXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8259CLXeon Platinum 8272CLXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off408012016020090.82123.98124.15128.00129.03129.15175.19179.24185.23185.45186.31193.44195.11197.29

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: python_startupXeon Platinum 82802 x Xeon Platinum 82802 x Xeon Platinum 8259CLXeon Platinum 8259CLXeon Platinum 8272CL2 x Xeon Platinum 8272CLEPYC 2 7R32Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8480+2 x Xeon Platinum 8272CL NUMA off612182430SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 311.211.211.311.411.611.816.317.719.819.820.422.623.425.2

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offEPYC 2 7R32Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL110220330440550SE +/- 0.76, N = 12SE +/- 0.62, N = 15SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 3SE +/- 0.92, N = 4SE +/- 0.00, N = 12SE +/- 0.72, N = 5SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.36, N = 3500.00100.0099.2469.5266.6766.6762.5061.5858.8254.3938.4637.0434.4832.97MIN: 333.33MIN: 55.56MIN: 71.43 / MAX: 100

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaXeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8375CEPYC 2 7R322 x Xeon Platinum 8272CLXeon W-3275M2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8272CLXeon Platinum 8259CL2 x Xeon Platinum 8280Xeon Platinum 82801530456075SE +/- 0.30, N = 15SE +/- 0.34, N = 3SE +/- 0.17, N = 3SE +/- 0.12, N = 3SE +/- 0.48, N = 15SE +/- 0.40, N = 3SE +/- 0.11, N = 3SE +/- 0.52, N = 3SE +/- 0.35, N = 3SE +/- 0.80, N = 3SE +/- 0.06, N = 3SE +/- 0.34, N = 3SE +/- 0.59, N = 3SE +/- 0.21, N = 343.1944.0344.6845.5049.4754.8454.9155.9057.6958.4858.5761.2063.3567.36

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: Fishy Cat - Compute: CPU-Only2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL306090120150SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.23, N = 3SE +/- 0.19, N = 3SE +/- 0.11, N = 3SE +/- 0.14, N = 3SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 0.05, N = 3SE +/- 0.14, N = 3SE +/- 0.08, N = 353.6953.9661.4965.6766.3970.1270.2279.0079.2281.64110.67118.09119.36134.22

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVis2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL48121620SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.04, N = 3SE +/- 0.00, N = 12SE +/- 0.02, N = 7SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 12SE +/- 0.00, N = 8SE +/- 0.00, N = 1217.8617.8612.0511.6311.6311.0310.4210.329.386.256.025.725.38MIN: 12.5 / MAX: 18.18MIN: 16.95 / MAX: 18.18MIN: 9.71 / MAX: 12.2MIN: 9.8 / MAX: 11.9MIN: 11.11 / MAX: 11.76MIN: 10.75 / MAX: 11.24MIN: 8.2 / MAX: 10.53MIN: 10.1 / MAX: 10.42MIN: 9.26 / MAX: 9.43MIN: 6.13 / MAX: 6.29MIN: 5.95 / MAX: 6.1MIN: 5.56 / MAX: 5.78MIN: 5.29 / MAX: 5.41

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.10.0Test: Apache Spark PageRankXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MEPYC 2 7R32Xeon Platinum 8272CL2 x Xeon Platinum 8375C NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CLXeon Platinum 8480+2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off5K10K15K20K25KSE +/- 173.94, N = 5SE +/- 169.44, N = 5SE +/- 165.27, N = 5SE +/- 198.48, N = 5SE +/- 170.65, N = 5SE +/- 263.18, N = 5SE +/- 154.16, N = 5SE +/- 172.53, N = 25SE +/- 286.23, N = 6SE +/- 167.92, N = 5SE +/- 260.73, N = 25SE +/- 205.74, N = 5SE +/- 278.82, N = 5SE +/- 232.21, N = 519555.0620621.2321015.5421641.9521895.6422022.8122226.8522456.6123588.4023916.2324744.4324909.6625275.1625636.69

Radiance Benchmark

This is a benchmark of NREL Radiance, a synthetic imaging system that is open-source and developed by the Lawrence Berkeley National Laboratory in California. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRadiance Benchmark 5.0Test: SMP ParallelXeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8375CEPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8259CL60120180240300161.40198.50200.80201.22219.65219.84226.08235.00241.06246.16255.59257.23258.34273.13

Chaos Group V-RAY

This is a test of Chaos Group's V-RAY benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgKsamples, More Is BetterChaos Group V-RAY 4.10.07Mode: CPU2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL20K40K60K80K100KSE +/- 778.30, N = 3SE +/- 771.09, N = 3SE +/- 291.10, N = 3SE +/- 474.70, N = 3SE +/- 183.86, N = 3SE +/- 219.47, N = 3SE +/- 584.43, N = 3SE +/- 48.95, N = 3SE +/- 332.60, N = 3SE +/- 481.50, N = 7SE +/- 306.00, N = 3SE +/- 15.25, N = 3SE +/- 401.71, N = 3SE +/- 88.49, N = 38793787388727366433162897621756137853051518034476034710339473301429061

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: floatXeon W-3275MXeon Platinum 82802 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8375CXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+306090120150SE +/- 0.12, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 394.5109.0109.0117.0117.0122.0123.0123.0133.0134.0134.0153.0

Build2

This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.12Time To Compile2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+EPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA offXeon W-3275M2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CL20406080100SE +/- 0.14, N = 3SE +/- 0.58, N = 3SE +/- 0.48, N = 3SE +/- 0.40, N = 3SE +/- 0.18, N = 3SE +/- 0.09, N = 3SE +/- 0.62, N = 3SE +/- 0.14, N = 3SE +/- 0.49, N = 3SE +/- 0.46, N = 3SE +/- 0.19, N = 3SE +/- 0.13, N = 3SE +/- 0.42, N = 3SE +/- 0.03, N = 356.0156.1657.1059.9660.3362.7865.7766.4266.6768.5669.2074.1274.2776.15

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.10.0Test: Apache Spark BayesXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CLEPYC 2 7R322 x Xeon Platinum 82802 x Xeon Platinum 8272CLXeon Platinum 8375C2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+13002600390052006500SE +/- 19.41, N = 5SE +/- 18.35, N = 20SE +/- 30.38, N = 7SE +/- 26.07, N = 25SE +/- 26.28, N = 25SE +/- 36.17, N = 25SE +/- 41.01, N = 25SE +/- 39.37, N = 7SE +/- 47.41, N = 25SE +/- 55.66, N = 25SE +/- 77.21, N = 20SE +/- 90.83, N = 25SE +/- 172.31, N = 25SE +/- 93.28, N = 242674.182739.002743.623017.333057.063228.633402.053492.993495.613813.753859.384402.825550.066203.16

LuxCoreRender

LuxCoreRender is an open-source physically based renderer. This test profile is focused on running LuxCoreRender on the CPU as opposed to the OpenCL version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: DLSC2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL3691215SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.13, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 310.1510.098.787.287.196.996.955.885.865.583.793.633.533.04MIN: 9.9 / MAX: 10.91MIN: 9.82 / MAX: 10.81MIN: 6.87 / MAX: 9.77MIN: 7.01 / MAX: 7.86MIN: 6.83 / MAX: 7.6MIN: 6.72 / MAX: 7.38MIN: 6.73 / MAX: 7.34MIN: 5.75 / MAX: 6.19MIN: 5.7 / MAX: 6.16MIN: 5.31 / MAX: 5.97MIN: 3.65 / MAX: 3.94MIN: 3.44 / MAX: 3.8MIN: 3.46 / MAX: 3.66MIN: 2.93 / MAX: 3.16

Stockfish

This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 9Total Time2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL30M60M90M120M150MSE +/- 2015291.67, N = 3SE +/- 1266085.84, N = 3SE +/- 541700.90, N = 3SE +/- 902995.35, N = 3SE +/- 1223161.47, N = 3SE +/- 1934115.41, N = 3SE +/- 1396988.17, N = 4SE +/- 102308.18, N = 3SE +/- 336693.20, N = 3SE +/- 828671.04, N = 3SE +/- 829358.84, N = 3SE +/- 709349.35, N = 3SE +/- 54908.55, N = 3SE +/- 114508.53, N = 315311349215290686013098360211971579511676502811236272310975337710410667610351385385313242628913126117065158172107544371171. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: MD52 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280EPYC 2 7R32Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL2M4M6M8M10MSE +/- 29179.52, N = 3SE +/- 103333.33, N = 3SE +/- 14571.66, N = 3SE +/- 67263.16, N = 3SE +/- 32186.95, N = 3SE +/- 27545.92, N = 3SE +/- 11259.56, N = 3SE +/- 41921.89, N = 3SE +/- 1855.92, N = 3SE +/- 14468.36, N = 3SE +/- 30550.50, N = 3SE +/- 34226.37, N = 3SE +/- 14294.52, N = 3SE +/- 9073.77, N = 310714333104973339133000788000071500007102667660833365546675714333409000039720003727667365400033840001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

Appleseed

Appleseed is an open-source production renderer focused on physically-based global illumination rendering engine primarily designed for animation and visual effects. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterAppleseed 2.0 BetaScene: Disney MaterialEPYC 2 7R322 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8480+Xeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL30609012015062.4364.7564.9768.6272.8573.2273.5073.5877.3378.58103.83106.64117.15120.17

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: regex_compileXeon Platinum 8480+Xeon W-3275MXeon Platinum 82802 x Xeon Platinum 82802 x Xeon Platinum 8375CXeon Platinum 8375C2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8375C NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLEPYC 2 7R3250100150200250SE +/- 2.72, N = 15SE +/- 1.20, N = 3SE +/- 2.08, N = 3SE +/- 1.45, N = 3SE +/- 2.08, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3146157164164183186189190191200200200213

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: ETC1SXeon Platinum 8480+Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CLEPYC 2 7R322 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off1326395265SE +/- 0.35, N = 3SE +/- 0.48, N = 3SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 339.5847.6548.2348.9249.4051.4552.6954.1954.3155.2855.4657.4057.8358.281. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

Blender

Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: BMW27 - Compute: CPU-Only2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLEPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL20406080100SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.21, N = 3SE +/- 0.10, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.26, N = 3SE +/- 0.14, N = 331.0931.1035.2738.7941.2241.3341.7647.6348.0052.2869.8074.9675.8686.89

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: u8s8f322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL15003000450060007500SE +/- 10.36, N = 3SE +/- 7.16, N = 3SE +/- 5.84, N = 3SE +/- 7.97, N = 3SE +/- 25.92, N = 3SE +/- 16.05, N = 3SE +/- 15.22, N = 3SE +/- 11.90, N = 3SE +/- 7.22, N = 3SE +/- 1.23, N = 3SE +/- 11.11, N = 3SE +/- 41.02, N = 3SE +/- 14.11, N = 3SE +/- 5.52, N = 32382.112540.742550.213005.483076.763243.003271.403548.603551.384454.225919.946183.116437.607095.62MIN: 2282.91MIN: 2390.74MIN: 2528.47MIN: 2934.55MIN: 3011.81MIN: 3183.9MIN: 3209.1MIN: 3529.41MIN: 3527.13MIN: 4426.84MIN: 5895.99MIN: 5934.01MIN: 6341.76MIN: 7085.951. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon W-3275MEPYC 2 7R322 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CLXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off15003000450060007500SE +/- 4.09, N = 3SE +/- 1.99, N = 3SE +/- 0.44, N = 3SE +/- 6.81, N = 3SE +/- 3.72, N = 3SE +/- 22.45, N = 3SE +/- 5.39, N = 3SE +/- 4.11, N = 3SE +/- 1.29, N = 3SE +/- 6.12, N = 3SE +/- 10.51, N = 3SE +/- 1.29, N = 3SE +/- 1.57, N = 3SE +/- 2.97, N = 36769.274214.944211.214198.513866.093769.723514.603510.213331.383318.983315.303152.503147.913130.861. (CC) gcc options: -O3 -mavx2

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.13ATPase Simulation - 327,506 Atoms2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL0.19560.39120.58680.78240.978SE +/- 0.00026, N = 3SE +/- 0.00107, N = 3SE +/- 0.00028, N = 3SE +/- 0.00133, N = 3SE +/- 0.00016, N = 3SE +/- 0.00066, N = 3SE +/- 0.00038, N = 3SE +/- 0.00012, N = 3SE +/- 0.00002, N = 3SE +/- 0.00084, N = 3SE +/- 0.00060, N = 3SE +/- 0.01127, N = 4SE +/- 0.00775, N = 15SE +/- 0.00063, N = 30.249410.256320.291960.362260.388400.401730.445510.454220.469480.486140.707090.777240.814340.86925

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: u8s8f32Xeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CLXeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275MEPYC 2 7R32306090120150SE +/- 0.00297, N = 3SE +/- 0.00536, N = 3SE +/- 0.00823, N = 3SE +/- 0.00773, N = 3SE +/- 0.00908, N = 3SE +/- 0.01481, N = 3SE +/- 0.01090, N = 3SE +/- 0.04146, N = 4SE +/- 0.00398, N = 3SE +/- 0.03153, N = 3SE +/- 0.00539, N = 3SE +/- 0.00119, N = 3SE +/- 0.03226, N = 3SE +/- 0.23997, N = 31.292401.446131.498651.845152.781933.048663.105303.110023.231213.296153.344273.432363.59218149.33900MIN: 144.741. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch All - Data Type: f32Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA offXeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CLEPYC 2 7R322 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL3691215SE +/- 0.00487, N = 3SE +/- 0.00676, N = 3SE +/- 0.00218, N = 3SE +/- 0.00521, N = 3SE +/- 0.00556, N = 3SE +/- 0.00603, N = 3SE +/- 0.00854, N = 3SE +/- 0.01006, N = 3SE +/- 0.10112, N = 3SE +/- 0.03341, N = 3SE +/- 0.01277, N = 3SE +/- 0.01419, N = 3SE +/- 0.01633, N = 3SE +/- 0.03588, N = 32.914523.730314.595506.002646.386526.658426.795097.407047.544397.890718.883909.524669.713559.97669MIN: 2.69MIN: 3.56MIN: 4.48MIN: 5.79MIN: 6.06MIN: 6.25MIN: 6.68MIN: 7.28MIN: 6.71MIN: 7.76MIN: 8.36MIN: 9.28MIN: 9.49MIN: 9.621. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.D2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL2K4K6K8K10KSE +/- 102.38, N = 3SE +/- 143.88, N = 3SE +/- 307.30, N = 12SE +/- 148.86, N = 12SE +/- 89.09, N = 3SE +/- 91.47, N = 3SE +/- 89.39, N = 3SE +/- 84.22, N = 3SE +/- 5.66, N = 3SE +/- 5.75, N = 3SE +/- 46.76, N = 4SE +/- 4.09, N = 3SE +/- 39.87, N = 5SE +/- 38.58, N = 49339.049334.316288.826284.656088.085973.185310.165284.344730.263570.463283.383210.523033.282677.641. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Xeon Platinum 8375C: Open MPI 3.1.33. 2 x Xeon Platinum 8375C NUMA off: Open MPI 3.1.34. Xeon Platinum 8480+: Open MPI 3.1.35. 2 x Xeon Platinum 8280: Open MPI 4.0.3rc46. 2 x Xeon Platinum 8272CL: Open MPI 3.1.37. 2 x Xeon Platinum 8272CL NUMA off: Open MPI 3.1.38. 2 x Xeon Platinum 8259CL: Open MPI 3.1.39. 2 x Xeon Platinum 8259CL NUMA off: Open MPI 3.1.310. Xeon Platinum 8375C: Open MPI 3.1.311. EPYC 2 7R32: Open MPI 3.1.312. Xeon Platinum 8280: Open MPI 4.0.3rc413. Xeon W-3275M: Open MPI 3.1.314. Xeon Platinum 8272CL: Open MPI 3.1.315. Xeon Platinum 8259CL: Open MPI 3.1.3

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path Tracer2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL612182430SE +/- 0.000, N = 12SE +/- 0.000, N = 12SE +/- 0.000, N = 12SE +/- 0.000, N = 12SE +/- 0.101, N = 3SE +/- 0.068, N = 4SE +/- 0.000, N = 3SE +/- 0.080, N = 3SE +/- 0.067, N = 3SE +/- 0.030, N = 3SE +/- 0.000, N = 12SE +/- 0.006, N = 12SE +/- 0.022, N = 327.03027.03018.18017.54017.44016.60015.62515.54014.2209.4909.0908.6908.090MIN: 16.67 / MAX: 27.78MIN: 26.32 / MAX: 27.78MIN: 12.2 / MAX: 18.52MIN: 12.5 / MAX: 17.86MIN: 16.39 / MAX: 17.86MIN: 15.87 / MAX: 16.95MIN: 11.11 / MAX: 15.87MIN: 14.93 / MAX: 15.87MIN: 14.08 / MAX: 14.49MIN: 9.26 / MAX: 9.62MIN: 8.85 / MAX: 9.35MIN: 8.33 / MAX: 9.01MIN: 7.94 / MAX: 8.2

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CLXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL40K80K120K160K200KSE +/- 2821.73, N = 12SE +/- 116.37, N = 3SE +/- 78.84, N = 3SE +/- 125.50, N = 3SE +/- 2505.03, N = 15SE +/- 500.22, N = 3SE +/- 107.83, N = 3SE +/- 156.22, N = 3SE +/- 120.43, N = 3SE +/- 64.74, N = 3SE +/- 52.95, N = 3SE +/- 127.07, N = 3SE +/- 149.79, N = 3SE +/- 15.49, N = 3188359.36164568.74136046.53131794.23116616.82106415.53104107.99102113.0596192.0584766.1471842.3469347.5869011.3551975.351. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Xeon Platinum 8375C: Open MPI 3.1.33. 2 x Xeon Platinum 8375C NUMA off: Open MPI 3.1.34. 2 x Xeon Platinum 8280: Open MPI 4.0.3rc45. 2 x Xeon Platinum 8272CL: Open MPI 3.1.36. Xeon Platinum 8480+: Open MPI 3.1.37. EPYC 2 7R32: Open MPI 3.1.38. 2 x Xeon Platinum 8259CL: Open MPI 3.1.39. 2 x Xeon Platinum 8272CL NUMA off: Open MPI 3.1.310. Xeon Platinum 8375C: Open MPI 3.1.311. 2 x Xeon Platinum 8259CL NUMA off: Open MPI 3.1.312. Xeon Platinum 8280: Open MPI 4.0.3rc413. Xeon W-3275M: Open MPI 3.1.314. Xeon Platinum 8272CL: Open MPI 3.1.315. Xeon Platinum 8259CL: Open MPI 3.1.3

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080p2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon W-3275MXeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLEPYC 2 7R32Xeon Platinum 8272CL2 x Xeon Platinum 8375C NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8375CXeon Platinum 8480+Xeon Platinum 8375C246810SE +/- 0.064, N = 3SE +/- 0.081, N = 3SE +/- 0.062, N = 3SE +/- 0.023, N = 3SE +/- 0.032, N = 3SE +/- 0.075, N = 4SE +/- 0.002, N = 3SE +/- 0.072, N = 5SE +/- 0.052, N = 3SE +/- 0.110, N = 15SE +/- 0.052, N = 3SE +/- 0.083, N = 3SE +/- 0.011, N = 3SE +/- 0.043, N = 157.0336.4556.3996.3896.0305.9965.9555.7145.6535.2045.1124.8374.5024.0981. (CXX) g++ options: -fPIE -fPIC -pie

PyPerformance

PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyPerformance 1.0.0Benchmark: django_template2 x Xeon Platinum 8280Xeon Platinum 8280Xeon W-3275MXeon Platinum 8272CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8375CXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8375C NUMA offXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8480+20406080100SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.67, N = 3SE +/- 0.09, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.75, N = 15SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.50, N = 3SE +/- 0.35, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 345.946.249.960.660.660.863.264.164.164.265.365.466.478.2

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_3d - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8272CLXeon Platinum 8280Xeon W-3275MXeon Platinum 8259CL48121620SE +/- 0.00292, N = 3SE +/- 0.00440, N = 3SE +/- 0.05559, N = 3SE +/- 0.00581, N = 3SE +/- 0.01148, N = 3SE +/- 0.00328, N = 3SE +/- 0.00811, N = 3SE +/- 0.00172, N = 3SE +/- 0.00609, N = 3SE +/- 0.00634, N = 3SE +/- 0.00390, N = 3SE +/- 0.24390, N = 4SE +/- 0.00403, N = 32.981065.719635.812437.655617.675177.929129.187519.2066510.7508014.0747014.0780016.5906017.32170MIN: 5.59MIN: 5.59MIN: 7.54MIN: 7.54MIN: 7.75MIN: 9.03MIN: 9.04MIN: 10.65MIN: 13.94MIN: 13.95MIN: 14.98MIN: 17.211. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Asian Dragon Obj2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL1632486480SE +/- 0.04, N = 3SE +/- 0.19, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.30, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 370.3970.0551.7951.5949.2849.1147.5842.7642.4940.1829.2127.6727.2023.72MIN: 67.53 / MAX: 74.84MIN: 66.65 / MAX: 74.31MIN: 50.62 / MAX: 53.17MIN: 49.79 / MAX: 53.4MIN: 47.13 / MAX: 50.81MIN: 48.15 / MAX: 50.25MIN: 46.92 / MAX: 48.74MIN: 42.06 / MAX: 43.59MIN: 41.76 / MAX: 43.28MIN: 39.6 / MAX: 41.68MIN: 28.89 / MAX: 29.61MIN: 27.31 / MAX: 27.95MIN: 26.82 / MAX: 27.54MIN: 23.57 / MAX: 23.92

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: Blowfish2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLEPYC 2 7R32Xeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL20K40K60K80K100KSE +/- 33.58, N = 3SE +/- 1536.33, N = 3SE +/- 186.70, N = 3SE +/- 700.53, N = 3SE +/- 275.51, N = 3SE +/- 355.34, N = 3SE +/- 24.01, N = 3SE +/- 107.83, N = 3SE +/- 203.51, N = 3SE +/- 56.93, N = 3SE +/- 99.39, N = 3SE +/- 149.28, N = 3SE +/- 233.27, N = 3SE +/- 8.37, N = 31118471099789340984084784207830170955708886647556326428644032639604355991. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian Dragon Obj2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL20406080100SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 382.3281.9062.0558.6956.5156.4650.9850.8646.1445.4333.0531.5730.3528.29MIN: 79.32 / MAX: 84.86MIN: 79.15 / MAX: 84.55MIN: 59.57 / MAX: 64.55MIN: 56.97 / MAX: 60.91MIN: 55.22 / MAX: 58.38MIN: 55.08 / MAX: 57.97MIN: 49.91 / MAX: 52.27MIN: 49.88 / MAX: 52.05MIN: 45.35 / MAX: 47.2MIN: 44.82 / MAX: 46.53MIN: 32.72 / MAX: 33.4MIN: 31.3 / MAX: 31.91MIN: 29.95 / MAX: 30.77MIN: 28.06 / MAX: 28.6

Tungsten Renderer

Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Water CausticXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 8280EPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8259CLXeon Platinum 8480+714212835SE +/- 0.10, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.24, N = 3SE +/- 0.27, N = 3SE +/- 0.14, N = 3SE +/- 0.04, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.17, N = 3SE +/- 0.06, N = 3SE +/- 0.17, N = 321.5521.5922.0222.3223.4223.4923.6423.8223.9124.2124.5424.7825.3628.21-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a -ljpeg-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-march=znver1 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw1. (CXX) g++ options: -std=c++0x -msse2 -msse3 -mssse3 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lpthread -ldl

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: f322 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R32Xeon Platinum 8272CLXeon Platinum 8280Xeon Platinum 8259CLXeon W-3275M0.64511.29021.93532.58043.2255SE +/- 0.003943, N = 3SE +/- 0.007993, N = 3SE +/- 0.002860, N = 3SE +/- 0.004282, N = 3SE +/- 0.019533, N = 3SE +/- 0.014831, N = 15SE +/- 0.014870, N = 15SE +/- 0.009266, N = 3SE +/- 0.014367, N = 15SE +/- 0.006731, N = 3SE +/- 0.005715, N = 3SE +/- 0.010722, N = 3SE +/- 0.008928, N = 3SE +/- 0.005753, N = 30.7317530.7652790.7703061.0855301.5372901.5768701.6064101.6310601.6398601.6756002.3860602.3945102.6997702.867150MIN: 0.67MIN: 0.69MIN: 0.71MIN: 1.01MIN: 1.39MIN: 1.29MIN: 1.36MIN: 1.44MIN: 1.39MIN: 1.59MIN: 2.32MIN: 2.28MIN: 2.46MIN: 2.531. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Zstd Compression

This test measures the time needed to compress a sample file (an Ubuntu file-system image) using Zstd compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterZstd Compression 1.3.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19Xeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275M2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CLXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off3691215SE +/- 0.065, N = 15SE +/- 0.071, N = 13SE +/- 0.111, N = 12SE +/- 0.016, N = 3SE +/- 0.062, N = 14SE +/- 0.023, N = 3SE +/- 0.058, N = 15SE +/- 0.148, N = 3SE +/- 0.062, N = 15SE +/- 0.093, N = 15SE +/- 0.128, N = 3SE +/- 0.094, N = 15SE +/- 0.111, N = 7SE +/- 0.128, N = 56.1507.2838.1798.2038.3078.3808.4229.1379.2309.6539.6729.94910.08810.294-llzma1. (CC) gcc options: -O3 -pthread -lz

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: DecompressionXeon W-3275MXeon Platinum 8480+Xeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLEPYC 2 7R32110220330440550SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.67, N = 35094604404284274224164154134124043963943801. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: CompressionXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon W-3275M2 x Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offEPYC 2 7R32306090120150SE +/- 1.00, N = 3SE +/- 1.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 312011110910910610197969393918886811. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Intel MPI Benchmarks

Intel MPI Benchmarks for stressing MPI implementations. At this point the test profile aggregates results for some common MPI functionality. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgAverage Msg/sec, More Is BetterIntel MPI Benchmarks 2019.3Test: IMB-P2P PingPong2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8280Xeon Platinum 8480+Xeon Platinum 8375CEPYC 2 7R32Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CLXeon Platinum 828011M22M33M44M55MSE +/- 285425.17, N = 3SE +/- 568664.40, N = 3SE +/- 199886.33, N = 3SE +/- 90521.14, N = 3SE +/- 105849.64, N = 3SE +/- 80057.86, N = 3SE +/- 66348.18, N = 3SE +/- 70410.43, N = 3SE +/- 216886.06, N = 3SE +/- 40509.97, N = 3SE +/- 59477.15, N = 3SE +/- 60171.59, N = 3SE +/- 30074.61, N = 3SE +/- 33688.94, N = 350148418.0049151696.0636872935.1436807805.9332444647.4331988149.2531821690.5527196959.4324658155.9620348261.0320246604.6219096342.6516759832.2616653864.77MIN: 12462 / MAX: 134417572MIN: 9132 / MAX: 132488506MIN: 5799 / MAX: 95284327MIN: 9236 / MAX: 94667517MIN: 5770 / MAX: 85492119MIN: 9163 / MAX: 84595469MIN: 9950 / MAX: 90956260MIN: 10672 / MAX: 68333347MIN: 6906 / MAX: 64523629MIN: 6272 / MAX: 47546902MIN: 5167 / MAX: 52108336MIN: 4992 / MAX: 49169237MIN: 4962 / MAX: 43240261MIN: 5455 / MAX: 458333301. (CXX) g++ options: -O0 -pedantic -fopenmp -pthread -lmpi_cxx -lmpi

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.10.0Test: Scala DottyEPYC 2 7R32Xeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2K4K6K8K10KSE +/- 29.54, N = 5SE +/- 19.31, N = 5SE +/- 56.89, N = 13SE +/- 26.00, N = 5SE +/- 29.10, N = 5SE +/- 31.93, N = 5SE +/- 35.89, N = 5SE +/- 20.56, N = 5SE +/- 75.43, N = 8SE +/- 54.97, N = 5SE +/- 89.20, N = 5SE +/- 117.56, N = 25SE +/- 114.80, N = 25SE +/- 138.00, N = 56577.346664.256890.616900.196956.906970.907200.757256.927347.087566.348999.779170.229329.4311351.88

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8259CLXeon Platinum 8480+EPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon W-3275MXeon Platinum 8259CL40K80K120K160K200KSE +/- 2597.03, N = 15SE +/- 1042.35, N = 3SE +/- 188.25, N = 3SE +/- 55.32, N = 3SE +/- 1233.36, N = 3SE +/- 139.53, N = 3SE +/- 630.50, N = 3SE +/- 32.58, N = 3SE +/- 82.95, N = 3SE +/- 35.10, N = 3SE +/- 7.82, N = 3177387.18152798.56140090.32120961.91114346.4099821.8997414.0794642.9974255.6769606.8465522.461. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Xeon Platinum 8375C: Open MPI 3.1.33. 2 x Xeon Platinum 8375C NUMA off: Open MPI 3.1.34. 2 x Xeon Platinum 8280: Open MPI 4.0.3rc45. 2 x Xeon Platinum 8259CL: Open MPI 3.1.36. Xeon Platinum 8480+: Open MPI 3.1.37. EPYC 2 7R32: Open MPI 3.1.38. Xeon Platinum 8375C: Open MPI 3.1.39. 2 x Xeon Platinum 8259CL NUMA off: Open MPI 3.1.310. Xeon Platinum 8280: Open MPI 4.0.3rc411. Xeon W-3275M: Open MPI 3.1.312. Xeon Platinum 8259CL: Open MPI 3.1.3

Basis Universal

Basis Universal is a GPU texture codoec. This test times how long it takes to convert sRGB PNGs into Basis Univeral assets with various settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBasis Universal 1.12Settings: UASTC Level 3Xeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offEPYC 2 7R322 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon W-3275MXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CL816243240SE +/- 0.10, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.25, N = 6SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 316.2716.9217.3419.1919.8921.1221.4023.0923.1526.6030.6330.6932.2035.981. (CXX) g++ options: -std=c++11 -fvisibility=hidden -fPIC -fno-strict-aliasing -O3 -rdynamic -lm -lpthread

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: DecompressionXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375CXeon W-3275M2 x Xeon Platinum 8375C NUMA offEPYC 2 7R32Xeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8280Xeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off30060090012001500SE +/- 0.58, N = 3SE +/- 1.45, N = 3SE +/- 0.58, N = 3SE +/- 0.88, N = 3SE +/- 1.53, N = 3SE +/- 0.67, N = 3SE +/- 1.00, N = 3SE +/- 1.53, N = 3SE +/- 11.50, N = 315291412132012981291126612241145112410891058102210119661. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: CompressionXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL20406080100100898886848377757574737171691. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesXeon Platinum 8480+Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8272CLEPYC 2 7R32Xeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off30060090012001500SE +/- 3.38, N = 3SE +/- 1.45, N = 3SE +/- 3.21, N = 3SE +/- 1.45, N = 3SE +/- 1.86, N = 3SE +/- 1.00, N = 3SE +/- 1.15, N = 3SE +/- 0.58, N = 3SE +/- 0.58, N = 3SE +/- 2.33, N = 3SE +/- 3.18, N = 3SE +/- 0.67, N = 3SE +/- 1.00, N = 3SE +/- 1.33, N = 3693951978979980107310761182118211841226125012521253

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: DecompressionXeon W-3275MXeon Platinum 8480+Xeon Platinum 8280Xeon Platinum 8375C2 x Xeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R322 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off160320480640800SE +/- 1.45, N = 3SE +/- 1.15, N = 3SE +/- 0.67, N = 3SE +/- 3.18, N = 3SE +/- 1.33, N = 3SE +/- 0.58, N = 3SE +/- 5.55, N = 3SE +/- 1.00, N = 3SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 37217116736246236196176005875805715645515431. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: CompressionXeon Platinum 8480+Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 82802 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8272CLEPYC 2 7R32Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL50100150200250SE +/- 0.33, N = 3SE +/- 1.20, N = 3SE +/- 0.67, N = 3SE +/- 0.33, N = 3SE +/- 1.33, N = 32141961841831811781761681681661651601591571. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: DecompressionXeon Platinum 8480+Xeon Platinum 8375CXeon W-3275M2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 82802 x Xeon Platinum 8280EPYC 2 7R32Xeon Platinum 8272CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL30060090012001500SE +/- 4.04, N = 3SE +/- 0.58, N = 3SE +/- 2.65, N = 3SE +/- 0.67, N = 3SE +/- 1.53, N = 3SE +/- 2.00, N = 3SE +/- 5.03, N = 3SE +/- 6.01, N = 3SE +/- 0.67, N = 3SE +/- 0.67, N = 3140013211299127112581212116811481122108810781061103210271. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: CompressionXeon Platinum 8480+Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 82802 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8272CLEPYC 2 7R32Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL110220330440550SE +/- 3.21, N = 3SE +/- 1.00, N = 3SE +/- 0.33, N = 3SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 2.19, N = 3SE +/- 1.00, N = 3SE +/- 0.33, N = 3SE +/- 2.33, N = 3SE +/- 2.33, N = 35275084644614574534494304304234184104073991. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmXeon Platinum 8480+EPYC 2 7R32Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375CXeon Platinum 82802 x Xeon Platinum 82802 x Xeon Platinum 8375C NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8272CLXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off510152025SE +/- 0.02, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 37.9915.1615.7817.0517.0817.1917.1917.5417.8317.9418.6718.7619.0519.62

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32Xeon Platinum 8480+2 x Xeon Platinum 8375CXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 82802 x Xeon Platinum 8259CLXeon Platinum 8272CL2 x Xeon Platinum 8280Xeon Platinum 8259CLXeon W-3275MEPYC 2 7R322004006008001000SE +/- 0.002086, N = 3SE +/- 0.002925, N = 3SE +/- 0.000218, N = 3SE +/- 0.001434, N = 3SE +/- 0.002279, N = 3SE +/- 0.001261, N = 3SE +/- 0.001016, N = 3SE +/- 0.000361, N = 3SE +/- 0.000864, N = 3SE +/- 0.000242, N = 3SE +/- 0.006813, N = 3SE +/- 0.000985, N = 3SE +/- 0.001267, N = 3SE +/- 4.136428, N = 30.2811590.2994730.3010070.3063090.3364260.3517190.3614140.3710360.3762220.3782290.3944550.4156140.4584221131.130000MIN: 1118.791. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL246810SE +/- 0.00059, N = 3SE +/- 0.00390, N = 3SE +/- 0.00343, N = 3SE +/- 0.00220, N = 3SE +/- 0.00409, N = 3SE +/- 0.00661, N = 3SE +/- 0.00151, N = 3SE +/- 0.00489, N = 3SE +/- 0.00112, N = 3SE +/- 0.00167, N = 3SE +/- 0.00188, N = 3SE +/- 0.03740, N = 3SE +/- 0.02376, N = 32.157422.914953.077973.950673.983934.068474.465234.490394.978886.553746.597787.615957.98292MIN: 2.1MIN: 2.81MIN: 2.81MIN: 3.81MIN: 3.85MIN: 3.93MIN: 4.38MIN: 4.4MIN: 4.92MIN: 6.5MIN: 6.49MIN: 6.95MIN: 7.921. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVis2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offEPYC 2 7R32Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL306090120150SE +/- 1.62, N = 15SE +/- 2.25, N = 15SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 3SE +/- 0.00, N = 12140.48136.9090.9190.9190.9176.9276.9276.9271.4350.0047.6245.4543.48MIN: 34.48 / MAX: 142.86MIN: 66.67 / MAX: 142.86MIN: 29.41MIN: 28.57MIN: 71.43MIN: 27.78 / MAX: 83.33MIN: 71.43 / MAX: 83.33MIN: 71.43 / MAX: 83.33MIN: 62.5 / MAX: 76.92MIN: 43.48MIN: 41.67 / MAX: 50MIN: 43.48 / MAX: 47.62MIN: 37.04

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_1d - Data Type: f32Xeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275MEPYC 2 7R320.50071.00141.50212.00282.5035SE +/- 0.002767, N = 3SE +/- 0.005279, N = 3SE +/- 0.004528, N = 3SE +/- 0.000656, N = 3SE +/- 0.000614, N = 3SE +/- 0.003180, N = 3SE +/- 0.006932, N = 3SE +/- 0.002914, N = 3SE +/- 0.001218, N = 3SE +/- 0.001574, N = 3SE +/- 0.001211, N = 3SE +/- 0.000966, N = 3SE +/- 0.012727, N = 3SE +/- 0.019719, N = 30.9986291.0189001.0311501.2258901.2400101.2402901.2579501.3039501.3441101.4623201.4708401.6392801.7451502.225400MIN: 0.94MIN: 0.96MIN: 0.98MIN: 1.18MIN: 1.2MIN: 1.17MIN: 1.2MIN: 1.2MIN: 1.23MIN: 1.44MIN: 1.43MIN: 1.61MIN: 1.55MIN: 2.111. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL2004006008001000SE +/- 1.87, N = 4SE +/- 2.42, N = 3SE +/- 2.55, N = 3SE +/- 0.06, N = 3SE +/- 0.18, N = 3SE +/- 5.80, N = 3SE +/- 0.23, N = 3SE +/- 0.11, N = 3SE +/- 5.46, N = 3SE +/- 0.07, N = 3SE +/- 0.16, N = 3SE +/- 1.04, N = 3SE +/- 0.41, N = 3129.03251.83252.41320.06332.32337.88399.30399.31500.12637.37662.76702.07796.75MIN: 247.18MIN: 247.48MIN: 319.23MIN: 331.59MIN: 331.31MIN: 398.05MIN: 398.47MIN: 493.61MIN: 636.97MIN: 661.35MIN: 687.86MIN: 795.521. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

ArrayFire

ArrayFire is an GPU and CPU numeric processing library, this test uses the built-in CPU and OpenCL ArrayFire benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterArrayFire 3.7Test: BLAS CPU2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8480+2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CLXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CLEPYC 2 7R3214002800420056007000SE +/- 23.67, N = 3SE +/- 14.26, N = 3SE +/- 14.66, N = 3SE +/- 122.72, N = 12SE +/- 18.17, N = 3SE +/- 28.94, N = 3SE +/- 7.08, N = 3SE +/- 240.61, N = 12SE +/- 25.11, N = 3SE +/- 20.06, N = 3SE +/- 8.38, N = 3SE +/- 30.40, N = 3SE +/- 1.09, N = 3SE +/- 6.50, N = 36330.935996.405650.715319.584875.474631.384520.183972.323432.163430.283281.583128.392950.16990.801. (CXX) g++ options: -rdynamic

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080p2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon W-3275M2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8480+1530456075SE +/- 1.09, N = 3SE +/- 0.58, N = 15SE +/- 0.67, N = 3SE +/- 0.77, N = 3SE +/- 0.35, N = 3SE +/- 0.43, N = 3SE +/- 0.46, N = 3SE +/- 0.13, N = 3SE +/- 0.29, N = 3SE +/- 0.07, N = 3SE +/- 0.37, N = 3SE +/- 0.35, N = 3SE +/- 0.53, N = 5SE +/- 0.31, N = 367.0865.3056.9854.4754.1453.9653.8253.0452.3151.9251.5946.8741.8440.651. (CXX) g++ options: -fPIE -fPIC -pie

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Crown2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offEPYC 2 7R322 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL1530456075SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.16, N = 3SE +/- 0.18, N = 3SE +/- 0.31, N = 3SE +/- 0.17, N = 3SE +/- 0.15, N = 3SE +/- 0.31, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 369.2669.1447.6545.6145.4743.9943.8238.1238.1037.6326.2825.1224.5521.38MIN: 67.05 / MAX: 73.59MIN: 66.94 / MAX: 73.49MIN: 46.77 / MAX: 49.17MIN: 44.2 / MAX: 47.35MIN: 42.39 / MAX: 48.28MIN: 42.79 / MAX: 45.69MIN: 41.48 / MAX: 45.92MIN: 37.03 / MAX: 39.42MIN: 37.24 / MAX: 39.22MIN: 36.48 / MAX: 39.76MIN: 25.64 / MAX: 26.64MIN: 24.68 / MAX: 25.5MIN: 24.09 / MAX: 24.98MIN: 21.05 / MAX: 21.63

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: u8s8f322 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 8375CXeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275MEPYC 2 7R3230060090012001500SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.24, N = 3SE +/- 0.02, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 2.69, N = 310.9711.6812.3317.7218.3919.3719.8120.0922.3826.8427.9830.8831.991301.36MIN: 1286.291. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.6.0Video Input: Summer Nature 4KEPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 8375CXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8480+2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8272CLXeon Platinum 8259CLXeon Platinum 828070140210280350SE +/- 0.98, N = 3SE +/- 1.20, N = 3SE +/- 1.25, N = 3SE +/- 1.86, N = 3SE +/- 2.46, N = 8SE +/- 3.68, N = 3SE +/- 2.41, N = 3SE +/- 1.73, N = 3SE +/- 1.78, N = 3SE +/- 2.65, N = 5SE +/- 2.74, N = 3SE +/- 0.71, N = 3SE +/- 1.31, N = 3SE +/- 1.15, N = 3321.17256.60251.63247.86246.96221.31216.00215.64212.99211.21211.05208.50205.63199.08MIN: 150.11 / MAX: 347.95MIN: 164.1 / MAX: 284.35MIN: 70.21 / MAX: 290.5MIN: 139.62 / MAX: 276.02MIN: 58.67 / MAX: 300.85MIN: 81.4 / MAX: 247.6MIN: 87.7 / MAX: 239.38MIN: 93.88 / MAX: 236.99MIN: 89.5 / MAX: 233.99MIN: 53.65 / MAX: 250.48MIN: 77.75 / MAX: 238.31MIN: 127.92 / MAX: 223.11MIN: 134.32 / MAX: 229.88MIN: 121.12 / MAX: 214.691. (CC) gcc options: -pthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Crown2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLEPYC 2 7R32Xeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL20406080100SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.46, N = 3SE +/- 0.63, N = 5SE +/- 0.05, N = 3SE +/- 0.38, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 378.6378.5553.2353.2151.2050.9546.4246.1845.1942.6028.7727.7826.4024.73MIN: 75.59 / MAX: 83.02MIN: 75.52 / MAX: 82.95MIN: 51.36 / MAX: 55.83MIN: 50.57 / MAX: 56.71MIN: 47.31 / MAX: 53.67MIN: 45.23 / MAX: 53.81MIN: 45.24 / MAX: 47.89MIN: 42.99 / MAX: 47.93MIN: 44.15 / MAX: 46.52MIN: 41.98 / MAX: 43.65MIN: 28.26 / MAX: 29.24MIN: 27.34 / MAX: 28.17MIN: 25.71 / MAX: 26.95MIN: 24.4 / MAX: 25.05

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Convolution Batch conv_alexnet - Data Type: f322 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280EPYC 2 7R32Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275M20406080100SE +/- 0.02, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.11, N = 3SE +/- 0.50, N = 3SE +/- 0.20, N = 3SE +/- 0.23, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 1.13, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.11, N = 342.9143.6643.8649.8451.4453.0055.3656.6583.3694.1996.6998.67107.80108.57MIN: 41.86MIN: 43.12MIN: 42.5MIN: 48.76MIN: 50.35MIN: 50.88MIN: 54.28MIN: 54.78MIN: 82.91MIN: 93.73MIN: 93.63MIN: 97.89MIN: 106.14MIN: 102.461. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.B2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLEPYC 2 7R32Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL20K40K60K80K100KSE +/- 285.88, N = 3SE +/- 1751.95, N = 15SE +/- 1041.11, N = 15SE +/- 115.61, N = 3SE +/- 72.16, N = 3SE +/- 77.17, N = 3SE +/- 520.84, N = 15SE +/- 648.57, N = 15SE +/- 22.17, N = 3SE +/- 42.29, N = 3SE +/- 151.80, N = 3SE +/- 146.25, N = 3SE +/- 115.29, N = 3SE +/- 28.87, N = 3112350.4795294.9786029.3270893.4166865.4553978.9051225.1846438.5846295.7938493.5732682.1431426.5329919.0124288.341. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Xeon Platinum 8375C: Open MPI 3.1.33. Xeon Platinum 8480+: Open MPI 3.1.34. 2 x Xeon Platinum 8375C NUMA off: Open MPI 3.1.35. 2 x Xeon Platinum 8280: Open MPI 4.0.3rc46. 2 x Xeon Platinum 8272CL: Open MPI 3.1.37. 2 x Xeon Platinum 8259CL: Open MPI 3.1.38. EPYC 2 7R32: Open MPI 3.1.39. Xeon Platinum 8375C: Open MPI 3.1.310. 2 x Xeon Platinum 8272CL NUMA off: Open MPI 3.1.311. 2 x Xeon Platinum 8259CL NUMA off: Open MPI 3.1.312. Xeon Platinum 8280: Open MPI 4.0.3rc413. Xeon W-3275M: Open MPI 3.1.314. Xeon Platinum 8272CL: Open MPI 3.1.315. Xeon Platinum 8259CL: Open MPI 3.1.3

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer - Model: Asian Dragon2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R322 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL20406080100SE +/- 0.12, N = 3SE +/- 0.40, N = 3SE +/- 0.16, N = 3SE +/- 0.17, N = 3SE +/- 0.27, N = 3SE +/- 0.14, N = 3SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.38, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 378.7778.5457.4956.5754.2054.2052.9546.9146.7944.3132.2530.5730.1126.06MIN: 75.79 / MAX: 82.75MIN: 73.71 / MAX: 82.39MIN: 55.98 / MAX: 58.87MIN: 53.23 / MAX: 58.38MIN: 52.81 / MAX: 55.78MIN: 53.05 / MAX: 55.48MIN: 52.3 / MAX: 53.84MIN: 46.17 / MAX: 47.87MIN: 44.43 / MAX: 47.97MIN: 43.55 / MAX: 45.94MIN: 32 / MAX: 32.54MIN: 30.32 / MAX: 30.89MIN: 29.8 / MAX: 30.54MIN: 25.85 / MAX: 26.28

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. The number of iterations used is 1,000,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteXeon Platinum 8480+Xeon W-3275MXeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8259CLEPYC 2 7R32200K400K600K800K1000KSE +/- 1409.89, N = 3SE +/- 1014.34, N = 3SE +/- 307.66, N = 3SE +/- 3064.50, N = 3SE +/- 1813.04, N = 3SE +/- 884.68, N = 3SE +/- 425.09, N = 3SE +/- 505.94, N = 3SE +/- 894.52, N = 3SE +/- 1286.95, N = 3SE +/- 377.70, N = 3SE +/- 961.65, N = 3SE +/- 778.31, N = 3SE +/- 1283.26, N = 3893124734339712140708571704453649693649164606548606436604857574348573483569793500524

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: IP Batch 1D - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL1.28232.56463.84695.12926.4115SE +/- 0.00321, N = 3SE +/- 0.01294, N = 3SE +/- 0.00443, N = 3SE +/- 0.00267, N = 3SE +/- 0.00609, N = 3SE +/- 0.00405, N = 3SE +/- 0.01183, N = 3SE +/- 0.00913, N = 3SE +/- 0.00431, N = 3SE +/- 0.00377, N = 3SE +/- 0.02664, N = 3SE +/- 0.01234, N = 3SE +/- 0.00262, N = 31.472561.739501.742132.419062.853102.890863.659644.464404.489184.831705.563045.596805.69900MIN: 1.39MIN: 1.48MIN: 1.5MIN: 2.23MIN: 2.7MIN: 2.77MIN: 3.47MIN: 4.28MIN: 4.31MIN: 4.76MIN: 5.04MIN: 5.51MIN: 5.571. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.6.0Video Input: Summer Nature 1080pEPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8375C2 x Xeon Platinum 8259CL2 x Xeon Platinum 8280Xeon Platinum 8480+2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CLXeon Platinum 8280Xeon Platinum 8272CL150300450600750SE +/- 2.19, N = 3SE +/- 2.32, N = 3SE +/- 4.53, N = 15SE +/- 1.14, N = 3SE +/- 5.59, N = 3SE +/- 3.67, N = 3SE +/- 3.74, N = 15SE +/- 5.27, N = 4SE +/- 3.77, N = 15SE +/- 5.42, N = 4SE +/- 0.73, N = 3SE +/- 1.73, N = 3SE +/- 0.52, N = 3SE +/- 2.16, N = 3676.16537.84500.83478.05444.30431.76409.88403.30400.24391.25371.48353.78323.20315.51MIN: 302.17 / MAX: 751.14MIN: 317.28 / MAX: 605.87MIN: 107.59 / MAX: 628.01MIN: 104.25 / MAX: 592.47MIN: 180.09 / MAX: 495.08MIN: 205.51 / MAX: 479.42MIN: 126.27 / MAX: 481.12MIN: 84.43 / MAX: 499.24MIN: 106.81 / MAX: 466.09MIN: 158.31 / MAX: 448.01MIN: 133.17 / MAX: 428.07MIN: 207.11 / MAX: 387.94MIN: 200.79 / MAX: 358.32MIN: 164.62 / MAX: 350.651. (CC) gcc options: -pthread

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL9001800270036004500SE +/- 1.01, N = 3SE +/- 9.75, N = 3SE +/- 0.92, N = 3SE +/- 5.70, N = 3SE +/- 3.22, N = 3SE +/- 0.39, N = 3SE +/- 0.61, N = 3SE +/- 3.00, N = 3SE +/- 22.00, N = 3SE +/- 0.87, N = 3SE +/- 2.14, N = 3SE +/- 48.27, N = 3SE +/- 8.91, N = 3SE +/- 0.90, N = 31295.481320.601445.071893.721899.242063.232064.022238.052252.672530.963716.983852.373985.104426.06MIN: 1272.23MIN: 1274.59MIN: 1419.8MIN: 1871.79MIN: 1888.24MIN: 2061.82MIN: 2062.22MIN: 2230.65MIN: 2229.49MIN: 2512.49MIN: 3705.72MIN: 3728.62MIN: 3970.03MIN: 4423.171. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian Dragon2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL20406080100SE +/- 0.08, N = 3SE +/- 0.51, N = 3SE +/- 0.20, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.03, N = 395.9995.2771.6268.4265.5565.3759.0659.0553.5752.3538.3136.7535.1532.91MIN: 92.5 / MAX: 98.6MIN: 91.48 / MAX: 98.3MIN: 66.7 / MAX: 74.09MIN: 66.22 / MAX: 70.39MIN: 64.05 / MAX: 67.28MIN: 63.67 / MAX: 66.98MIN: 57.51 / MAX: 60.42MIN: 57.78 / MAX: 60.29MIN: 52.72 / MAX: 54.62MIN: 51.67 / MAX: 53.48MIN: 37.87 / MAX: 38.72MIN: 36.43 / MAX: 37.08MIN: 33.95 / MAX: 35.72MIN: 32.68 / MAX: 33.18

Scikit-Learn

Scikit-learn is a Python module for machine learning Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 0.22.1Xeon Platinum 8480+Xeon Platinum 82802 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8375C NUMA offXeon Platinum 8375CXeon W-3275MEPYC 2 7R32Xeon Platinum 8272CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off3691215SE +/- 0.062, N = 15SE +/- 0.042, N = 3SE +/- 0.032, N = 3SE +/- 0.122, N = 3SE +/- 0.058, N = 3SE +/- 0.053, N = 3SE +/- 0.035, N = 3SE +/- 0.037, N = 3SE +/- 0.026, N = 3SE +/- 0.025, N = 3SE +/- 0.047, N = 3SE +/- 0.056, N = 3SE +/- 0.039, N = 3SE +/- 0.043, N = 36.8309.1279.3329.3929.4729.5669.66110.41911.21211.76312.05512.08812.45912.954

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.1Tuning: Visual Quality Optimized - Input: Bosphorus 1080pEPYC 2 7R32Xeon W-3275MXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8259CLXeon Platinum 8272CL2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8480+60120180240300SE +/- 2.49, N = 3SE +/- 0.97, N = 3SE +/- 1.70, N = 3SE +/- 3.65, N = 4SE +/- 1.36, N = 3SE +/- 9.84, N = 12SE +/- 2.02, N = 13SE +/- 3.68, N = 15SE +/- 2.89, N = 4SE +/- 8.04, N = 12SE +/- 8.15, N = 12SE +/- 8.26, N = 12SE +/- 7.67, N = 13SE +/- 5.95, N = 12291.17282.18266.29265.40256.24256.02249.76243.24233.17229.91229.53227.90227.89205.611. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm

ArrayFire

ArrayFire is an GPU and CPU numeric processing library, this test uses the built-in CPU and OpenCL ArrayFire benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient CPUXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL2 x Xeon Platinum 8375C2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA offEPYC 2 7R32816243240SE +/- 0.056, N = 12SE +/- 0.009, N = 3SE +/- 0.073, N = 15SE +/- 0.067, N = 3SE +/- 0.005, N = 3SE +/- 0.075, N = 15SE +/- 0.022, N = 3SE +/- 0.032, N = 3SE +/- 0.129, N = 3SE +/- 0.086, N = 6SE +/- 0.243, N = 15SE +/- 0.091, N = 15SE +/- 0.188, N = 15SE +/- 0.217, N = 32.1492.8144.0364.1154.1964.4234.6777.3557.4677.4817.6048.4308.94735.1301. (CXX) g++ options: -rdynamic

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVis2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MEPYC 2 7R32Xeon Platinum 8259CL20406080100SE +/- 0.00, N = 15SE +/- 1.45, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.60, N = 3SE +/- 0.00, N = 12111.11108.3371.4371.4371.4362.5062.5062.5042.8740.0040.0040.0037.04MIN: 76.92MIN: 22.22 / MAX: 111.11MIN: 15.87 / MAX: 76.92MIN: 19.23MIN: 58.82MIN: 58.82 / MAX: 66.67MIN: 18.52 / MAX: 66.67MIN: 58.82 / MAX: 66.67MIN: 38.46 / MAX: 43.48MIN: 34.48 / MAX: 41.67MIN: 37.04MIN: 38.46 / MAX: 41.67MIN: 33.33

SMHasher

SMHasher is a hash function tester. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2020-02-29Hash: wyhashXeon W-3275MXeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8259CL6K12K18K24K30KSE +/- 105.54, N = 3SE +/- 3.98, N = 3SE +/- 5.31, N = 3SE +/- 4.10, N = 3SE +/- 0.70, N = 3SE +/- 6.86, N = 3SE +/- 8.73, N = 326298.7720556.9620554.9220553.0520231.5820225.1820221.251. (CXX) g++ options: -march=native -O3 -lpthread

OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2020-02-29Hash: wyhashXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8272CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8259CL612182430SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 318.5122.6822.6823.6523.6723.6824.1524.1524.191. (CXX) g++ options: -march=native -O3 -lpthread

Tungsten Renderer

Tungsten is a C++ physically based renderer that makes use of Intel's Embree ray tracing library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Hair2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+EPYC 2 7R322 x Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8259CLXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL48121620SE +/- 0.02429, N = 3SE +/- 0.06608, N = 3SE +/- 0.09675, N = 4SE +/- 0.05445, N = 3SE +/- 0.11249, N = 3SE +/- 0.00398, N = 3SE +/- 0.10478, N = 3SE +/- 0.05235, N = 3SE +/- 0.01484, N = 3SE +/- 0.05659, N = 3SE +/- 0.03914, N = 3SE +/- 0.05256, N = 3SE +/- 0.04470, N = 3SE +/- 0.03318, N = 36.056686.117906.665297.674557.827068.347448.667919.090739.213659.2621712.6863012.7104013.8818014.80460-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-march=znver1 -msse4.1 -msse4.2 -msse4a -mfma -mbmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-march=core2 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-fma -mno-bmi2 -mno-avx512f -mno-avx512vl -mno-avx512cd -mno-avx512dq -mno-avx512bw-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a -ljpeg-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a-msse4.1 -msse4.2 -mfma -mbmi2 -mavx512f -mavx512vl -mavx512cd -mavx512dq -mavx512bw -mno-sse4a1. (CXX) g++ options: -std=c++0x -msse2 -msse3 -mssse3 -mno-avx -mno-avx2 -mno-xop -mno-fma4 -mno-avx512pf -mno-avx512er -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -O3 -rdynamic -lpthread -ldl

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonXeon W-3275MXeon Platinum 82802 x Xeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CL2 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLEPYC 2 7R32Xeon Platinum 8480+2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off12002400360048006000SE +/- 10.08, N = 4SE +/- 15.39, N = 4SE +/- 19.30, N = 4SE +/- 15.06, N = 4SE +/- 22.03, N = 4SE +/- 37.95, N = 4SE +/- 25.41, N = 4SE +/- 7.82, N = 4SE +/- 11.32, N = 4SE +/- 56.00, N = 4SE +/- 10.59, N = 4SE +/- 12.06, N = 4SE +/- 24.79, N = 4SE +/- 43.39, N = 443214547471147865012501350375112516151725246525653385400

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 9Jan2020Model: Rhodopsin Protein2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offEPYC 2 7R32Xeon Platinum 8480+Xeon Platinum 8375CXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL612182430SE +/- 0.92, N = 15SE +/- 0.81, N = 15SE +/- 0.25, N = 14SE +/- 0.26, N = 6SE +/- 0.12, N = 3SE +/- 0.33, N = 3SE +/- 0.32, N = 3SE +/- 0.37, N = 15SE +/- 0.55, N = 12SE +/- 0.44, N = 15SE +/- 0.19, N = 3SE +/- 0.13, N = 15SE +/- 0.03, N = 3SE +/- 0.08, N = 327.3525.9824.7923.4322.6722.5421.8920.8219.7615.9015.3114.5614.0113.04-lpng -lz-lpng -lz-ljpeg -lpng -lz1. (CXX) g++ options: -O3 -rdynamic -lfftw3 -lm

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.0.0Scene: MemorialXeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8272CL2 x Xeon Platinum 82802 x Xeon Platinum 8259CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8280Xeon Platinum 8272CLEPYC 2 7R32Xeon Platinum 8259CLXeon W-3275M1326395265SE +/- 3.68, N = 12SE +/- 0.26, N = 15SE +/- 0.14, N = 3SE +/- 0.17, N = 3SE +/- 0.12, N = 3SE +/- 0.18, N = 3SE +/- 0.21, N = 3SE +/- 0.30, N = 10SE +/- 0.32, N = 9SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.30, N = 356.0136.5635.5234.4933.8933.5733.1032.9632.2931.1030.2228.7127.9927.35

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.C2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8375CEPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL20K40K60K80K100KSE +/- 1012.46, N = 3SE +/- 208.05, N = 3SE +/- 131.11, N = 3SE +/- 860.82, N = 5SE +/- 229.09, N = 3SE +/- 304.73, N = 3SE +/- 926.87, N = 15SE +/- 796.48, N = 3SE +/- 31.00, N = 3SE +/- 420.94, N = 3SE +/- 129.93, N = 3SE +/- 40.04, N = 3SE +/- 23.88, N = 3SE +/- 114.90, N = 3111939.6282633.6482390.2469175.1266470.5663393.3550840.7649330.1545188.2343490.1537089.0535376.1134708.6033490.031. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Xeon Platinum 8375C: Open MPI 3.1.33. Xeon Platinum 8480+: Open MPI 3.1.34. 2 x Xeon Platinum 8375C NUMA off: Open MPI 3.1.35. 2 x Xeon Platinum 8280: Open MPI 4.0.3rc46. 2 x Xeon Platinum 8272CL: Open MPI 3.1.37. 2 x Xeon Platinum 8259CL: Open MPI 3.1.38. Xeon Platinum 8375C: Open MPI 3.1.39. EPYC 2 7R32: Open MPI 3.1.310. 2 x Xeon Platinum 8272CL NUMA off: Open MPI 3.1.311. 2 x Xeon Platinum 8259CL NUMA off: Open MPI 3.1.312. Xeon Platinum 8280: Open MPI 4.0.3rc413. Xeon W-3275M: Open MPI 3.1.314. Xeon Platinum 8272CL: Open MPI 3.1.315. Xeon Platinum 8259CL: Open MPI 3.1.3

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16Xeon Platinum 8480+2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon W-3275MXeon Platinum 8259CL3691215SE +/- 0.02482, N = 13SE +/- 0.03881, N = 8SE +/- 0.04017, N = 7SE +/- 0.00369, N = 3SE +/- 0.00318, N = 3SE +/- 0.00376, N = 3SE +/- 0.00526, N = 3SE +/- 0.00712, N = 3SE +/- 0.07400, N = 6SE +/- 0.00251, N = 3SE +/- 0.00016, N = 3SE +/- 0.03985, N = 3SE +/- 0.00210, N = 31.740053.777693.783194.358174.494054.494644.856894.871946.071177.696058.316778.462739.37107MIN: 3.68MIN: 3.68MIN: 4.31MIN: 4.44MIN: 4.46MIN: 4.82MIN: 4.83MIN: 5.82MIN: 7.67MIN: 8.25MIN: 8.21MIN: 9.331. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.C2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375C2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8480+EPYC 2 7R32Xeon Platinum 8280Xeon W-3275MXeon Platinum 8272CLXeon Platinum 8259CL14002800420056007000SE +/- 138.06, N = 15SE +/- 202.13, N = 15SE +/- 88.25, N = 3SE +/- 25.51, N = 3SE +/- 40.42, N = 3SE +/- 34.33, N = 3SE +/- 77.32, N = 3SE +/- 107.97, N = 15SE +/- 3.84, N = 3SE +/- 31.12, N = 3SE +/- 44.67, N = 3SE +/- 40.50, N = 3SE +/- 17.03, N = 3SE +/- 14.22, N = 36744.576352.446104.945763.375667.765059.025052.634003.823863.003485.993194.183025.112974.152641.321. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. 2 x Xeon Platinum 8375C NUMA off: Open MPI 3.1.33. 2 x Xeon Platinum 8375C: Open MPI 3.1.34. 2 x Xeon Platinum 8280: Open MPI 4.0.3rc45. 2 x Xeon Platinum 8272CL: Open MPI 3.1.36. 2 x Xeon Platinum 8272CL NUMA off: Open MPI 3.1.37. 2 x Xeon Platinum 8259CL NUMA off: Open MPI 3.1.38. 2 x Xeon Platinum 8259CL: Open MPI 3.1.39. Xeon Platinum 8375C: Open MPI 3.1.310. Xeon Platinum 8480+: Open MPI 3.1.311. EPYC 2 7R32: Open MPI 3.1.312. Xeon Platinum 8280: Open MPI 4.0.3rc413. Xeon W-3275M: Open MPI 3.1.314. Xeon Platinum 8272CL: Open MPI 3.1.315. Xeon Platinum 8259CL: Open MPI 3.1.3

Nettle

GNU Nettle is a low-level cryptographic library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.5.1Test: aes256Xeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8259CL2 x Xeon Platinum 8259CLEPYC 2 7R322K4K6K8K10KSE +/- 2.51, N = 3SE +/- 9.18, N = 3SE +/- 15.44, N = 3SE +/- 31.31, N = 3SE +/- 39.45, N = 3SE +/- 3.29, N = 3SE +/- 7.98, N = 3SE +/- 2.12, N = 3SE +/- 5.48, N = 3SE +/- 9.93, N = 3SE +/- 5.89, N = 3SE +/- 20.53, N = 3SE +/- 24.71, N = 3SE +/- 0.67, N = 39126.498052.908044.398007.347070.606320.006313.165852.825852.225830.305523.315517.025513.084665.33MIN: 5913.05 / MAX: 15615.18MIN: 5267.38 / MAX: 13669.67MIN: 5265.65 / MAX: 13663.69MIN: 5102.53 / MAX: 13668.47MIN: 4721.3 / MAX: 11806.11MIN: 4254.97 / MAX: 10469.4MIN: 4256.89 / MAX: 10467.36MIN: 3956.27 / MAX: 9681.61MIN: 3943.77 / MAX: 9693.14MIN: 3880.03 / MAX: 9687.65MIN: 3739.77 / MAX: 9132.41MIN: 3740.26 / MAX: 9153.64MIN: 3744.15 / MAX: 9173.02MIN: 3372.31 / MAX: 7239.741. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.5.1Test: sha512Xeon Platinum 8480+Xeon W-3275MXeon Platinum 82802 x Xeon Platinum 8280EPYC 2 7R322 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8272CL2 x Xeon Platinum 8375C2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8259CLXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA off130260390520650SE +/- 4.00, N = 3SE +/- 2.72, N = 3SE +/- 0.15, N = 3SE +/- 0.97, N = 3SE +/- 0.28, N = 3SE +/- 0.20, N = 3SE +/- 0.19, N = 3SE +/- 1.50, N = 3SE +/- 9.89, N = 15SE +/- 1.80, N = 3SE +/- 2.61, N = 3SE +/- 2.12, N = 3SE +/- 0.08, N = 3SE +/- 0.13, N = 3616.59550.93501.98500.98478.09465.40465.18462.78456.39438.23437.62436.73396.69396.531. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto

MKL-DNN DNNL

This is a test of the Intel MKL-DNN (DNNL / Deep Neural Network Library) as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMKL-DNN DNNL 1.1Harness: Deconvolution Batch deconv_3d - Data Type: f322 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CXeon Platinum 8480+2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8375CXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLEPYC 2 7R32Xeon W-3275M0.51191.02381.53572.04762.5595SE +/- 0.001131, N = 3SE +/- 0.000385, N = 3SE +/- 0.013699, N = 3SE +/- 0.003732, N = 3SE +/- 0.003163, N = 3SE +/- 0.001856, N = 3SE +/- 0.003885, N = 3SE +/- 0.001182, N = 3SE +/- 0.005784, N = 3SE +/- 0.000323, N = 3SE +/- 0.002172, N = 3SE +/- 0.005714, N = 3SE +/- 0.006027, N = 3SE +/- 0.018012, N = 30.8510070.8525181.0219901.1360801.2512701.2529401.2639501.2664901.3409601.9627902.1658102.2005302.2047102.275260MIN: 0.83MIN: 0.83MIN: 0.99MIN: 1.1MIN: 1.22MIN: 1.22MIN: 1.24MIN: 1.24MIN: 1.3MIN: 1.95MIN: 2.07MIN: 2.17MIN: 2.14MIN: 2.121. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm8192.in input file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian Hydrodynamics2 x Xeon Platinum 8375C NUMA off2 x Xeon Platinum 8375CEPYC 2 7R32Xeon Platinum 8480+Xeon Platinum 8375C2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 82802 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CLXeon Platinum 8280Xeon Platinum 8272CLXeon Platinum 8259CLXeon W-3275M0.25650.5130.76951.0261.2825SE +/- 0.01, N = 3SE +/- 0.01, N = 15SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 15SE +/- 0.02, N = 15SE +/- 0.03, N = 15SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 150.330.350.360.460.510.610.630.690.710.780.961.051.111.141. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

Nettle

GNU Nettle is a low-level cryptographic library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.5.1Test: chachaXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CLXeon Platinum 8272CLXeon Platinum 8480+Xeon Platinum 8259CL2 x Xeon Platinum 8259CL2 x Xeon Platinum 8259CL NUMA offXeon Platinum 8375C2 x Xeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offEPYC 2 7R3230060090012001500SE +/- 5.99, N = 3SE +/- 0.37, N = 3SE +/- 0.86, N = 3SE +/- 0.28, N = 3SE +/- 0.37, N = 3SE +/- 2.25, N = 3SE +/- 0.16, N = 3SE +/- 1.52, N = 3SE +/- 4.50, N = 3SE +/- 4.18, N = 3SE +/- 9.77, N = 3SE +/- 8.54, N = 3SE +/- 6.97, N = 3SE +/- 0.08, N = 31172.151051.941051.11975.08974.46970.38968.52920.51917.75917.51822.11821.18815.18750.36MIN: 567.45 / MAX: 3234.29MIN: 535.04 / MAX: 2870.94MIN: 532.77 / MAX: 2871.25MIN: 495.02 / MAX: 2662.01MIN: 494.65 / MAX: 2659.52MIN: 491.16 / MAX: 2661.13MIN: 476.58 / MAX: 2735.75MIN: 461.1 / MAX: 2517.2MIN: 461.99 / MAX: 2517.41MIN: 463.47 / MAX: 2517.16MIN: 419.2 / MAX: 2294.54MIN: 418.33 / MAX: 2284.8MIN: 418.76 / MAX: 2286.72MIN: 382.32 / MAX: 2047.211. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto

OpenBenchmarking.orgMbyte/s, More Is BetterNettle 3.5.1Test: poly1305-aesXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 8280Xeon Platinum 8272CL2 x Xeon Platinum 8272CL NUMA off2 x Xeon Platinum 8272CL2 x Xeon Platinum 8259CL NUMA off2 x Xeon Platinum 8259CLXeon Platinum 8259CLEPYC 2 7R322 x Xeon Platinum 8375CXeon Platinum 8375C2 x Xeon Platinum 8375C NUMA offXeon Platinum 8480+7001400210028003500SE +/- 24.77, N = 3SE +/- 1.10, N = 3SE +/- 0.87, N = 3SE +/- 12.97, N = 3SE +/- 2.42, N = 3SE +/- 3.72, N = 3SE +/- 0.36, N = 3SE +/- 15.23, N = 3SE +/- 12.50, N = 3SE +/- 0.71, N = 3SE +/- 35.93, N = 15SE +/- 2.47, N = 3SE +/- 2.03, N = 3SE +/- 0.09, N = 33207.982868.692867.482661.612657.682654.232516.872497.862492.482046.351963.031909.601909.431439.351. (CC) gcc options: -O2 -ggdb3 -lnettle -lgmp -lm -lcrypto

119 Results Shown

Timed GCC Compilation
MKL-DNN DNNL:
  Convolution Batch conv_all - bf16bf16bf16
  Convolution Batch conv_all - u8s8f32
  Convolution Batch conv_all - f32
PostgreSQL pgbench
MKL-DNN DNNL:
  Deconvolution Batch deconv_all - bf16bf16bf16
  Deconvolution Batch deconv_all - f32
PostgreSQL pgbench
Blender
Radiance Benchmark
Timed LLVM Compilation
Numpy Benchmark
Renaissance
High Performance Conjugate Gradient
Mlpack Benchmark
Blender
MKL-DNN DNNL
Mlpack Benchmark
MKL-DNN DNNL
asmFish
Timed MrBayes Analysis
Timed GDB GNU Debugger Compilation
Blender
dav1d
MKL-DNN DNNL
LuxCoreRender
Appleseed
OSPray
Timed Linux Kernel Compilation
DeepSpeech
OSPray
Appleseed
PyPerformance
OSPray
Mlpack Benchmark
Blender
OSPray
Renaissance
Radiance Benchmark
Chaos Group V-RAY
PyPerformance
Build2
Renaissance
LuxCoreRender
Stockfish
John The Ripper
Appleseed
PyPerformance
Basis Universal
Blender
MKL-DNN DNNL
Himeno Benchmark
NAMD
MKL-DNN DNNL:
  IP Batch All - u8s8f32
  IP Batch All - f32
NAS Parallel Benchmarks
OSPray
NAS Parallel Benchmarks
SVT-AV1
PyPerformance
MKL-DNN DNNL
Embree
John The Ripper
Embree
Tungsten Renderer
MKL-DNN DNNL
Zstd Compression
lzbench:
  Crush 0 - Decompression
  Crush 0 - Compression
Intel MPI Benchmarks
Renaissance
NAS Parallel Benchmarks
Basis Universal
lzbench:
  Zstd 8 - Decompression
  Zstd 8 - Compression
PyBench
lzbench:
  Brotli 2 - Decompression
  Brotli 2 - Compression
  Zstd 1 - Decompression
  Zstd 1 - Compression
Mlpack Benchmark
MKL-DNN DNNL:
  Deconvolution Batch deconv_1d - u8s8f32
  Deconvolution Batch deconv_1d - bf16bf16bf16
OSPray
MKL-DNN DNNL:
  Deconvolution Batch deconv_1d - f32
  Convolution Batch conv_alexnet - bf16bf16bf16
ArrayFire
SVT-AV1
Embree
MKL-DNN DNNL
dav1d
Embree
MKL-DNN DNNL
NAS Parallel Benchmarks
Embree
PHPBench
MKL-DNN DNNL
dav1d
MKL-DNN DNNL
Embree
Scikit-Learn
SVT-VP9
ArrayFire
OSPray
SMHasher:
  wyhash:
    MiB/sec
    cycles/hash
Tungsten Renderer
DaCapo Benchmark
LAMMPS Molecular Dynamics Simulator
Intel Open Image Denoise
NAS Parallel Benchmarks
MKL-DNN DNNL
NAS Parallel Benchmarks
Nettle:
  aes256
  sha512
MKL-DNN DNNL
CloverLeaf
Nettle:
  chacha
  poly1305-aes