Xeon Tests

AMD EPYC 7R32 48-Core testing with a Supermicro H11SSL-i v2.00 (2.1 BIOS) and llvmpipe 126GB on Ubuntu 19.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2008226-VE-2004196VE87
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

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

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
Triggered
  Test
  Duration
2 x Xeon Gold 5220R
March 26 2020
  9 Hours, 36 Minutes
Xeon Gold 5220R
March 27 2020
  10 Hours, 11 Minutes
Xeon Platinum 8280
March 28 2020
  8 Hours, 39 Minutes
2 x Xeon Platinum 8280
March 28 2020
  8 Hours, 49 Minutes
Xeon Gold 5218
March 29 2020
  11 Hours, 15 Minutes
Xeon Silver 4216
March 30 2020
  11 Hours, 48 Minutes
Xeon Gold 6226R
March 30 2020
  9 Hours, 51 Minutes
2 x Xeon Gold 6226R
March 31 2020
  9 Hours, 29 Minutes
2 x Xeon Silver 4216
March 31 2020
  10 Hours, 35 Minutes
Xeon W-3275M
April 19 2020
  9 Hours, 52 Minutes
2 x Xeon Platinum 8275CL
June 02 2020
  11 Hours, 15 Minutes
2 x Xeon Platinum 8275CL NUMA off
June 03 2020
  9 Hours, 44 Minutes
EPYC 2 7R32
August 22 2020
  7 Hours, 24 Minutes
Invert Hiding All Results Option
  9 Hours, 53 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):


Xeon TestsProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkAudioOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionOpenGL2 x Xeon Gold 5220RXeon Gold 5220RXeon Platinum 82802 x Xeon Platinum 8280Xeon Gold 5218Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads)GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS)Intel Sky Lake-E DMI3 Registers12 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 9.3.0ext41920x1080Intel Xeon Gold 5220R @ 3.90GHz (18 Cores / 36 Threads)6 x 32 GB DDR4-2933MT/s HMA84GR7CJR4N-WM1024x768Intel Xeon Platinum 8280 @ 4.00GHz (28 Cores / 56 Threads)2 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads)12 x 32 GB DDR4-2933MT/s HMA84GR7CJR4N-WMIntel Xeon Gold 5218 @ 3.90GHz (16 Cores / 32 Threads)6 x 32 GB DDR4-2933MT/s HMA84GR7CJR4N-WM2 x VE2281920x1080Intel Xeon Silver 4216 @ 3.20GHz (16 Cores / 32 Threads)3 x VE228Intel Xeon Gold 6226R @ 3.90GHz (16 Cores / 32 Threads)4 x VE2282 x Intel Xeon Gold 6226R @ 3.90GHz (32 Cores / 64 Threads)12 x 32 GB DDR4-2933MT/s HMA84GR7CJR4N-WM5 x VE2282 x Intel Xeon Silver 4216 @ 3.20GHz (32 Cores / 64 Threads)6 x VE228Intel Xeon W-3275M @ 4.60GHz (28 Cores / 56 Threads)Supermicro X11SPA-T v1.01 (3.1 BIOS)188GB2 x 1024GB Samsung SSD 970 PRO 1TBeVGA NVIDIA GeForce RTX 2080 Ti 11GB (1350/7000MHz)Realtek ALC888-VDBenQ SW271Intel I210 + Aquantia AQC107 NBase-T/IEEEUbuntu 19.105.3.0-46-generic (x86_64)GNOME Shell 3.34.3X Server 1.20.5NVIDIA 430.504.6.0GCC 9.2.1 20191008 + CUDA 10.13840x21602 x Intel Xeon Platinum 8275CL @ 3.90GHz (48 Cores / 96 Threads)ASRockRack EP2C621D12 WS (P2.10 BIOS)32 GB + 32 GB + 16384 MB + 32 GB + 16384 MB + 32 GB + 32 GB + 32 GB DDR4-3200MT/s256GB TOSHIBA THNSNJ25 + 600GB INTEL SSDSC2BB60 + 6001GB HGST HUS726060ALllvmpipe 220GBCreative Labs EMU20k2DELL UP2516D4 x Intel X722 for 1GbE5.3.0-55-generic (x86_64)GNOME Shell 3.34.1X Server3.3 Mesa 19.2.1 (LLVM 9.0 256 bits)GCC 9.2.1 201910082560x14403.3 Mesa 19.2.8 (LLVM 9.0 256 bits)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 I2105.3.0-64-generic (x86_64)3.3 Mesa 19.2.8 (LLVM 9.0 128 bits)800x600OpenBenchmarking.orgCompiler Details- 2 x Xeon Gold 5220R: --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 Gold 5220R: --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 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 Gold 5218: --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 Silver 4216: --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 Gold 6226R: --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 Gold 6226R: --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 Silver 4216: --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 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- 2 x Xeon Platinum 8275CL: --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 8275CL 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- 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 -vProcessor Details- 2 x Xeon Gold 5220R: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- Xeon Gold 5220R: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- 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 Gold 5218: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- Xeon Silver 4216: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- Xeon Gold 6226R: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- 2 x Xeon Gold 6226R: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- 2 x Xeon Silver 4216: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- Xeon W-3275M: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- 2 x Xeon Platinum 8275CL: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002b- 2 x Xeon Platinum 8275CL NUMA off: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c- EPYC 2 7R32: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301034Java Details- 2 x Xeon Gold 5220R: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- Xeon Gold 5220R: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- 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 Gold 5218: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- Xeon Silver 4216: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- Xeon Gold 6226R: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- 2 x Xeon Gold 6226R: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- 2 x Xeon Silver 4216: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- Xeon W-3275M: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-1ubuntu119.10.1)- 2 x Xeon Platinum 8275CL: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- 2 x Xeon Platinum 8275CL NUMA off: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)- EPYC 2 7R32: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-2ubuntu219.10)Python Details- 2 x Xeon Gold 5220R: + Python 3.8.2- Xeon Gold 5220R: + Python 3.8.2- Xeon Platinum 8280: + Python 3.8.2- 2 x Xeon Platinum 8280: + Python 3.8.2- Xeon Gold 5218: + Python 3.8.2- Xeon Silver 4216: + Python 3.8.2- Xeon Gold 6226R: + Python 3.8.2- 2 x Xeon Gold 6226R: + Python 3.8.2- 2 x Xeon Silver 4216: + Python 3.8.2- Xeon W-3275M: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8275CL: Python 2.7.17 + Python 3.7.5- 2 x Xeon Platinum 8275CL NUMA off: Python 2.7.17 + Python 3.7.5- EPYC 2 7R32: Python 2.7.17 + Python 3.7.5Security Details- 2 x Xeon Gold 5220R: 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 Gold 5220R: 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 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 Gold 5218: 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 Silver 4216: 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 Gold 6226R: 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 Gold 6226R: 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 Silver 4216: 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 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- 2 x Xeon Platinum 8275CL: 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 8275CL 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 + tsx_async_abort: Mitigation of TSX disabled- 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 affectedOpenCL Details- Xeon W-3275M: GPU Compute Cores: 4352

2 x Xeon Gold 5220RXeon Gold 5220RXeon Platinum 82802 x Xeon Platinum 8280Xeon Gold 5218Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R32Logarithmic Result OverviewPhoronix Test Suite 10.4.0m1MKL-DNN DNNLArrayFireOSPrayHigh Performance Conjugate GradientLuxCoreRenderJohn The RipperEmbreeIntel MPI BenchmarksStockfishasmFishChaos Group V-RAYBlenderPostgreSQL pgbenchLAMMPS Molecular Dynamics SimulatorTimed Linux Kernel Compilationdav1dIntel Open Image DenoiseTimed LLVM CompilationSVT-AV1Timed MrBayes AnalysisTungsten RendererAppleseedBasis UniversalDeepSpeechBuild2SVT-VP9Timed GCC CompilationTimed GDB GNU Debugger CompilationMlpack BenchmarkPHPBenchNettlePyBenchNumpy BenchmarklzbenchZstd CompressionRenaissanceScikit-LearnHimeno BenchmarkRadiance BenchmarkDaCapo Benchmark

Xeon Testsmkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8f32arrayfire: Conjugate Gradient CPUarrayfire: BLAS CPUmkl-dnn: Convolution Batch conv_alexnet - f32mkl-dnn: Convolution Batch conv_all - u8s8f32mkl-dnn: Convolution Batch conv_googlenet_v3 - f32mkl-dnn: Convolution Batch conv_alexnet - u8s8f32mkl-dnn: Deconvolution Batch deconv_3d - f32mkl-dnn: Convolution Batch conv_all - f32mkl-dnn: Convolution Batch conv_alexnet - bf16bf16bf16mkl-dnn: Convolution Batch conv_all - bf16bf16bf16mkl-dnn: Convolution Batch conv_googlenet_v3 - bf16bf16bf16ospray: NASA Streamlines - Path Tracerospray: XFrog Forest - SciVisospray: XFrog Forest - Path Tracermkl-dnn: Convolution Batch conv_3d - bf16bf16bf16ospray: NASA Streamlines - SciVismkl-dnn: IP Batch All - u8s8f32mkl-dnn: Deconvolution Batch deconv_3d - bf16bf16bf16ospray: San Miguel - SciVisembree: Pathtracer ISPC - Crownospray: San Miguel - Path Tracerhpcg: embree: Pathtracer ISPC - Asian Dragonembree: Pathtracer ISPC - Asian Dragon Objluxcorerender: Rainbow Colors and Prismblender: Classroom - CPU-Onlymkl-dnn: Deconvolution Batch deconv_1d - f32namd: ATPase Simulation - 327,506 Atomsluxcorerender: DLSCmkl-dnn: Deconvolution Batch deconv_1d - bf16bf16bf16john-the-ripper: MD5blender: Pabellon Barcelona - CPU-Onlyjohn-the-ripper: Blowfishmkl-dnn: Convolution Batch conv_3d - u8s8f32intel-mpi: IMB-P2P PingPongmkl-dnn: Deconvolution Batch deconv_1d - u8s8f32npb: EP.Dospray: Magnetic Reconnection - SciVisnpb: EP.Cmkl-dnn: Deconvolution Batch deconv_all - bf16bf16bf16mkl-dnn: Deconvolution Batch deconv_3d - u8s8f32blender: BMW27 - CPU-Onlyembree: Pathtracer - Crownstockfish: Total Timeasmfish: 1024 Hash Memory, 26 Depthv-ray: CPUembree: Pathtracer - Asian Dragonembree: Pathtracer - Asian Dragon Objblender: Fishy Cat - CPU-Onlymkl-dnn: IP Batch 1D - bf16bf16bf16npb: SP.Bblender: Barbershop - CPU-Onlylammps: Rhodopsin Proteinmkl-dnn: IP Batch 1D - f32appleseed: Disney Materialmkl-dnn: Deconvolution Batch deconv_all - f32tungsten: Hairbasis: UASTC Level 3npb: LU.Cbuild-linux-kernel: Time To Compilenpb: BT.Cdav1d: Summer Nature 1080ppgbench: Buffer Test - Normal Load - Read Onlymkl-dnn: IP Batch All - f32dav1d: Chimera 1080p 10-bitoidn: Memorialbuild-llvm: Time To Compilenpb: MG.Cdav1d: Summer Nature 4Kappleseed: Emilysvt-av1: Enc Mode 8 - 1080psvt-av1: Enc Mode 4 - 1080pmrbayes: Primate Phylogeny Analysismlpack: scikit_qdapgbench: Buffer Test - Normal Load - Read Writerenaissance: Apache Spark Bayesmlpack: scikit_icadeepspeech: CPUbuild2: Time To Compilemlpack: scikit_linearridgeregressionsvt-vp9: Visual Quality Optimized - Bosphorus 1080pbuild-gcc: Time To Compilebuild-gdb: Time To Compilepyperformance: python_startupappleseed: Material Testernettle: poly1305-aesnettle: chachanettle: aes256lzbench: Crush 0 - Decompressionphpbench: PHP Benchmark Suitepyperformance: floatpyperformance: django_templatelzbench: Brotli 2 - Decompressionbasis: ETC1Slzbench: Zstd 1 - Compressionpybench: Total For Average Test Timessmhasher: wyhashsmhasher: wyhashnumpy: lzbench: Zstd 8 - Decompressioncompress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19mlpack: scikit_svmlzbench: Brotli 2 - Compressionlzbench: Zstd 1 - Decompressionnettle: sha512scikit-learn: himeno: Poisson Pressure Solverradiance: Serialradiance: SMP Parallellzbench: Crush 0 - Compressionrenaissance: Apache Spark PageRankpyperformance: regex_compiledacapobench: Jythontungsten: Water Causticlzbench: Zstd 8 - Compressionrenaissance: Scala Dottycloverleaf: Lagrangian-Eulerian Hydrodynamicsrenaissance: Savina Reactors.IO2 x Xeon Gold 5220RXeon Gold 5220RXeon Platinum 82802 x Xeon Platinum 8280Xeon Gold 5218Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R3215.75429.8982359.13122.7612479.2854.819035.97782.69700999.544712.2013917.52187.16810.427.193.9815.217755.564.086059.4259143.4833.44693.9525.630343.098337.14984.46165.962.126800.603334.447.229754821000208.30513244896.4119650424.3333330.5608903780.5947.623768.663152.413112.3660.6729.515374235829855018514027035.875732.330198.245.7279150932.88260.5617.3782.2300988.8566341278.2511.072627.981108883.4036.699100761.29321.84598798.57132410.634870.8127.97286.78258984.18177.89180.64579749.2095.302134.92842.6934227.7249033662.76777.41125.1186372.1512.39237.571060.578146.43911.2189.4596532797.221024.9756159.2141963128311347.361354.554437110018.806300.73110110.44917.921721147489.1510.4583404.690861746.918230.5619324540.720167487723.3367747414.6191.0626257.00828.45326.4091201.54244.7454826.71108.16166.73524.889661948.771433.477839.95369.2875.383.702.0629.917328.574.6105416.825323.2617.66042.1413.352923.604220.34182.48328.813.364321.157452.3213.14282506000404.78255479696.9210276849.5217390.8437091966.6127.031863.155695.586072.57113.3016.051438167326432179392188619.610517.7957175.958.6859023346.56456.479.5354.49985158.772381920.7418.984046.70856744.8759.01250787.79274.98383120.6905098.9079558.0518.41430.42631197.22160.59284.2797433.0884.285134.48334.2445518.5935153150.09668.19126.0780191.6412.25224.011167.683151.90911.3162.8603532779.781020.0936133.9142862969011447.165156.018444111418.815313.76119410.35517.821761180485.8610.1123398.397455749.923242.5159821704.485168483425.0080767556.3291.2822776.29511.29004.1153432.1694.19372846.1242.075826.83911.96279765.745637.3743497.47166.9849.496.253.4614.0780503.231217.6960538.4628.77303.5214.909738.310033.04904.11198.251.462320.707093.796.553744090000246.31428645919.9416653864.7681160.3710363283.3842.873194.182935.993716.9869.8026.280062891312687620753471032.247429.2061110.674.8317032682.14287.8015.3142.39451103.8330991113.5412.686330.68974255.6740.89871842.34323.20560887.7666966.7950965.4031.10308.74137089.05199.08197.64950551.9176.030128.84830.4553139.0577772674.17567.36118.0937774.1181.67265.401063.705144.64311.2123.9791032867.481051.1116313.1644064916410946.267351.446457107622.678330.3312248.20317.191811212501.989.1273510.205159728.36123510120621.229164454722.0167776664.2540.9621975.1227.460507.4813430.2849.84371517.4423.177219.81431.13608496.816320.0601766.3586.539118.1812.056.677.9291290.913.110024.3581769.5253.23046.5328.854268.424158.69457.15100.471.240290.362267.194.068477880000129.01840843005.4831821690.5507250.3944556284.6571.436104.941779.151893.7238.7945.61211197157951338398966433157.487051.785366.393.6596470893.41187.3324.7881.6398668.618272881.7837.8270619.885140090.3227.309136046.53409.88909916.4900869.5246685.1434.49221.51569175.12221.31154.98075256.9827.033129.98638.3524466.9112273228.63363.35122.2246862.7761.82233.171005.530142.69511.2193.442412868.691051.9366320.0042864969310945.962352.694449107322.677321.8211249.13717.191761168500.989.3923514.604106728.246226.0769623588.397164471121.5896756890.6060.6926764.94134.60317.0421002.09298.3005840.26131.08881.07055.804202363.001874.4710255.2484.5944.373.011.6839.998523.264.8387521.429018.8714.41631.7513.013819.432116.76851.88433.423.994881.511681.7416.54841912667531.982140311821.39066471.04347830.9767301635.7822.221621.486721.027189.03149.3912.388232424332365618171698115.310413.8526229.288.827122329.34589.078.1584.75696182.5579721699.6922.486655.20948143.4967.15648617.05265.46322051.6311879.3095955.8815.80501.21830328.65136.86328.13913327.1933.339147.73437.4140894.8967563477.33168.78128.40692101.7162.46172.821218.874156.29911.1179.7515052798.561024.4936164.7943063245511447.265558.473446110119.806315.14119511.12217.981761184489.5110.2863405.035897747.844241.8979623703.029167509625.9635748286.1231.3727198.08045.05027.819765.491395.4416241.16171.428105.8267.636463129.931875.2310258.0484.8953.412.341.3140.019218.185.5927521.475514.7111.38271.4011.888315.591113.49181.88432.115.310871.517211.7716.55161893333531.392042412300.38923779.55072461.273751571.4917.941545.506721.547457.86149.0212.431631174576350854961710515.279613.8827229.7110.901421149.68593.137.8105.16658194.4018432201.0423.612958.24646723.0171.77046389.79232.35313376.77991710.193050.1012.85531.15228301.13132.46345.38996126.4603.309193.46640.7540373.3048943588.69281.36135.99744109.6992.97173.181398.020177.77215.4188.7767562291.69837.6125024.3935351569713757.554267.461365134222.13122004.08272.1299811.43020.89146973399.9912.0302823.595364909.318299.7398225271.433204576828.8979648634.6911.4922798.38217.97356.4512137.03155.3684478.9369.143142.22023.130061227.071042.345709.17269.9915.523.682.0222.424329.414.5382112.327122.6916.80992.0213.958122.459819.32282.51321.532.231261.135862.359.903622583000400.96271269374.6010620968.6086950.5286322094.5325.482044.243966.735864.26111.8215.994339776292446149252207419.347217.4223173.955.7143923859.79453.2210.3254.41366155.228014983.91318.612944.94358594.5057.77256576.12307.75376860.3186158.8423362.3222.01433.19131633.6170.07276.76865434.6364.340123.70231.9646503.2830402856.85061.82119.7871788.9771.93221.951092.587139.12011.0156.6056142795.491024.9926161.4242963243111547.365354.790445110325.13419429.87318.87119310.33017.921761182488.819.9243401.157135746.687242.0029521743.019168460924.0874746999.4971.2727198.85611.32827.1553912.3081.14002377.9136.371626.53631.70749694.324527.0172871.11137.73410.567.093.8811.607655.564.117796.2964742.2731.73183.7126.591040.432534.80304.32163.321.497240.582374.51835.384024977667207.41534854821.2420467429.5217390.3690603969.6045.454005.262169.932987.0859.4528.960275203729838945143474934.691931.311796.413.2128149232.24257.0118.1502.1897584.844108711.38910.665327.226106268.3336.59577470.67358.62587882.23269910.523973.9533.06279.68258681.42203.10177.66114851.7105.431124.25844.5337715.2162713405.41474.63123.3870270.6082.08256.91996.904133.44410.9185.3571542797.431025.3846157.4941863260111347.160854.947437110425.12919424.11304.16110010.20617.771721141489.4210.4493410.383938745.686233.0429124522.220167480623.1916727231.0140.9327972.99823.702816.921497.75197.9143215.1787.543455.33613.855121578.24938.9105156.69246.4466.624.572.5520.516334.484.8260310.937927.8421.74682.6323.157428.973225.06783.65218.063.065900.769823.438.869133764000272.72410216329.9417176230.5362320.7703003144.7032.532984.023600.953849.1378.7223.267460569696668958102759228.203225.4971125.355.5650042220.94333.4014.1612.66699107.4980891364.2613.603134.63285606.5345.31565509.98272.55501697.23823111.421559.5120.43347.82752832.47159.60222.96851440.5484.216194.50450.1034916.0308524199.41194.45133.0116086.3252.99206.421268.815169.70415.4197.7404422294.74841.6495061.6034451996413757.550965.644359134322.12922011.20262.7893010.97821.09142943401.1312.5492817.673053907.757281.2518028045.920204580227.3948638583.0141.0028596.10513.45044.4233128.39108.5682967.9249.874431.98932.27526878.390702.0653880.44188.6118.695.723.1716.590645.453.592188.4627334.4826.40103.2514.162335.149530.34583.79213.361.745150.814343.537.615953654000265.41403266183.1120246604.6231880.4584223210.52403025.113293.943852.3774.9624.550361170651662256413301430.105727.1952118.095.5630431426.53304.7914.5622.86715106.6433921292.6212.710430.63469606.8436.36769347.58537.84567696.5059127.5443981.6027.35281.67835376.11256.60196.92978653.8196.389143.34832.6511704.2318042743.62354.91123.8525266.4151.99282.18994.583134.82417.7124.1473013207.981172.1537070.6050973433994.549.972147.65350895118.50826298.77367.5612988.38015.781961299550.939.6613866.090988665.7219.83610921015.543157432121.5517846970.9041.1425186.66617.754913.226259.0179.11092023.4733.630559.90401.173851206.22348.2431919.5093.359617.0211.366.258.7123482.625.367084.3795263.5449.04615.828.4140538.765836.66527.09108.971.331096.733.969067388333139.11777713250.8935954425.3623190.35536250.881730.802059.0341.6841.78991098743071195499015832640.078735.667669.934.10369189.5522.3011.9599268.606542981.0588.2906721.06027.100422.89629984.51425319.103090.2815.06218.050218.14152.90204955.3745.243125.86762.023890.7708825185.62153.50113.1439664.5032.36191.66966.599128.950189.2961662785.991025.6036172.5543464017359252.608439113326.08618770.15338.44105310.94516.871701099487.5511.0653440.518277753.635232.6959424505.962468522.7599736733.16026906.88610.48419.6546325.0757.51111708.0525.053330.64271.17278650.981348.3261918.2893.284216.9511.366.248.6470783.333.847464.3600266.6749.95595.876.3297151.874144.71517.31108.941.251166.753.965107350667139.00783273252.4835888907.9855070.35471458.821705.352057.9341.7241.39441084125141030339555852545.767440.850270.044.06041190.0120.9841.6359968.202167910.1428.1933720.99429.258406.66725740.72379512.682090.7819.55244.910216.86152.09368156.1225.942124.29757.253197.7429544510.83053.66138.2562964.4212.46228.381032.978145.510194.8468612803.551027.5926184.2043263902658253.135443112326.12918775.94315.6699910.28918.421721090490.2611.4903430.766164753.433227.1599524666.363475922.7088757042.35427361.597730.30635.13990.79896.688111728.842.54131301.362.20471710.14016.6011.035.8576.92149.33958.8245.18694.7317.060852.354145.43308.12101.342.225400.486147.283972000132.04664753076.7620348261.0289851131.133570.46403485.991899.2441.7647.64641167650281132789776289752.953247.580665.6751225.18177.7920.8211.6756062.4310471485.477.6745519.19399821.8924.005106415.53676.16891303.1219438.88390140.1428.71200.03949330.15321.17147.68585654.4725.71492.36139.546105.7289573057.05949.4772.8159959.9561.79291.17849.192108.67516.3175.1937082046.35750.3574665.3338050052466.456454.3104181226323.5812669.23015.161651148478.0910.4193769.723664730.7219.6528121641.954213524622.3247836577.3430.3619303.561OpenBenchmarking.org

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: u8s8f32EPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Gold 6226RXeon Platinum 82802 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280160320480640800SE +/- 7.24186, N = 3SE +/- 0.02504, N = 3SE +/- 0.01482, N = 3SE +/- 0.02138, N = 3SE +/- 0.01146, N = 3SE +/- 0.01762, N = 3SE +/- 0.29660, N = 3SE +/- 0.00643, N = 3SE +/- 0.01662, N = 3SE +/- 0.01218, N = 3SE +/- 0.01386, N = 3SE +/- 0.02611, N = 3SE +/- 0.03597, N = 3730.3060045.0502034.6031028.4532023.7028017.9735017.7549015.7542013.4504011.3282011.2900010.484107.46050MIN: 698.321. (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_googlenet_v3 - Data Type: u8s8f32EPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Gold 6226RXeon Platinum 82802 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280130260390520650Min: 717.38 / Avg: 730.31 / Max: 742.42Min: 45.01 / Avg: 45.05 / Max: 45.1Min: 34.58 / Avg: 34.6 / Max: 34.63Min: 28.41 / Avg: 28.45 / Max: 28.48Min: 23.68 / Avg: 23.7 / Max: 23.72Min: 17.95 / Avg: 17.97 / Max: 18.01Min: 17.22 / Avg: 17.75 / Max: 18.24Min: 15.75 / Avg: 15.75 / Max: 15.77Min: 13.43 / Avg: 13.45 / Max: 13.48Min: 11.31 / Avg: 11.33 / Max: 11.35Min: 11.27 / Avg: 11.29 / Max: 11.32Min: 10.45 / Avg: 10.48 / Max: 10.54Min: 7.4 / Avg: 7.46 / Max: 7.521. (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.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient CPUEPYC 2 7R322 x Xeon Silver 42162 x Xeon Platinum 8275CL2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offXeon Silver 42162 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon W-3275MXeon Platinum 8280816243240SE +/- 0.217, N = 3SE +/- 0.372, N = 15SE +/- 0.220, N = 3SE +/- 0.099, N = 15SE +/- 0.272, N = 15SE +/- 0.047, N = 3SE +/- 0.086, N = 6SE +/- 0.072, N = 15SE +/- 0.069, N = 9SE +/- 0.026, N = 3SE +/- 0.094, N = 3SE +/- 0.075, N = 15SE +/- 0.067, N = 335.13016.92013.2209.8989.6547.8197.4817.1557.0426.4516.4094.4234.1151. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgms x Core, Fewer Is BetterArrayFire 3.7Performance Per Core - Test: Conjugate Gradient CPUEPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Silver 42162 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226RXeon Silver 4216Xeon W-3275MXeon Gold 5220RXeon Platinum 8280Xeon Gold 5218Xeon Gold 6226R4008001200160020001686.24634.56541.44463.39418.94356.33228.96125.10123.84115.36115.22112.67103.221. EPYC 2 7R32: Detected core count of 482. 2 x Xeon Platinum 8275CL: Detected core count of 483. 2 x Xeon Silver 4216: Detected core count of 324. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 485. 2 x Xeon Platinum 8280: Detected core count of 566. 2 x Xeon Gold 5220R: Detected core count of 367. 2 x Xeon Gold 6226R: Detected core count of 328. Xeon Silver 4216: Detected core count of 169. Xeon W-3275M: Detected core count of 2810. Xeon Gold 5220R: Detected core count of 1811. Xeon Platinum 8280: Detected core count of 2812. Xeon Gold 5218: Detected core count of 1613. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgms x Thread, Fewer Is BetterArrayFire 3.7Performance Per Thread - Test: Conjugate Gradient CPUEPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Silver 42162 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226RXeon Silver 4216Xeon W-3275MXeon Gold 5220RXeon Platinum 8280Xeon Gold 5218Xeon Gold 6226R70014002100280035003372.481269.121082.88926.78837.87712.66457.92250.21247.69230.72230.44225.34206.431. EPYC 2 7R32: Detected thread count of 962. 2 x Xeon Platinum 8275CL: Detected thread count of 963. 2 x Xeon Silver 4216: Detected thread count of 644. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 965. 2 x Xeon Platinum 8280: Detected thread count of 1126. 2 x Xeon Gold 5220R: Detected thread count of 727. 2 x Xeon Gold 6226R: Detected thread count of 648. Xeon Silver 4216: Detected thread count of 329. Xeon W-3275M: Detected thread count of 5610. Xeon Gold 5220R: Detected thread count of 3611. Xeon Platinum 8280: Detected thread count of 5612. Xeon Gold 5218: Detected thread count of 3213. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgms x GHz, Fewer Is BetterArrayFire 3.7Performance Per Clock - Test: Conjugate Gradient CPUEPYC 2 7R322 x Xeon Silver 42162 x Xeon Platinum 8275CL2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Gold 5218Xeon Gold 6226RXeon Silver 4216Xeon Gold 5220RXeon W-3275MXeon Platinum 82802040608010098.3654.1451.5638.6037.6529.9227.9127.4625.1625.0225.0020.3516.461. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.802. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 3. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 6. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 7. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 8. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 11. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 13. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00
OpenBenchmarking.orgms x Channel, Fewer Is BetterArrayFire 3.7Performance Per Memory Channel - Test: Conjugate Gradient CPUEPYC 2 7R322 x Xeon Silver 42162 x Xeon Gold 5220RXeon Silver 42162 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon Platinum 828060120180240300281.04101.5259.3946.9144.8942.9342.2538.7138.4524.691. EPYC 2 7R32: Detected channel count of 82. 2 x Xeon Silver 4216: Detected channel count of 63. 2 x Xeon Gold 5220R: Detected channel count of 64. Xeon Silver 4216: Detected channel count of 65. 2 x Xeon Platinum 8280: Detected channel count of 66. 2 x Xeon Gold 6226R: Detected channel count of 67. Xeon Gold 5218: Detected channel count of 68. Xeon Gold 6226R: Detected channel count of 69. Xeon Gold 5220R: Detected channel count of 610. Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgms, Fewer Is BetterArrayFire 3.7Test: Conjugate Gradient CPUEPYC 2 7R322 x Xeon Silver 42162 x Xeon Platinum 8275CL2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offXeon Silver 42162 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon W-3275MXeon Platinum 8280816243240Min: 34.75 / Avg: 35.13 / Max: 35.5Min: 13.72 / Avg: 16.92 / Max: 18.32Min: 12.78 / Avg: 13.22 / Max: 13.45Min: 9.29 / Avg: 9.9 / Max: 10.81Min: 7.39 / Avg: 9.65 / Max: 11.48Min: 7.73 / Avg: 7.82 / Max: 7.87Min: 7.21 / Avg: 7.48 / Max: 7.81Min: 6.54 / Avg: 7.16 / Max: 7.52Min: 6.91 / Avg: 7.04 / Max: 7.56Min: 6.42 / Avg: 6.45 / Max: 6.5Min: 6.29 / Avg: 6.41 / Max: 6.59Min: 4.12 / Avg: 4.42 / Max: 5.06Min: 4 / Avg: 4.12 / Max: 4.231. (CXX) g++ options: -rdynamic

OpenBenchmarking.orgGFLOPS, More Is BetterArrayFire 3.7Test: BLAS CPUXeon Silver 4216EPYC 2 7R32Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off14002800420056007000SE +/- 1.52, N = 3SE +/- 6.50, N = 3SE +/- 0.62, N = 3SE +/- 0.37, N = 3SE +/- 0.79, N = 3SE +/- 4.16, N = 3SE +/- 3.91, N = 3SE +/- 30.40, N = 3SE +/- 20.06, N = 3SE +/- 25.11, N = 3SE +/- 31.78, N = 3SE +/- 61.47, N = 3SE +/- 4.90, N = 3765.49990.801002.091201.541497.752137.032359.133128.393430.283432.163912.306259.016325.071. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgGFLOPS Per Core, More Is BetterArrayFire 3.7Performance Per Core - Test: BLAS CPUEPYC 2 7R322 x Xeon Silver 4216Xeon Silver 42162 x Xeon Platinum 8280Xeon Gold 52182 x Xeon Gold 5220RXeon Gold 5220RXeon W-3275M2 x Xeon Gold 6226RXeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offXeon Gold 6226R30609012015020.6446.8047.8461.2662.6365.5366.75111.73122.26122.58130.40131.77133.561. EPYC 2 7R32: Detected core count of 482. 2 x Xeon Silver 4216: Detected core count of 323. Xeon Silver 4216: Detected core count of 164. 2 x Xeon Platinum 8280: Detected core count of 565. Xeon Gold 5218: Detected core count of 166. 2 x Xeon Gold 5220R: Detected core count of 367. Xeon Gold 5220R: Detected core count of 188. Xeon W-3275M: Detected core count of 289. 2 x Xeon Gold 6226R: Detected core count of 3210. Xeon Platinum 8280: Detected core count of 2811. 2 x Xeon Platinum 8275CL: Detected core count of 4812. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4813. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgGFLOPS Per Thread, More Is BetterArrayFire 3.7Performance Per Thread - Test: BLAS CPUEPYC 2 7R322 x Xeon Silver 4216Xeon Silver 42162 x Xeon Platinum 8280Xeon Gold 52182 x Xeon Gold 5220RXeon Gold 5220RXeon W-3275M2 x Xeon Gold 6226RXeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offXeon Gold 6226R153045607510.3223.4023.9230.6331.3232.7733.3855.8661.1361.2965.2065.8966.781. EPYC 2 7R32: Detected thread count of 962. 2 x Xeon Silver 4216: Detected thread count of 643. Xeon Silver 4216: Detected thread count of 324. 2 x Xeon Platinum 8280: Detected thread count of 1125. Xeon Gold 5218: Detected thread count of 326. 2 x Xeon Gold 5220R: Detected thread count of 727. Xeon Gold 5220R: Detected thread count of 368. Xeon W-3275M: Detected thread count of 569. 2 x Xeon Gold 6226R: Detected thread count of 6410. Xeon Platinum 8280: Detected thread count of 5611. 2 x Xeon Platinum 8275CL: Detected thread count of 9612. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9613. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgGFLOPS Per GHz, More Is BetterArrayFire 3.7Performance Per Clock - Test: BLAS CPUXeon Silver 4216Xeon Gold 5218Xeon Gold 5220REPYC 2 7R322 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off30060090012001500239.22256.95308.09353.86468.05547.96604.91680.08857.57858.041003.151604.871621.811. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.805. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 6. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 7. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 8. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 9. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 10. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 11. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 13. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90
OpenBenchmarking.orgGFLOPS Per Channel, More Is BetterArrayFire 3.7Performance Per Memory Channel - Test: BLAS CPUEPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 6226R140280420560700123.85127.58167.02200.26249.63356.17393.19571.71572.03652.051. EPYC 2 7R32: Detected channel count of 82. Xeon Silver 4216: Detected channel count of 63. Xeon Gold 5218: Detected channel count of 64. Xeon Gold 5220R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Gold 6226R: Detected channel count of 67. 2 x Xeon Gold 5220R: Detected channel count of 68. 2 x Xeon Platinum 8280: Detected channel count of 69. Xeon Platinum 8280: Detected channel count of 610. 2 x Xeon Gold 6226R: Detected channel count of 6
OpenBenchmarking.orgGFLOPS, More Is BetterArrayFire 3.7Test: BLAS CPUXeon Silver 4216EPYC 2 7R32Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off11002200330044005500Min: 763.23 / Avg: 765.49 / Max: 768.38Min: 983.21 / Avg: 990.8 / Max: 1003.73Min: 1001.43 / Avg: 1002.09 / Max: 1003.32Min: 1200.98 / Avg: 1201.54 / Max: 1202.24Min: 1496.17 / Avg: 1497.75 / Max: 1498.69Min: 2129.01 / Avg: 2137.03 / Max: 2142.94Min: 2351.37 / Avg: 2359.13 / Max: 2363.88Min: 3086.09 / Avg: 3128.39 / Max: 3187.37Min: 3391.32 / Avg: 3430.28 / Max: 3458.08Min: 3385.97 / Avg: 3432.16 / Max: 3472.32Min: 3849.99 / Avg: 3912.3 / Max: 3954.29Min: 6136.22 / Avg: 6259.01 / Max: 6325.81Min: 6318.16 / Avg: 6325.07 / Max: 6334.541. (CXX) g++ options: -rdynamic

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: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828090180270360450SE +/- 0.95, N = 3SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 1.13, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.88, N = 3SE +/- 0.79, N = 3SE +/- 0.08, N = 3395.44298.30244.75197.91155.37122.76108.5796.6994.1981.1479.1157.5149.84MIN: 390.56MIN: 297.84MIN: 242.31MIN: 195.97MIN: 154.79MIN: 121.68MIN: 102.46MIN: 93.63MIN: 93.73MIN: 79.72MIN: 74.841. (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: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828070140210280350Min: 393.64 / Avg: 395.44 / Max: 396.87Min: 298.15 / Avg: 298.3 / Max: 298.51Min: 244.44 / Avg: 244.75 / Max: 244.98Min: 197.83 / Avg: 197.91 / Max: 197.98Min: 155.26 / Avg: 155.37 / Max: 155.43Min: 122.69 / Avg: 122.76 / Max: 122.81Min: 108.37 / Avg: 108.57 / Max: 108.74Min: 94.45 / Avg: 96.69 / Max: 98.06Min: 94.07 / Avg: 94.19 / Max: 94.27Min: 81.1 / Avg: 81.14 / Max: 81.2Min: 77.71 / Avg: 79.11 / Max: 80.74Min: 56.48 / Avg: 57.51 / Max: 59.07Min: 49.69 / Avg: 49.84 / Max: 49.951. (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: u8s8f32EPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82803K6K9K12K15KSE +/- 19.01, N = 3SE +/- 4.87, N = 3SE +/- 1.57, N = 3SE +/- 4.48, N = 3SE +/- 7.74, N = 3SE +/- 23.72, N = 3SE +/- 15.10, N = 3SE +/- 1.24, N = 3SE +/- 2.07, N = 3SE +/- 10.67, N = 3SE +/- 3.97, N = 3SE +/- 0.40, N = 3SE +/- 3.12, N = 311728.806241.165840.264826.714478.933215.172967.922846.122479.282377.912023.471708.051517.44MIN: 11565.1MIN: 6214.68MIN: 5831.48MIN: 4806.63MIN: 4461.44MIN: 3181.24MIN: 2893.12MIN: 2813.781. (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: u8s8f32EPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82802K4K6K8K10KMin: 11690.8 / Avg: 11728.8 / Max: 11749.1Min: 6234.77 / Avg: 6241.16 / Max: 6250.73Min: 5838.56 / Avg: 5840.26 / Max: 5843.39Min: 4817.98 / Avg: 4826.71 / Max: 4832.8Min: 4466.85 / Avg: 4478.93 / Max: 4493.36Min: 3191.41 / Avg: 3215.17 / Max: 3262.6Min: 2951.08 / Avg: 2967.92 / Max: 2998.05Min: 2843.64 / Avg: 2846.12 / Max: 2847.53Min: 2475.39 / Avg: 2479.28 / Max: 2482.46Min: 2366.63 / Avg: 2377.91 / Max: 2399.24Min: 2015.56 / Avg: 2023.47 / Max: 2028.08Min: 1707.25 / Avg: 1708.05 / Max: 1708.54Min: 1513.64 / Avg: 1517.44 / Max: 1523.621. (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_googlenet_v3 - Data Type: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82804080120160200SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.24, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 3171.43131.09108.1687.5469.1454.8249.8742.5442.0836.3733.6325.0523.18MIN: 170.75MIN: 130.56MIN: 106.43MIN: 86.64MIN: 68.48MIN: 54.08MIN: 44.78MIN: 40.72MIN: 41.31MIN: 34.65MIN: 32.491. (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_googlenet_v3 - Data Type: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280306090120150Min: 171.33 / Avg: 171.43 / Max: 171.54Min: 131.05 / Avg: 131.09 / Max: 131.15Min: 107.84 / Avg: 108.16 / Max: 108.62Min: 87.46 / Avg: 87.54 / Max: 87.61Min: 69.03 / Avg: 69.14 / Max: 69.27Min: 54.7 / Avg: 54.82 / Max: 54.9Min: 49.75 / Avg: 49.87 / Max: 49.99Min: 42.21 / Avg: 42.54 / Max: 42.87Min: 42.04 / Avg: 42.08 / Max: 42.1Min: 36.35 / Avg: 36.37 / Max: 36.39Min: 33.37 / Avg: 33.63 / Max: 33.95Min: 25.04 / Avg: 25.05 / Max: 25.08Min: 23.03 / Avg: 23.18 / Max: 23.351. (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: u8s8f32EPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8275CL NUMA offXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 828030060090012001500SE +/- 2.69, N = 3SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.73, N = 3SE +/- 0.14, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.25, N = 3SE +/- 0.03, N = 31301.36105.8381.0766.7459.9055.3442.2235.9831.9930.6426.8426.5419.81MIN: 1286.291. (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: u8s8f32EPYC 2 7R32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8275CL NUMA offXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 82802004006008001000Min: 1295.98 / Avg: 1301.36 / Max: 1304.14Min: 105.74 / Avg: 105.83 / Max: 105.95Min: 81 / Avg: 81.07 / Max: 81.13Min: 66.57 / Avg: 66.74 / Max: 66.89Min: 58.62 / Avg: 59.9 / Max: 61.17Min: 55.17 / Avg: 55.34 / Max: 55.61Min: 42.21 / Avg: 42.22 / Max: 42.24Min: 35.92 / Avg: 35.98 / Max: 36.02Min: 31.94 / Avg: 31.99 / Max: 32.06Min: 30.58 / Avg: 30.64 / Max: 30.68Min: 26.8 / Avg: 26.84 / Max: 26.88Min: 26.05 / Avg: 26.54 / Max: 26.86Min: 19.76 / Avg: 19.81 / Max: 19.861. (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_3d - Data Type: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280246810SE +/- 0.01443, N = 3SE +/- 0.01731, N = 3SE +/- 0.01927, N = 3SE +/- 0.00590, N = 3SE +/- 0.00634, N = 3SE +/- 0.00092, N = 3SE +/- 0.01801, N = 3SE +/- 0.00603, N = 3SE +/- 0.00032, N = 3SE +/- 0.00200, N = 3SE +/- 0.00161, N = 3SE +/- 0.00132, N = 3SE +/- 0.00373, N = 37.636465.804204.889663.855123.130062.697002.275262.204711.962791.707491.173851.172781.13608MIN: 7.57MIN: 5.77MIN: 4.84MIN: 3.82MIN: 3.1MIN: 2.66MIN: 2.12MIN: 2.14MIN: 1.951. (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_3d - Data Type: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82803691215Min: 7.61 / Avg: 7.64 / Max: 7.66Min: 5.78 / Avg: 5.8 / Max: 5.84Min: 4.87 / Avg: 4.89 / Max: 4.93Min: 3.84 / Avg: 3.86 / Max: 3.86Min: 3.12 / Avg: 3.13 / Max: 3.14Min: 2.7 / Avg: 2.7 / Max: 2.7Min: 2.25 / Avg: 2.28 / Max: 2.31Min: 2.2 / Avg: 2.2 / Max: 2.22Min: 1.96 / Avg: 1.96 / Max: 1.96Min: 1.7 / Avg: 1.71 / Max: 1.71Min: 1.17 / Avg: 1.17 / Max: 1.18Min: 1.17 / Avg: 1.17 / Max: 1.18Min: 1.13 / Avg: 1.14 / Max: 1.141. (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: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Gold 5220RXeon W-3275MXeon Platinum 8280EPYC 2 7R322 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82807001400210028003500SE +/- 1.30, N = 3SE +/- 0.47, N = 3SE +/- 1.16, N = 3SE +/- 1.69, N = 3SE +/- 0.59, N = 3SE +/- 7.07, N = 3SE +/- 0.29, N = 3SE +/- 0.68, N = 3SE +/- 0.39, N = 3SE +/- 2.05, N = 3SE +/- 0.79, N = 3SE +/- 5.95, N = 3SE +/- 0.78, N = 33129.932363.001948.771578.241227.071206.22999.54878.39765.75710.14694.32650.98496.82MIN: 3085.85MIN: 2357.87MIN: 1927.2MIN: 1571.3MIN: 1218.34MIN: 1181.19MIN: 995.71MIN: 819.94MIN: 755.35MIN: 700.52MIN: 686.24MIN: 639.11. (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: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Gold 5220RXeon W-3275MXeon Platinum 8280EPYC 2 7R322 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82805001000150020002500Min: 3127.45 / Avg: 3129.93 / Max: 3131.87Min: 2362.17 / Avg: 2363 / Max: 2363.81Min: 1946.47 / Avg: 1948.77 / Max: 1950.16Min: 1575.89 / Avg: 1578.24 / Max: 1581.52Min: 1226.38 / Avg: 1227.07 / Max: 1228.24Min: 1192.1 / Avg: 1206.22 / Max: 1214.06Min: 999.25 / Avg: 999.54 / Max: 1000.12Min: 877.4 / Avg: 878.39 / Max: 879.69Min: 765.14 / Avg: 765.75 / Max: 766.49Min: 706.04 / Avg: 710.14 / Max: 712.25Min: 692.88 / Avg: 694.32 / Max: 695.62Min: 644.85 / Avg: 650.98 / Max: 662.89Min: 495.3 / Avg: 496.82 / Max: 497.871. (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 Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 8280400800120016002000SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 1.48, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 1.04, N = 3SE +/- 0.07, N = 3SE +/- 4.23, N = 3SE +/- 0.33, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 31875.231874.471433.471042.34938.91712.20702.07637.37527.02348.33348.24320.06MIN: 1873.9MIN: 1873.87MIN: 1421.77MIN: 1041.43MIN: 937.57MIN: 710.93MIN: 687.86MIN: 636.97MIN: 521.14MIN: 347.28MIN: 346.92MIN: 319.231. (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 Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 828030060090012001500Min: 1875.18 / Avg: 1875.23 / Max: 1875.27Min: 1874.2 / Avg: 1874.47 / Max: 1874.72Min: 1430.58 / Avg: 1433.47 / Max: 1435.46Min: 1042.2 / Avg: 1042.34 / Max: 1042.5Min: 938.8 / Avg: 938.91 / Max: 939.07Min: 712.08 / Avg: 712.2 / Max: 712.3Min: 700.45 / Avg: 702.07 / Max: 704.02Min: 637.27 / Avg: 637.37 / Max: 637.49Min: 522.71 / Avg: 527.02 / Max: 535.48Min: 347.69 / Avg: 348.33 / Max: 348.81Min: 348.13 / Avg: 348.24 / Max: 348.38Min: 319.97 / Avg: 320.06 / Max: 320.181. (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: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82802K4K6K8K10KSE +/- 0.35, N = 3SE +/- 0.34, N = 3SE +/- 4.68, N = 3SE +/- 0.12, N = 3SE +/- 6.81, N = 3SE +/- 3.93, N = 3SE +/- 0.11, N = 3SE +/- 0.26, N = 3SE +/- 2.73, N = 3SE +/- 0.16, N = 3SE +/- 0.26, N = 3SE +/- 1.13, N = 310258.0010255.207839.955709.175156.693917.523880.443497.472871.111919.501918.281766.35MIN: 10250.4MIN: 10250.1MIN: 7795.37MIN: 5699.77MIN: 5142.07MIN: 3906.67MIN: 3767MIN: 3490.41MIN: 2860.81MIN: 1911.94MIN: 1911.421. (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: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82802K4K6K8K10KMin: 10257.3 / Avg: 10257.97 / Max: 10258.5Min: 10254.5 / Avg: 10255.17 / Max: 10255.6Min: 7834.19 / Avg: 7839.95 / Max: 7849.22Min: 5708.98 / Avg: 5709.17 / Max: 5709.4Min: 5149.22 / Avg: 5156.69 / Max: 5170.3Min: 3912.84 / Avg: 3917.52 / Max: 3925.33Min: 3880.24 / Avg: 3880.44 / Max: 3880.6Min: 3497.1 / Avg: 3497.47 / Max: 3497.97Min: 2867.97 / Avg: 2871.11 / Max: 2876.54Min: 1919.22 / Avg: 1919.5 / Max: 1919.79Min: 1918.01 / Avg: 1918.28 / Max: 1918.81Min: 1765.16 / Avg: 1766.35 / Max: 1768.61. (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_googlenet_v3 - Data Type: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Gold 5220RXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280100200300400500SE +/- 0.08, N = 3SE +/- 0.02, N = 3SE +/- 0.42, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.20, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3484.90484.59369.29269.99246.45188.61187.17166.98137.7393.3693.2886.54MIN: 484.15MIN: 484.13MIN: 365.68MIN: 268.99MIN: 245.49MIN: 178.56MIN: 186.33MIN: 166.21MIN: 136.67MIN: 92.35MIN: 92.44MIN: 85.561. (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_googlenet_v3 - Data Type: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Gold 5220RXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828090180270360450Min: 484.78 / Avg: 484.9 / Max: 485.04Min: 484.56 / Avg: 484.59 / Max: 484.62Min: 368.87 / Avg: 369.29 / Max: 370.12Min: 269.85 / Avg: 269.99 / Max: 270.19Min: 246.4 / Avg: 246.45 / Max: 246.54Min: 188.22 / Avg: 188.61 / Max: 188.81Min: 187.14 / Avg: 187.17 / Max: 187.21Min: 166.94 / Avg: 166.98 / Max: 167.04Min: 137.7 / Avg: 137.73 / Max: 137.77Min: 93.34 / Avg: 93.36 / Max: 93.38Min: 93.23 / Avg: 93.28 / Max: 93.32Min: 86.53 / Avg: 86.54 / Max: 86.551. (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: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 828048121620SE +/- 0.00, N = 15SE +/- 0.01, N = 3SE +/- 0.00, N = 12SE +/- 0.01, N = 4SE +/- 0.00, N = 12SE +/- 0.01, N = 12SE +/- 0.03, N = 3SE +/- 0.00, N = 12SE +/- 0.04, N = 3SE +/- 0.07, N = 4SE +/- 0.00, N = 12SE +/- 0.07, N = 4SE +/- 0.00, N = 123.414.375.385.526.628.699.4910.4210.5616.6016.9517.0218.18MIN: 3.33 / MAX: 3.45MIN: 4.29 / MAX: 4.44MIN: 5.29 / MAX: 5.46MIN: 5.43 / MAX: 5.62MIN: 5.52 / MAX: 6.76MIN: 8.33 / MAX: 9.01MIN: 9.26 / MAX: 9.62MIN: 8.47 / MAX: 10.53MIN: 8.77 / MAX: 10.75MIN: 15.87 / MAX: 16.95MIN: 16.13 / MAX: 17.54MIN: 11.76 / MAX: 17.54MIN: 12.2 / MAX: 18.52
OpenBenchmarking.orgFPS Per Core, More Is BetterOSPray 1.8.5Performance Per Core - Demo: NASA Streamlines - Renderer: Path Tracer2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon Gold 5220RXeon W-3275M2 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Platinum 8280Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL0.07980.15960.23940.31920.3990.20690.21310.27310.28940.29890.31040.32460.33000.33890.34500.34580.35310.35461. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. Xeon Gold 5218: Detected core count of 164. 2 x Xeon Gold 5220R: Detected core count of 365. Xeon Gold 5220R: Detected core count of 186. Xeon W-3275M: Detected core count of 287. 2 x Xeon Platinum 8280: Detected core count of 568. 2 x Xeon Gold 6226R: Detected core count of 329. Xeon Platinum 8280: Detected core count of 2810. Xeon Gold 6226R: Detected core count of 1611. EPYC 2 7R32: Detected core count of 4812. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4813. 2 x Xeon Platinum 8275CL: Detected core count of 48
OpenBenchmarking.orgFPS Per Thread, More Is BetterOSPray 1.8.5Performance Per Thread - Demo: NASA Streamlines - Renderer: Path Tracer2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon Gold 5220RXeon W-3275M2 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Platinum 8280Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL0.03990.07980.11970.15960.19950.10340.10660.13660.14470.14940.15520.16230.16500.16950.17250.17290.17660.17731. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. Xeon Gold 5218: Detected thread count of 324. 2 x Xeon Gold 5220R: Detected thread count of 725. Xeon Gold 5220R: Detected thread count of 366. Xeon W-3275M: Detected thread count of 567. 2 x Xeon Platinum 8280: Detected thread count of 1128. 2 x Xeon Gold 6226R: Detected thread count of 649. Xeon Platinum 8280: Detected thread count of 5610. Xeon Gold 6226R: Detected thread count of 3211. EPYC 2 7R32: Detected thread count of 9612. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9613. 2 x Xeon Platinum 8275CL: Detected thread count of 96
OpenBenchmarking.orgFPS Per GHz, More Is BetterOSPray 1.8.5Performance Per Clock - Demo: NASA Streamlines - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 8280EPYC 2 7R321.33432.66864.00295.33726.67151.071.121.381.421.892.072.372.672.714.354.364.555.931. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 7. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 8. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgFPS Per Channel, More Is BetterOSPray 1.8.5Performance Per Memory Channel - Demo: NASA Streamlines - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 82800.68181.36362.04542.72723.4090.56830.72830.89670.92001.10001.58001.74001.76002.08003.03001. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 5220R: Detected channel count of 64. Xeon Gold 6226R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. 2 x Xeon Gold 5220R: Detected channel count of 68. 2 x Xeon Gold 6226R: Detected channel count of 69. EPYC 2 7R32: Detected channel count of 810. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 8280510152025Min: 3.41 / Avg: 3.41 / Max: 3.41Min: 4.37 / Avg: 4.37 / Max: 4.39Min: 5.38 / Avg: 5.38 / Max: 5.38Min: 5.49 / Avg: 5.52 / Max: 5.52Min: 6.62 / Avg: 6.62 / Max: 6.62Min: 8.62 / Avg: 8.69 / Max: 8.7Min: 9.43 / Avg: 9.49 / Max: 9.52Min: 10.42 / Avg: 10.42 / Max: 10.42Min: 10.53 / Avg: 10.56 / Max: 10.64Min: 16.39 / Avg: 16.6 / Max: 16.67Min: 16.95 / Avg: 16.95 / Max: 16.95Min: 16.95 / Avg: 17.02 / Max: 17.24Min: 18.18 / Avg: 18.18 / Max: 18.18

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82803691215SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 6SE +/- 0.00, N = 12SE +/- 0.00, N = 8SE +/- 0.00, N = 3SE +/- 0.00, N = 12SE +/- 0.00, N = 6SE +/- 0.04, N = 3SE +/- 0.00, N = 13SE +/- 0.00, N = 13SE +/- 0.00, N = 122.343.013.683.704.575.726.257.097.1911.0311.3611.3612.05MIN: 2.3 / MAX: 2.36MIN: 2.98 / MAX: 3.05MIN: 3.64 / MAX: 3.7MIN: 3.66 / MAX: 3.73MIN: 4.08 / MAX: 4.61MIN: 5.56 / MAX: 5.78MIN: 6.13 / MAX: 6.29MIN: 6.37 / MAX: 7.14MIN: 6.45 / MAX: 7.25MIN: 10.75 / MAX: 11.24MIN: 9.8 / MAX: 11.49MIN: 10.99 / MAX: 11.49MIN: 9.71 / MAX: 12.2
OpenBenchmarking.orgFPS Per Core, More Is BetterOSPray 1.8.5Performance Per Core - Demo: XFrog Forest - Renderer: SciVis2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon W-3275MXeon Gold 5220R2 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Platinum 8280EPYC 2 7R32Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off0.05330.10660.15990.21320.26650.14280.14630.18810.19970.20430.20560.21520.22160.22320.22980.23000.23670.23671. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. Xeon Gold 5218: Detected core count of 164. 2 x Xeon Gold 5220R: Detected core count of 365. Xeon W-3275M: Detected core count of 286. Xeon Gold 5220R: Detected core count of 187. 2 x Xeon Platinum 8280: Detected core count of 568. 2 x Xeon Gold 6226R: Detected core count of 329. Xeon Platinum 8280: Detected core count of 2810. EPYC 2 7R32: Detected core count of 4811. Xeon Gold 6226R: Detected core count of 1612. 2 x Xeon Platinum 8275CL: Detected core count of 4813. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 48
OpenBenchmarking.orgFPS Per Thread, More Is BetterOSPray 1.8.5Performance Per Thread - Demo: XFrog Forest - Renderer: SciVis2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon W-3275MXeon Gold 5220R2 x Xeon Platinum 82802 x Xeon Gold 6226RXeon Platinum 8280EPYC 2 7R32Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off0.02660.05320.07980.10640.1330.07140.07310.09410.09990.10210.10280.10760.11080.11160.11490.11500.11830.11831. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. Xeon Gold 5218: Detected thread count of 324. 2 x Xeon Gold 5220R: Detected thread count of 725. Xeon W-3275M: Detected thread count of 566. Xeon Gold 5220R: Detected thread count of 367. 2 x Xeon Platinum 8280: Detected thread count of 1128. 2 x Xeon Gold 6226R: Detected thread count of 649. Xeon Platinum 8280: Detected thread count of 5610. EPYC 2 7R32: Detected thread count of 9611. Xeon Gold 6226R: Detected thread count of 3212. 2 x Xeon Platinum 8275CL: Detected thread count of 9613. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 96
OpenBenchmarking.orgFPS Per GHz, More Is BetterOSPray 1.8.5Performance Per Clock - Demo: XFrog Forest - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R320.88651.7732.65953.5464.43250.73130.77180.94360.94871.24001.43001.56001.82001.84002.91002.91003.01003.94001. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 7. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 8. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgFPS Per Channel, More Is BetterOSPray 1.8.5Performance Per Memory Channel - Demo: XFrog Forest - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 82800.45230.90461.35691.80922.26150.39000.50170.61330.61670.76171.04001.18001.20001.38002.01001. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 6226R: Detected channel count of 64. Xeon Gold 5220R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. 2 x Xeon Gold 6226R: Detected channel count of 68. 2 x Xeon Gold 5220R: Detected channel count of 69. EPYC 2 7R32: Detected channel count of 810. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828048121620Min: 2.34 / Avg: 2.34 / Max: 2.34Min: 3.01 / Avg: 3.01 / Max: 3.01Min: 3.68 / Avg: 3.68 / Max: 3.68Min: 3.69 / Avg: 3.7 / Max: 3.7Min: 4.57 / Avg: 4.57 / Max: 4.57Min: 5.71 / Avg: 5.72 / Max: 5.75Min: 6.25 / Avg: 6.25 / Max: 6.25Min: 7.09 / Avg: 7.09 / Max: 7.09Min: 7.19 / Avg: 7.19 / Max: 7.19Min: 10.99 / Avg: 11.03 / Max: 11.11Min: 11.36 / Avg: 11.36 / Max: 11.36Min: 11.36 / Avg: 11.36 / Max: 11.36Min: 12.05 / Avg: 12.05 / Max: 12.05

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 8280246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 6SE +/- 0.00, N = 5SE +/- 0.00, N = 7SE +/- 0.00, N = 4SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 121.311.682.022.062.553.173.463.883.985.856.246.256.67MIN: 1.3 / MAX: 1.32MIN: 1.67MIN: 2 / MAX: 2.04MIN: 2.04 / MAX: 2.07MIN: 2.28 / MAX: 2.57MIN: 3.09 / MAX: 3.19MIN: 3.41 / MAX: 3.47MIN: 3.57 / MAX: 3.92MIN: 3.64 / MAX: 4.02MIN: 5.65 / MAX: 5.99MIN: 6.1 / MAX: 6.29MIN: 5.68 / MAX: 6.29MIN: 5.85 / MAX: 6.76
OpenBenchmarking.orgFPS Per Core, More Is BetterOSPray 1.8.5Performance Per Core - Demo: XFrog Forest - Renderer: Path Tracer2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon W-3275MXeon Gold 5220R2 x Xeon Platinum 82802 x Xeon Gold 6226REPYC 2 7R32Xeon Platinum 8280Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL0.02930.05860.08790.11720.14650.07970.08190.10500.11060.11320.11440.11910.12130.12190.12360.12630.13000.13021. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. Xeon Gold 5218: Detected core count of 164. 2 x Xeon Gold 5220R: Detected core count of 365. Xeon W-3275M: Detected core count of 286. Xeon Gold 5220R: Detected core count of 187. 2 x Xeon Platinum 8280: Detected core count of 568. 2 x Xeon Gold 6226R: Detected core count of 329. EPYC 2 7R32: Detected core count of 4810. Xeon Platinum 8280: Detected core count of 2811. Xeon Gold 6226R: Detected core count of 1612. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4813. 2 x Xeon Platinum 8275CL: Detected core count of 48
OpenBenchmarking.orgFPS Per Thread, More Is BetterOSPray 1.8.5Performance Per Thread - Demo: XFrog Forest - Renderer: Path Tracer2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon W-3275MXeon Gold 5220R2 x Xeon Platinum 82802 x Xeon Gold 6226REPYC 2 7R32Xeon Platinum 8280Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL0.01460.02920.04380.05840.0730.03980.04090.05250.05530.05660.05720.05960.06060.06090.06180.06310.06500.06511. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. Xeon Gold 5218: Detected thread count of 324. 2 x Xeon Gold 5220R: Detected thread count of 725. Xeon W-3275M: Detected thread count of 566. Xeon Gold 5220R: Detected thread count of 367. 2 x Xeon Platinum 8280: Detected thread count of 1128. 2 x Xeon Gold 6226R: Detected thread count of 649. EPYC 2 7R32: Detected thread count of 9610. Xeon Platinum 8280: Detected thread count of 5611. Xeon Gold 6226R: Detected thread count of 3212. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9613. 2 x Xeon Platinum 8275CL: Detected thread count of 96
OpenBenchmarking.orgFPS Per GHz, More Is BetterOSPray 1.8.5Performance Per Clock - Demo: XFrog Forest - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R320.47030.94061.41091.88122.35150.40940.43080.51790.52820.68910.79690.86500.99491.02001.60001.60001.67002.09001. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 7. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 8. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgFPS Per Channel, More Is BetterOSPray 1.8.5Performance Per Memory Channel - Demo: XFrog Forest - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 82800.24980.49960.74940.99921.2490.21830.28000.33670.34330.42500.57670.64670.66330.73131.11001. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 6226R: Detected channel count of 64. Xeon Gold 5220R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. 2 x Xeon Gold 6226R: Detected channel count of 68. 2 x Xeon Gold 5220R: Detected channel count of 69. EPYC 2 7R32: Detected channel count of 810. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 82803691215Min: 1.31 / Avg: 1.31 / Max: 1.31Min: 1.68 / Avg: 1.68 / Max: 1.68Min: 2.02 / Avg: 2.02 / Max: 2.02Min: 2.05 / Avg: 2.06 / Max: 2.06Min: 2.55 / Avg: 2.55 / Max: 2.56Min: 3.16 / Avg: 3.17 / Max: 3.17Min: 3.45 / Avg: 3.46 / Max: 3.46Min: 3.88 / Avg: 3.88 / Max: 3.88Min: 3.98 / Avg: 3.98 / Max: 3.98Min: 5.81 / Avg: 5.85 / Max: 5.88Min: 6.21 / Avg: 6.24 / Max: 6.25Min: 6.25 / Avg: 6.25 / Max: 6.25Min: 6.67 / Avg: 6.67 / Max: 6.67

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 Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Gold 5220RXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280918273645SE +/- 0.00904, N = 3SE +/- 0.00339, N = 3SE +/- 0.01559, N = 3SE +/- 0.00584, N = 3SE +/- 0.00466, N = 3SE +/- 0.24390, N = 4SE +/- 0.00301, N = 3SE +/- 0.00390, N = 3SE +/- 0.00770, N = 3SE +/- 0.00882, N = 3SE +/- 0.00469, N = 3SE +/- 0.00328, N = 340.0192039.9985029.9173022.4243020.5163016.5906015.2177014.0780011.607608.712348.647077.92912MIN: 39.9MIN: 39.89MIN: 29.42MIN: 22.32MIN: 20.35MIN: 14.98MIN: 15.05MIN: 13.95MIN: 11.46MIN: 8.5MIN: 8.431. (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_3d - Data Type: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Gold 5220RXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280816243240Min: 40.01 / Avg: 40.02 / Max: 40.04Min: 39.99 / Avg: 40 / Max: 40.01Min: 29.89 / Avg: 29.92 / Max: 29.94Min: 22.41 / Avg: 22.42 / Max: 22.43Min: 20.51 / Avg: 20.52 / Max: 20.52Min: 16.33 / Avg: 16.59 / Max: 17.32Min: 15.21 / Avg: 15.22 / Max: 15.22Min: 14.07 / Avg: 14.08 / Max: 14.08Min: 11.59 / Avg: 11.61 / Max: 11.62Min: 8.7 / Avg: 8.71 / Max: 8.73Min: 8.64 / Avg: 8.65 / Max: 8.66Min: 7.92 / Avg: 7.93 / Max: 7.931. (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: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828020406080100SE +/- 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 = 15SE +/- 0.00, N = 15SE +/- 0.00, N = 12SE +/- 0.71, N = 9SE +/- 0.00, N = 12SE +/- 0.00, N = 1318.1823.2628.5729.4134.4845.4550.0055.5655.5676.9282.6283.3390.91MIN: 17.86 / MAX: 18.52MIN: 22.73 / MAX: 23.81MIN: 26.32 / MAX: 29.41MIN: 26.32MIN: 18.87 / MAX: 35.71MIN: 43.48 / MAX: 47.62MIN: 43.48MIN: 24.39MIN: 19.23MIN: 71.43 / MAX: 83.33MIN: 26.32 / MAX: 90.91MIN: 71.43 / MAX: 90.91MIN: 29.41
OpenBenchmarking.orgFPS Per Core, More Is BetterOSPray 1.8.5Performance Per Core - Demo: NASA Streamlines - Renderer: SciVis2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8280Xeon W-3275M2 x Xeon Platinum 8275CL2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA offXeon Platinum 8280Xeon Gold 6226R0.4140.8281.2421.6562.071.081.141.451.541.591.601.621.621.721.741.741.791.841. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. Xeon Gold 5218: Detected core count of 164. 2 x Xeon Gold 5220R: Detected core count of 365. Xeon Gold 5220R: Detected core count of 186. EPYC 2 7R32: Detected core count of 487. 2 x Xeon Platinum 8280: Detected core count of 568. Xeon W-3275M: Detected core count of 289. 2 x Xeon Platinum 8275CL: Detected core count of 4810. 2 x Xeon Gold 6226R: Detected core count of 3211. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4812. Xeon Platinum 8280: Detected core count of 2813. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgFPS Per Thread, More Is BetterOSPray 1.8.5Performance Per Thread - Demo: NASA Streamlines - Renderer: SciVis2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon Gold 5220REPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Gold 6226RXeon Platinum 8280Xeon Gold 6226R0.20680.41360.62040.82721.0340.53880.56810.72690.77170.79360.80130.81160.81170.86060.86800.86810.89290.91911. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. Xeon Gold 5218: Detected thread count of 324. 2 x Xeon Gold 5220R: Detected thread count of 725. Xeon Gold 5220R: Detected thread count of 366. EPYC 2 7R32: Detected thread count of 967. Xeon W-3275M: Detected thread count of 568. 2 x Xeon Platinum 8280: Detected thread count of 1129. 2 x Xeon Platinum 8275CL: Detected thread count of 9610. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9611. 2 x Xeon Gold 6226R: Detected thread count of 6412. Xeon Platinum 8280: Detected thread count of 5613. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgFPS Per GHz, More Is BetterOSPray 1.8.5Performance Per Clock - Demo: NASA Streamlines - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R326121824305.685.967.337.549.8810.7812.5014.2514.2521.1821.3722.7327.471. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 7. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 8. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgFPS Per Channel, More Is BetterOSPray 1.8.5Performance Per Memory Channel - Demo: NASA Streamlines - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8280481216203.033.884.764.905.758.339.269.269.6215.151. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 5220R: Detected channel count of 64. Xeon Gold 6226R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. 2 x Xeon Gold 5220R: Detected channel count of 68. 2 x Xeon Gold 6226R: Detected channel count of 69. EPYC 2 7R32: Detected channel count of 810. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828020406080100Min: 18.18 / Avg: 18.18 / Max: 18.18Min: 23.26 / Avg: 23.26 / Max: 23.26Min: 28.57 / Avg: 28.57 / Max: 28.57Min: 29.41 / Avg: 29.41 / Max: 29.41Min: 34.48 / Avg: 34.48 / Max: 34.48Min: 45.45 / Avg: 45.45 / Max: 45.45Min: 55.56 / Avg: 55.56 / Max: 55.56Min: 55.56 / Avg: 55.56 / Max: 55.56Min: 76.92 / Avg: 76.92 / Max: 76.92Min: 76.92 / Avg: 82.62 / Max: 83.33Min: 83.33 / Avg: 83.33 / Max: 83.33Min: 90.91 / Avg: 90.91 / Max: 90.91

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: u8s8f32EPYC 2 7R32Xeon Silver 42162 x Xeon Platinum 8275CLXeon Gold 52182 x Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offXeon W-3275MXeon Platinum 82802 x Xeon Platinum 8280306090120150SE +/- 0.23997, N = 3SE +/- 0.03131, N = 3SE +/- 0.06319, N = 3SE +/- 0.00870, N = 3SE +/- 0.04012, N = 3SE +/- 0.02448, N = 3SE +/- 0.01160, N = 3SE +/- 0.03959, N = 3SE +/- 0.03777, N = 3SE +/- 0.01731, N = 3SE +/- 0.03226, N = 3SE +/- 0.00398, N = 3SE +/- 0.04146, N = 4149.339005.592755.367084.838754.826034.610544.538214.117794.086053.847463.592183.231213.11002MIN: 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: u8s8f32EPYC 2 7R32Xeon Silver 42162 x Xeon Platinum 8275CLXeon Gold 52182 x Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offXeon W-3275MXeon Platinum 82802 x Xeon Platinum 8280306090120150Min: 148.88 / Avg: 149.34 / Max: 149.7Min: 5.53 / Avg: 5.59 / Max: 5.64Min: 5.3 / Avg: 5.37 / Max: 5.49Min: 4.82 / Avg: 4.84 / Max: 4.85Min: 4.76 / Avg: 4.83 / Max: 4.89Min: 4.56 / Avg: 4.61 / Max: 4.64Min: 4.52 / Avg: 4.54 / Max: 4.56Min: 4.06 / Avg: 4.12 / Max: 4.19Min: 4.01 / Avg: 4.09 / Max: 4.14Min: 3.81 / Avg: 3.85 / Max: 3.87Min: 3.53 / Avg: 3.59 / Max: 3.63Min: 3.23 / Avg: 3.23 / Max: 3.24Min: 3 / Avg: 3.11 / Max: 3.21. (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_3d - Data Type: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280510152025SE +/- 0.02388, N = 3SE +/- 0.05476, N = 3SE +/- 0.00134, N = 3SE +/- 0.00297, N = 3SE +/- 0.00821, N = 3SE +/- 0.01392, N = 3SE +/- 0.03985, N = 3SE +/- 0.00251, N = 3SE +/- 0.00159, N = 3SE +/- 0.01549, N = 3SE +/- 0.00222, N = 3SE +/- 0.00369, N = 321.4755021.4290016.8253012.3271010.937909.425918.462737.696056.296474.379524.360024.35817MIN: 21.24MIN: 21.29MIN: 16.72MIN: 12.27MIN: 10.86MIN: 9.32MIN: 8.21MIN: 7.67MIN: 6.251. (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_3d - Data Type: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280510152025Min: 21.44 / Avg: 21.48 / Max: 21.52Min: 21.37 / Avg: 21.43 / Max: 21.54Min: 16.82 / Avg: 16.83 / Max: 16.83Min: 12.32 / Avg: 12.33 / Max: 12.33Min: 10.92 / Avg: 10.94 / Max: 10.95Min: 9.4 / Avg: 9.43 / Max: 9.45Min: 8.39 / Avg: 8.46 / Max: 8.53Min: 7.69 / Avg: 7.7 / Max: 7.7Min: 6.29 / Avg: 6.3 / Max: 6.3Min: 4.36 / Avg: 4.38 / Max: 4.41Min: 4.36 / Avg: 4.36 / Max: 4.36Min: 4.35 / Avg: 4.36 / Max: 4.361. (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: San Miguel - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82801530456075SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.04, N = 12SE +/- 0.00, N = 12SE +/- 0.07, N = 12SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.60, N = 3SE +/- 0.00, N = 12SE +/- 0.00, N = 12SE +/- 0.54, N = 12SE +/- 0.00, N = 12SE +/- 0.62, N = 1514.7118.8722.6923.2627.8434.4838.4642.2743.4858.8263.5466.6769.52MIN: 14.29 / MAX: 14.93MIN: 18.52 / MAX: 19.23MIN: 21.74 / MAX: 22.73MIN: 21.74 / MAX: 23.81MIN: 17.86 / MAX: 28.57MIN: 30.3 / MAX: 35.71MIN: 37.04MIN: 28.57 / MAX: 43.48MIN: 28.57 / MAX: 45.45MIN: 52.63 / MAX: 62.5MIN: 35.71 / MAX: 66.67MIN: 55.56MIN: 37.04 / MAX: 71.43
OpenBenchmarking.orgFPS Per Core, More Is BetterOSPray 1.8.5Performance Per Core - Demo: San Miguel - Renderer: SciVis2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R322 x Xeon Platinum 8280Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CLXeon Platinum 82802 x Xeon Platinum 8275CL NUMA offXeon Gold 6226R0.31950.6390.95851.2781.59750.87000.91941.18001.21001.23001.23001.24001.29001.32001.32001.37001.39001.42001. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. Xeon Gold 5218: Detected core count of 164. 2 x Xeon Gold 5220R: Detected core count of 365. Xeon W-3275M: Detected core count of 286. EPYC 2 7R32: Detected core count of 487. 2 x Xeon Platinum 8280: Detected core count of 568. Xeon Gold 5220R: Detected core count of 189. 2 x Xeon Gold 6226R: Detected core count of 3210. 2 x Xeon Platinum 8275CL: Detected core count of 4811. Xeon Platinum 8280: Detected core count of 2812. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4813. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgFPS Per Thread, More Is BetterOSPray 1.8.5Performance Per Thread - Demo: San Miguel - Renderer: SciVis2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 52182 x Xeon Gold 5220REPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 8280Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CLXeon Platinum 82802 x Xeon Platinum 8275CL NUMA offXeon Gold 6226R0.15950.3190.47850.6380.79750.43500.45970.58970.60390.61270.61570.62070.64610.66050.66190.68680.69450.70911. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. Xeon Gold 5218: Detected thread count of 324. 2 x Xeon Gold 5220R: Detected thread count of 725. EPYC 2 7R32: Detected thread count of 966. Xeon W-3275M: Detected thread count of 567. 2 x Xeon Platinum 8280: Detected thread count of 1128. Xeon Gold 5220R: Detected thread count of 369. 2 x Xeon Gold 6226R: Detected thread count of 6410. 2 x Xeon Platinum 8275CL: Detected thread count of 9611. Xeon Platinum 8280: Detected thread count of 5612. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9613. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgFPS Per GHz, More Is BetterOSPray 1.8.5Performance Per Clock - Demo: San Miguel - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R325101520254.604.845.825.967.508.709.6210.8411.1516.2917.0917.3821.011. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 7. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 8. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgFPS Per Channel, More Is BetterOSPray 1.8.5Performance Per Memory Channel - Demo: San Miguel - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 828036912152.453.153.783.884.646.417.057.257.3511.591. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 6226R: Detected channel count of 64. Xeon Gold 5220R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. 2 x Xeon Gold 6226R: Detected channel count of 68. 2 x Xeon Gold 5220R: Detected channel count of 69. EPYC 2 7R32: Detected channel count of 810. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82801326395265Min: 14.71 / Avg: 14.71 / Max: 14.71Min: 18.87 / Avg: 18.87 / Max: 18.87Min: 22.22 / Avg: 22.69 / Max: 22.73Min: 23.26 / Avg: 23.26 / Max: 23.26Min: 27.78 / Avg: 27.84 / Max: 28.57Min: 34.48 / Avg: 34.48 / Max: 34.48Min: 38.46 / Avg: 38.46 / Max: 38.46Min: 41.67 / Avg: 42.27 / Max: 43.48Min: 43.48 / Avg: 43.48 / Max: 43.48Min: 58.82 / Avg: 58.82 / Max: 58.82Min: 62.5 / Avg: 63.54 / Max: 66.67Min: 66.67 / Avg: 66.67 / Max: 66.67Min: 66.67 / Avg: 69.52 / Max: 71.43

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: CrownXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82801224364860SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.49, N = 12SE +/- 0.03, N = 3SE +/- 0.05, N = 311.3814.4216.8117.6621.7526.4028.7731.7333.4545.1949.0549.9653.23MIN: 11.32 / MAX: 11.49MIN: 14.3 / MAX: 14.63MIN: 16.67 / MAX: 17MIN: 17.5 / MAX: 17.85MIN: 21.43 / MAX: 22.13MIN: 25.71 / MAX: 26.95MIN: 28.26 / MAX: 29.24MIN: 31.27 / MAX: 32.36MIN: 32.87 / MAX: 34.13MIN: 44.15 / MAX: 46.52MIN: 37.93 / MAX: 51.79MIN: 48.68 / MAX: 51.86MIN: 51.36 / MAX: 55.83
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: CrownXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82801122334455Min: 11.38 / Avg: 11.38 / Max: 11.39Min: 14.4 / Avg: 14.42 / Max: 14.43Min: 16.78 / Avg: 16.81 / Max: 16.84Min: 17.65 / Avg: 17.66 / Max: 17.67Min: 21.73 / Avg: 21.75 / Max: 21.76Min: 26.36 / Avg: 26.4 / Max: 26.48Min: 28.75 / Avg: 28.77 / Max: 28.8Min: 31.68 / Avg: 31.73 / Max: 31.78Min: 33.41 / Avg: 33.45 / Max: 33.48Min: 45.07 / Avg: 45.19 / Max: 45.32Min: 43.85 / Avg: 49.05 / Max: 49.92Min: 49.93 / Avg: 49.96 / Max: 50.01Min: 53.18 / Avg: 53.23 / Max: 53.32

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 Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280246810SE +/- 0.00, N = 3SE +/- 0.00, N = 6SE +/- 0.00, N = 3SE +/- 0.00, N = 4SE +/- 0.00, N = 4SE +/- 0.00, N = 12SE +/- 0.00, N = 3SE +/- 0.00, N = 8SE +/- 0.00, N = 7SE +/- 0.01, N = 3SE +/- 0.00, N = 12SE +/- 0.01, N = 4SE +/- 0.01, N = 51.401.752.022.142.633.253.523.713.954.735.825.876.53MAX: 1.41MIN: 1.74 / MAX: 1.76MAX: 2.03MIN: 2.13 / MAX: 2.15MIN: 2.35 / MAX: 2.64MIN: 3.16 / MAX: 3.28MIN: 3.48 / MAX: 3.53MIN: 2.99 / MAX: 3.73MIN: 3.41 / MAX: 3.98MIN: 4.69 / MAX: 4.78MIN: 4.69 / MAX: 5.88MIN: 5.75 / MAX: 5.92MIN: 5.18 / MAX: 6.58
OpenBenchmarking.orgFPS Per Core, More Is BetterOSPray 1.8.5Performance Per Core - Demo: San Miguel - Renderer: Path Tracer2 x Xeon Silver 4216Xeon Silver 4216EPYC 2 7R32Xeon Gold 52182 x Xeon Gold 5220R2 x Xeon Gold 6226RXeon W-3275M2 x Xeon Platinum 8280Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offXeon Platinum 8280Xeon Gold 6226R0.02840.05680.08520.11360.1420.08220.08750.09850.10940.10970.11590.11610.11660.11890.12130.12230.12570.12631. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. EPYC 2 7R32: Detected core count of 484. Xeon Gold 5218: Detected core count of 165. 2 x Xeon Gold 5220R: Detected core count of 366. 2 x Xeon Gold 6226R: Detected core count of 327. Xeon W-3275M: Detected core count of 288. 2 x Xeon Platinum 8280: Detected core count of 569. Xeon Gold 5220R: Detected core count of 1810. 2 x Xeon Platinum 8275CL: Detected core count of 4811. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4812. Xeon Platinum 8280: Detected core count of 2813. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgFPS Per Thread, More Is BetterOSPray 1.8.5Performance Per Thread - Demo: San Miguel - Renderer: Path Tracer2 x Xeon Silver 4216Xeon Silver 4216EPYC 2 7R32Xeon Gold 52182 x Xeon Gold 5220R2 x Xeon Gold 6226RXeon W-3275M2 x Xeon Platinum 8280Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offXeon Platinum 8280Xeon Gold 6226R0.01420.02840.04260.05680.0710.04110.04380.04930.05470.05490.05800.05800.05830.05940.06060.06110.06290.06311. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. EPYC 2 7R32: Detected thread count of 964. Xeon Gold 5218: Detected thread count of 325. 2 x Xeon Gold 5220R: Detected thread count of 726. 2 x Xeon Gold 6226R: Detected thread count of 647. Xeon W-3275M: Detected thread count of 568. 2 x Xeon Platinum 8280: Detected thread count of 1129. Xeon Gold 5220R: Detected thread count of 3610. 2 x Xeon Platinum 8275CL: Detected thread count of 9611. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9612. Xeon Platinum 8280: Detected thread count of 5613. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgFPS Per GHz, More Is BetterOSPray 1.8.5Performance Per Clock - Demo: San Miguel - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R320.38030.76061.14091.52121.90150.43750.44870.51790.54870.70650.82190.88000.95131.01001.49001.51001.63001.69001. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 2. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 7. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 8. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgFPS Per Channel, More Is BetterOSPray 1.8.5Performance Per Memory Channel - Demo: San Miguel - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon Platinum 8280EPYC 2 7R322 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 82800.24530.49060.73590.98121.22650.23330.29170.33670.35670.43830.58670.59130.61830.65831.09001. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 6226R: Detected channel count of 64. Xeon Gold 5220R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. EPYC 2 7R32: Detected channel count of 88. 2 x Xeon Gold 6226R: Detected channel count of 69. 2 x Xeon Gold 5220R: Detected channel count of 610. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82803691215Min: 1.4 / Avg: 1.4 / Max: 1.4Min: 1.75 / Avg: 1.75 / Max: 1.75Min: 2.02 / Avg: 2.02 / Max: 2.02Min: 2.14 / Avg: 2.14 / Max: 2.14Min: 2.62 / Avg: 2.63 / Max: 2.63Min: 3.25 / Avg: 3.25 / Max: 3.26Min: 3.51 / Avg: 3.52 / Max: 3.52Min: 3.7 / Avg: 3.71 / Max: 3.72Min: 3.95 / Avg: 3.95 / Max: 3.97Min: 4.72 / Avg: 4.73 / Max: 4.74Min: 5.81 / Avg: 5.82 / Max: 5.85Min: 5.85 / Avg: 5.87 / Max: 5.88Min: 6.49 / Avg: 6.53 / Max: 6.54

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 8275CL NUMA off2 x Xeon Platinum 8275CLXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275MXeon Platinum 8280EPYC 2 7R322 x Xeon Silver 42162 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8280714212835SE +/- 0.06827, N = 7SE +/- 0.60187, N = 9SE +/- 0.03282, N = 3SE +/- 0.01417, N = 3SE +/- 0.08231, N = 3SE +/- 0.02780, N = 3SE +/- 0.15186, N = 3SE +/- 0.04073, N = 3SE +/- 0.00395, N = 3SE +/- 0.00847, N = 3SE +/- 0.15366, N = 3SE +/- 0.12533, N = 3SE +/- 0.24605, N = 36.329718.4140511.8883013.0138013.3529013.9581014.1623014.9097017.0608023.1574025.6303026.5910028.854201. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s Per Core, More Is BetterHigh Performance Conjugate Gradient 3.1Performance Per Core2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CLEPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 5220RXeon Silver 4216Xeon Gold 52182 x Xeon Gold 6226RXeon Gold 6226R0.19630.39260.58890.78520.98150.13190.17530.35540.50580.51530.53250.71200.72370.74180.74300.81340.83100.87241. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 482. 2 x Xeon Platinum 8275CL: Detected core count of 483. EPYC 2 7R32: Detected core count of 484. Xeon W-3275M: Detected core count of 285. 2 x Xeon Platinum 8280: Detected core count of 566. Xeon Platinum 8280: Detected core count of 287. 2 x Xeon Gold 5220R: Detected core count of 368. 2 x Xeon Silver 4216: Detected core count of 329. Xeon Gold 5220R: Detected core count of 1810. Xeon Silver 4216: Detected core count of 1611. Xeon Gold 5218: Detected core count of 1612. 2 x Xeon Gold 6226R: Detected core count of 3213. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgGFLOP/s Per Thread, More Is BetterHigh Performance Conjugate Gradient 3.1Performance Per Thread2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CLEPYC 2 7R32Xeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 5220RXeon Silver 4216Xeon Gold 52182 x Xeon Gold 6226RXeon Gold 6226R0.09810.19620.29430.39240.49050.06590.08760.17770.25290.25760.26620.35600.36180.37090.37150.40670.41550.43621. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 962. 2 x Xeon Platinum 8275CL: Detected thread count of 963. EPYC 2 7R32: Detected thread count of 964. Xeon W-3275M: Detected thread count of 565. 2 x Xeon Platinum 8280: Detected thread count of 1126. Xeon Platinum 8280: Detected thread count of 567. 2 x Xeon Gold 5220R: Detected thread count of 728. 2 x Xeon Silver 4216: Detected thread count of 649. Xeon Gold 5220R: Detected thread count of 3610. Xeon Silver 4216: Detected thread count of 3211. Xeon Gold 5218: Detected thread count of 3212. 2 x Xeon Gold 6226R: Detected thread count of 6413. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgGFLOP/s Per GHz, More Is BetterHigh Performance Conjugate Gradient 3.1Performance Per Clock2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CLXeon W-3275MXeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon Silver 4216Xeon Platinum 8280EPYC 2 7R322 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 82802 x Xeon Silver 42162468101.622.163.083.343.423.583.723.736.096.576.827.217.241. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 2. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 3. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 4. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 6. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 7. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 8. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 9. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.8010. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20
OpenBenchmarking.orgGFLOP/s Per Channel, More Is BetterHigh Performance Conjugate Gradient 3.1Performance Per Memory ChannelXeon Silver 4216EPYC 2 7R32Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon Platinum 82802 x Xeon Silver 42162 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 82801.08232.16463.24694.32925.41151.982.132.172.232.332.483.864.274.434.811. Xeon Silver 4216: Detected channel count of 62. EPYC 2 7R32: Detected channel count of 83. Xeon Gold 5218: Detected channel count of 64. Xeon Gold 5220R: Detected channel count of 65. Xeon Gold 6226R: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. 2 x Xeon Silver 4216: Detected channel count of 68. 2 x Xeon Gold 5220R: Detected channel count of 69. 2 x Xeon Gold 6226R: Detected channel count of 610. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.12 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CLXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275MXeon Platinum 8280EPYC 2 7R322 x Xeon Silver 42162 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8280612182430Min: 6 / Avg: 6.33 / Max: 6.51Min: 6.44 / Avg: 8.41 / Max: 10.67Min: 11.82 / Avg: 11.89 / Max: 11.93Min: 13 / Avg: 13.01 / Max: 13.04Min: 13.19 / Avg: 13.35 / Max: 13.44Min: 13.92 / Avg: 13.96 / Max: 14.01Min: 13.94 / Avg: 14.16 / Max: 14.45Min: 14.87 / Avg: 14.91 / Max: 14.99Min: 17.05 / Avg: 17.06 / Max: 17.07Min: 23.15 / Avg: 23.16 / Max: 23.17Min: 25.33 / Avg: 25.63 / Max: 25.84Min: 26.34 / Avg: 26.59 / Max: 26.74Min: 28.36 / Avg: 28.85 / Max: 29.141. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

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 DragonXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 82801530456075SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.53, N = 4SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 315.5919.4322.4623.6028.9735.1538.3138.7740.4343.1051.8752.3568.42MIN: 15.54 / MAX: 15.7MIN: 19.37 / MAX: 19.6MIN: 22.38 / MAX: 22.62MIN: 23.45 / MAX: 23.8MIN: 28.72 / MAX: 29.29MIN: 33.95 / MAX: 35.72MIN: 37.87 / MAX: 38.72MIN: 35.85 / MAX: 45.98MIN: 40.04 / MAX: 41.18MIN: 42.68 / MAX: 43.7MIN: 50.96 / MAX: 54.15MIN: 51.67 / MAX: 53.48MIN: 66.22 / MAX: 70.39
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian DragonXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 82801326395265Min: 15.58 / Avg: 15.59 / Max: 15.62Min: 19.42 / Avg: 19.43 / Max: 19.45Min: 22.45 / Avg: 22.46 / Max: 22.48Min: 23.56 / Avg: 23.6 / Max: 23.64Min: 28.93 / Avg: 28.97 / Max: 29Min: 35.01 / Avg: 35.15 / Max: 35.27Min: 38.21 / Avg: 38.31 / Max: 38.41Min: 37.26 / Avg: 38.77 / Max: 39.72Min: 40.29 / Avg: 40.43 / Max: 40.52Min: 43.07 / Avg: 43.1 / Max: 43.14Min: 51.84 / Avg: 51.87 / Max: 51.92Min: 52.31 / Avg: 52.35 / Max: 52.41Min: 68.36 / Avg: 68.42 / Max: 68.47

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 82801326395265SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.63, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 313.4916.7719.3220.3425.0730.3533.0534.8036.6737.1544.7245.4358.69MIN: 13.43 / MAX: 13.59MIN: 16.69 / MAX: 16.88MIN: 19.25 / MAX: 19.46MIN: 20.23 / MAX: 20.53MIN: 24.86 / MAX: 25.34MIN: 29.95 / MAX: 30.77MIN: 32.72 / MAX: 33.4MIN: 34.51 / MAX: 35.32MIN: 34.56 / MAX: 39.51MIN: 36.74 / MAX: 37.69MIN: 43.9 / MAX: 46.32MIN: 44.82 / MAX: 46.53MIN: 56.97 / MAX: 60.91
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.6.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjXeon Silver 4216Xeon Gold 5218Xeon Gold 6226RXeon Gold 5220R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 82801224364860Min: 13.47 / Avg: 13.49 / Max: 13.52Min: 16.75 / Avg: 16.77 / Max: 16.78Min: 19.31 / Avg: 19.32 / Max: 19.34Min: 20.33 / Avg: 20.34 / Max: 20.35Min: 25.04 / Avg: 25.07 / Max: 25.09Min: 30.31 / Avg: 30.35 / Max: 30.38Min: 33.01 / Avg: 33.05 / Max: 33.08Min: 34.77 / Avg: 34.8 / Max: 34.84Min: 35.6 / Avg: 36.67 / Max: 37.78Min: 37.08 / Avg: 37.15 / Max: 37.2Min: 44.7 / Avg: 44.72 / Max: 44.73Min: 45.4 / Avg: 45.43 / Max: 45.5Min: 58.6 / Avg: 58.69 / Max: 58.78

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 PrismXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 82802 x Xeon Platinum 8275CL NUMA offEPYC 2 7R32246810SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 15SE +/- 0.04, N = 3SE +/- 0.13, N = 15SE +/- 0.30, N = 12SE +/- 0.11, N = 15SE +/- 0.40, N = 15SE +/- 0.05, N = 3SE +/- 0.06, N = 31.881.882.482.513.653.794.114.324.467.097.157.318.12MIN: 1.85 / MAX: 1.92MIN: 1.84 / MAX: 1.92MIN: 2.4 / MAX: 2.53MIN: 2.44 / MAX: 2.56MIN: 3.56 / MAX: 3.67MIN: 3.45 / MAX: 3.89MIN: 3.96 / MAX: 4.16MIN: 2.71 / MAX: 4.87MIN: 1.23 / MAX: 4.91MIN: 5.64 / MAX: 7.47MIN: 2.92 / MAX: 8.11MIN: 6.62 / MAX: 7.52MIN: 7.87 / MAX: 8.3
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: Rainbow Colors and PrismXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Gold 5220R2 x Xeon Platinum 8275CL2 x Xeon Platinum 82802 x Xeon Platinum 8275CL NUMA offEPYC 2 7R323691215Min: 1.88 / Avg: 1.88 / Max: 1.88Min: 1.88 / Avg: 1.88 / Max: 1.89Min: 2.45 / Avg: 2.48 / Max: 2.5Min: 2.49 / Avg: 2.51 / Max: 2.53Min: 3.64 / Avg: 3.65 / Max: 3.65Min: 3.48 / Avg: 3.79 / Max: 3.88Min: 4.03 / Avg: 4.11 / Max: 4.16Min: 2.88 / Avg: 4.32 / Max: 4.83Min: 1.25 / Avg: 4.46 / Max: 4.88Min: 5.76 / Avg: 7.09 / Max: 7.37Min: 2.98 / Avg: 7.15 / Max: 7.98Min: 7.22 / Avg: 7.31 / Max: 7.41Min: 8 / Avg: 8.12 / Max: 8.18

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-OnlyXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 828090180270360450SE +/- 0.29, N = 3SE +/- 0.20, N = 3SE +/- 0.39, N = 3SE +/- 0.16, N = 3SE +/- 0.19, N = 3SE +/- 0.32, N = 3SE +/- 0.36, N = 3SE +/- 0.18, N = 3SE +/- 0.52, N = 3SE +/- 0.11, N = 3SE +/- 0.29, N = 3SE +/- 0.40, N = 3SE +/- 0.26, N = 3433.42432.11328.81321.53218.06213.36198.25165.96163.32108.97108.94101.34100.47
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: Classroom - Compute: CPU-OnlyXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 828080160240320400Min: 433.1 / Avg: 433.42 / Max: 433.99Min: 431.84 / Avg: 432.11 / Max: 432.51Min: 328.21 / Avg: 328.81 / Max: 329.54Min: 321.25 / Avg: 321.53 / Max: 321.79Min: 217.79 / Avg: 218.06 / Max: 218.43Min: 212.88 / Avg: 213.36 / Max: 213.97Min: 197.81 / Avg: 198.25 / Max: 198.96Min: 165.68 / Avg: 165.96 / Max: 166.31Min: 162.78 / Avg: 163.32 / Max: 164.35Min: 108.85 / Avg: 108.97 / Max: 109.2Min: 108.61 / Avg: 108.94 / Max: 109.52Min: 100.56 / Avg: 101.34 / Max: 101.89Min: 99.99 / Avg: 100.47 / Max: 100.86

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 Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226REPYC 2 7R322 x Xeon Gold 5220RXeon W-3275M2 x Xeon Gold 6226RXeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 82801.19492.38983.58474.77965.9745SE +/- 0.01140, N = 3SE +/- 0.00570, N = 3SE +/- 0.00527, N = 3SE +/- 0.00576, N = 3SE +/- 0.00272, N = 3SE +/- 0.01972, N = 3SE +/- 0.00174, N = 3SE +/- 0.01273, N = 3SE +/- 0.00475, N = 3SE +/- 0.00157, N = 3SE +/- 0.00979, N = 3SE +/- 0.00024, N = 3SE +/- 0.00318, N = 35.310873.994883.364323.065902.231262.225402.126801.745151.497241.462321.331091.251161.24029MIN: 5.16MIN: 3.95MIN: 3.3MIN: 2.99MIN: 2.17MIN: 2.11MIN: 2.06MIN: 1.55MIN: 1.43MIN: 1.44MIN: 1.26MIN: 1.19MIN: 1.171. (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: f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220R2 x Xeon Silver 4216Xeon Gold 6226REPYC 2 7R322 x Xeon Gold 5220RXeon W-3275M2 x Xeon Gold 6226RXeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280246810Min: 5.29 / Avg: 5.31 / Max: 5.33Min: 3.98 / Avg: 3.99 / Max: 4Min: 3.36 / Avg: 3.36 / Max: 3.37Min: 3.06 / Avg: 3.07 / Max: 3.08Min: 2.23 / Avg: 2.23 / Max: 2.24Min: 2.19 / Avg: 2.23 / Max: 2.26Min: 2.12 / Avg: 2.13 / Max: 2.13Min: 1.73 / Avg: 1.75 / Max: 1.77Min: 1.49 / Avg: 1.5 / Max: 1.5Min: 1.46 / Avg: 1.46 / Max: 1.47Min: 1.32 / Avg: 1.33 / Max: 1.35Min: 1.25 / Avg: 1.25 / Max: 1.25Min: 1.23 / Avg: 1.24 / Max: 1.251. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

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 AtomsXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 82800.34140.68281.02421.36561.707SE +/- 0.00193, N = 3SE +/- 0.00100, N = 3SE +/- 0.00122, N = 3SE +/- 0.00117, N = 3SE +/- 0.00775, N = 15SE +/- 0.00193, N = 3SE +/- 0.00060, N = 3SE +/- 0.00612, N = 3SE +/- 0.00104, N = 3SE +/- 0.00084, N = 3SE +/- 0.00133, N = 31.517211.511681.157451.135860.814340.769820.707090.603330.582370.486140.36226
OpenBenchmarking.orgdays/ns x Core, Fewer Is BetterNAMD 2.13Performance Per Core - ATPase Simulation - 327,506 Atoms2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 5218EPYC 2 7R32Xeon W-3275M2 x Xeon Gold 5220RXeon Gold 5220R2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 6226RXeon Gold 6226R61218243024.6324.2824.1923.3422.8021.7220.8320.2919.8018.6418.171. 2 x Xeon Silver 4216: Detected core count of 322. Xeon Silver 4216: Detected core count of 163. Xeon Gold 5218: Detected core count of 164. EPYC 2 7R32: Detected core count of 485. Xeon W-3275M: Detected core count of 286. 2 x Xeon Gold 5220R: Detected core count of 367. Xeon Gold 5220R: Detected core count of 188. 2 x Xeon Platinum 8280: Detected core count of 569. Xeon Platinum 8280: Detected core count of 2810. 2 x Xeon Gold 6226R: Detected core count of 3211. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgdays/ns x Thread, Fewer Is BetterNAMD 2.13Performance Per Thread - ATPase Simulation - 327,506 Atoms2 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 5218EPYC 2 7R32Xeon W-3275M2 x Xeon Gold 5220RXeon Gold 5220R2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Gold 6226RXeon Gold 6226R112233445549.2748.5548.3746.6745.6043.4441.6740.5739.6037.2736.351. 2 x Xeon Silver 4216: Detected thread count of 642. Xeon Silver 4216: Detected thread count of 323. Xeon Gold 5218: Detected thread count of 324. EPYC 2 7R32: Detected thread count of 965. Xeon W-3275M: Detected thread count of 566. 2 x Xeon Gold 5220R: Detected thread count of 727. Xeon Gold 5220R: Detected thread count of 368. 2 x Xeon Platinum 8280: Detected thread count of 1129. Xeon Platinum 8280: Detected thread count of 5610. 2 x Xeon Gold 6226R: Detected thread count of 6411. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgdays/ns x GHz, Fewer Is BetterNAMD 2.13Performance Per Clock - ATPase Simulation - 327,506 AtomsXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226RXeon W-3275MXeon Platinum 82802 x Xeon Silver 42162 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8280EPYC 2 7R321.32662.65323.97985.30646.6335.8964.8554.5144.4303.7462.8282.4632.3532.2711.4491.3611. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 2. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 3. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 7. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 8. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 11. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgdays/ns x Channel, Fewer Is BetterNAMD 2.13Performance Per Memory Channel - ATPase Simulation - 327,506 AtomsXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon Platinum 8280EPYC 2 7R322 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 828036912159.1039.0706.9456.8154.6194.2433.8893.6203.4942.1741. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 5220R: Detected channel count of 64. Xeon Gold 6226R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. EPYC 2 7R32: Detected channel count of 88. 2 x Xeon Gold 5220R: Detected channel count of 69. 2 x Xeon Gold 6226R: Detected channel count of 610. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.13ATPase Simulation - 327,506 AtomsXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8280246810Min: 1.51 / Avg: 1.52 / Max: 1.52Min: 1.51 / Avg: 1.51 / Max: 1.51Min: 1.16 / Avg: 1.16 / Max: 1.16Min: 1.13 / Avg: 1.14 / Max: 1.14Min: 0.78 / Avg: 0.81 / Max: 0.87Min: 0.77 / Avg: 0.77 / Max: 0.77Min: 0.71 / Avg: 0.71 / Max: 0.71Min: 0.59 / Avg: 0.6 / Max: 0.62Min: 0.58 / Avg: 0.58 / Max: 0.58Min: 0.48 / Avg: 0.49 / Max: 0.49Min: 0.36 / Avg: 0.36 / Max: 0.36

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: DLSCXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R32246810SE +/- 0.0034, N = 3SE +/- 0.0165, N = 3SE +/- 0.0243, N = 3SE +/- 0.0145, N = 3SE +/- 0.0265, N = 3SE +/- 0.0035, N = 3SE +/- 0.0032, N = 3SE +/- 0.0120, N = 3SE +/- 0.0036, N = 3SE +/- 0.0262, N = 3SE +/- 0.0164, N = 3SE +/- 0.0948, N = 3SE +/- 0.0970, N = 31.74001.77002.32002.35003.43003.53003.79004.44004.51836.73006.75007.19007.2800MIN: 1.67 / MAX: 1.79MIN: 1.68 / MAX: 1.85MIN: 2.2 / MAX: 2.43MIN: 2.24 / MAX: 2.46MIN: 3.33 / MAX: 3.63MIN: 3.46 / MAX: 3.66MIN: 3.65 / MAX: 3.94MIN: 4.34 / MAX: 4.63MIN: 4.4 / MAX: 4.76MIN: 6.54 / MAX: 7.08MIN: 6.62 / MAX: 7.08MIN: 6.83 / MAX: 7.6MIN: 7.01 / MAX: 7.86
OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.3Scene: DLSCXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R323691215Min: 1.74 / Avg: 1.74 / Max: 1.75Min: 1.75 / Avg: 1.77 / Max: 1.8Min: 2.29 / Avg: 2.32 / Max: 2.36Min: 2.33 / Avg: 2.35 / Max: 2.38Min: 3.39 / Avg: 3.43 / Max: 3.48Min: 3.52 / Avg: 3.53 / Max: 3.53Min: 3.79 / Avg: 3.79 / Max: 3.79Min: 4.42 / Avg: 4.44 / Max: 4.46Min: 4.51 / Avg: 4.52 / Max: 4.53Min: 6.68 / Avg: 6.73 / Max: 6.76Min: 6.72 / Avg: 6.75 / Max: 6.78Min: 7 / Avg: 7.19 / Max: 7.31Min: 7.13 / Avg: 7.28 / Max: 7.46

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: bf16bf16bf16Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Gold 5220RXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off48121620SE +/- 0.00867, N = 3SE +/- 0.00105, N = 3SE +/- 0.04579, N = 3SE +/- 0.00392, N = 3SE +/- 0.01001, N = 3SE +/- 0.03740, N = 3SE +/- 0.00772, N = 3SE +/- 0.00167, N = 3SE +/- 0.00747, N = 3SE +/- 0.00661, N = 3SE +/- 0.01030, N = 3SE +/- 0.00401, N = 316.5516016.5484013.142809.903628.869137.615957.229756.553745.384024.068473.969063.96510MIN: 16.51MIN: 16.51MIN: 12.96MIN: 9.85MIN: 8.78MIN: 6.95MIN: 7.13MIN: 6.5MIN: 5.3MIN: 3.93MIN: 3.86MIN: 3.881. (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 Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275M2 x Xeon Gold 5220RXeon Platinum 82802 x Xeon Gold 6226R2 x Xeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off48121620Min: 16.54 / Avg: 16.55 / Max: 16.57Min: 16.55 / Avg: 16.55 / Max: 16.55Min: 13.07 / Avg: 13.14 / Max: 13.23Min: 9.9 / Avg: 9.9 / Max: 9.91Min: 8.86 / Avg: 8.87 / Max: 8.89Min: 7.54 / Avg: 7.62 / Max: 7.66Min: 7.21 / Avg: 7.23 / Max: 7.24Min: 6.55 / Avg: 6.55 / Max: 6.56Min: 5.37 / Avg: 5.38 / Max: 5.4Min: 4.06 / Avg: 4.07 / Max: 4.08Min: 3.95 / Avg: 3.97 / Max: 3.98Min: 3.96 / Avg: 3.97 / Max: 3.971. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

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: MD5Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216EPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 82802M4M6M8M10MSE +/- 1452.97, N = 3SE +/- 8171.77, N = 3SE +/- 6806.86, N = 3SE +/- 11015.14, N = 3SE +/- 14294.52, N = 3SE +/- 4509.25, N = 3SE +/- 30550.50, N = 3SE +/- 14468.36, N = 3SE +/- 9643.65, N = 3SE +/- 17032.65, N = 3SE +/- 39218.76, N = 3SE +/- 5783.12, N = 3SE +/- 67263.16, N = 318933331912667250600025830003654000376400039720004090000482100049776677350667738833378800001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
OpenBenchmarking.orgReal C/S Per Core, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Core - Test: MD5EPYC 2 7R322 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 5218Xeon W-3275M2 x Xeon Gold 5220RXeon Gold 5220R2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Gold 6226RXeon Gold 6226R30K60K90K120K150K82750.00117625.00118333.31119541.69130500.00133916.67139222.22140714.29146071.43153138.90153923.60155552.09161437.501. EPYC 2 7R32: Detected core count of 482. 2 x Xeon Silver 4216: Detected core count of 323. Xeon Silver 4216: Detected core count of 164. Xeon Gold 5218: Detected core count of 165. Xeon W-3275M: Detected core count of 286. 2 x Xeon Gold 5220R: Detected core count of 367. Xeon Gold 5220R: Detected core count of 188. 2 x Xeon Platinum 8280: Detected core count of 569. Xeon Platinum 8280: Detected core count of 2810. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4811. 2 x Xeon Platinum 8275CL: Detected core count of 4812. 2 x Xeon Gold 6226R: Detected core count of 3213. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgReal C/S Per Thread, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Thread - Test: MD5EPYC 2 7R322 x Xeon Silver 4216Xeon Silver 4216Xeon Gold 5218Xeon W-3275M2 x Xeon Gold 5220RXeon Gold 5220R2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Gold 6226RXeon Gold 6226R20K40K60K80K100K41375.0058812.5059166.6659770.8465250.0066958.3369611.1170357.1473035.7176569.4576961.8077776.0580718.751. EPYC 2 7R32: Detected thread count of 962. 2 x Xeon Silver 4216: Detected thread count of 643. Xeon Silver 4216: Detected thread count of 324. Xeon Gold 5218: Detected thread count of 325. Xeon W-3275M: Detected thread count of 566. 2 x Xeon Gold 5220R: Detected thread count of 727. Xeon Gold 5220R: Detected thread count of 368. 2 x Xeon Platinum 8280: Detected thread count of 1129. Xeon Platinum 8280: Detected thread count of 5610. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9611. 2 x Xeon Platinum 8275CL: Detected thread count of 9612. 2 x Xeon Gold 6226R: Detected thread count of 6413. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgReal C/S Per GHz, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Clock - Test: MD5Xeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226RXeon W-3275MXeon Platinum 82802 x Xeon Silver 42162 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 8280400K800K1200K1600K2000K490427.44591666.56642564.10662307.69794347.831022500.001176250.001236153.851276324.871418571.431884786.411894444.361970000.001. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 2. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 3. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 7. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 8. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 13. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00
OpenBenchmarking.orgReal C/S Per Channel, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Memory Channel - Test: MD5Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226REPYC 2 7R322 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8280300K600K900K1200K1500K315555.50318777.83417666.67430500.00496500.00627333.33681666.67803500.00829611.171313333.331. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 5220R: Detected channel count of 64. Xeon Gold 6226R: Detected channel count of 65. EPYC 2 7R32: Detected channel count of 86. 2 x Xeon Silver 4216: Detected channel count of 67. Xeon Platinum 8280: Detected channel count of 68. 2 x Xeon Gold 5220R: Detected channel count of 69. 2 x Xeon Gold 6226R: Detected channel count of 610. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: MD5Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216EPYC 2 7R32Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL2 x Xeon Platinum 82801.4M2.8M4.2M5.6M7MMin: 1891000 / Avg: 1893333.33 / Max: 1896000Min: 1904000 / Avg: 1912666.67 / Max: 1929000Min: 2493000 / Avg: 2506000 / Max: 2516000Min: 2561000 / Avg: 2583000 / Max: 2595000Min: 3626000 / Avg: 3654000 / Max: 3673000Min: 3755000 / Avg: 3764000 / Max: 3769000Min: 3932000 / Avg: 3972000 / Max: 4032000Min: 4066000 / Avg: 4090000 / Max: 4116000Min: 4806000 / Avg: 4821000 / Max: 4839000Min: 4944000 / Avg: 4977666.67 / Max: 4999000Min: 7273000 / Avg: 7350666.67 / Max: 7399000Min: 7377000 / Avg: 7388333.33 / Max: 7396000Min: 7749000 / Avg: 7880000 / Max: 79720001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

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-OnlyXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 8280120240360480600SE +/- 0.31, N = 3SE +/- 0.46, N = 3SE +/- 0.03, N = 3SE +/- 0.22, N = 3SE +/- 0.50, N = 3SE +/- 0.73, N = 3SE +/- 0.05, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.11, N = 3SE +/- 0.51, N = 3SE +/- 0.26, N = 3531.98531.39404.78400.96272.72265.41246.31208.30207.41139.11139.00132.04129.01
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.82Blend File: Pabellon Barcelona - Compute: CPU-OnlyXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA offEPYC 2 7R322 x Xeon Platinum 828090180270360450Min: 531.42 / Avg: 531.98 / Max: 532.47Min: 530.72 / Avg: 531.39 / Max: 532.27Min: 404.72 / Avg: 404.78 / Max: 404.82Min: 400.53 / Avg: 400.96 / Max: 401.29Min: 272.04 / Avg: 272.72 / Max: 273.7Min: 264.01 / Avg: 265.41 / Max: 266.46Min: 246.23 / Avg: 246.31 / Max: 246.39Min: 208.1 / Avg: 208.3 / Max: 208.45Min: 207.29 / Avg: 207.41 / Max: 207.58Min: 138.96 / Avg: 139.11 / Max: 139.4Min: 138.83 / Avg: 139 / Max: 139.2Min: 131.01 / Avg: 132.04 / Max: 132.59Min: 128.65 / Avg: 129.01 / Max: 129.51

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: BlowfishXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828020K40K60K80K100KSE +/- 24.29, N = 3SE +/- 6.03, N = 3SE +/- 77.98, N = 3SE +/- 60.93, N = 3SE +/- 149.28, N = 3SE +/- 91.33, N = 3SE +/- 99.39, N = 3SE +/- 12.50, N = 3SE +/- 111.31, N = 3SE +/- 203.51, N = 3SE +/- 465.66, N = 3SE +/- 78.50, N = 3SE +/- 700.53, N = 3204242140325547271264032641021428645132453485664757777178327840841. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
OpenBenchmarking.orgReal C/S Per Core, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Core - Test: BlowfishXeon Silver 42162 x Xeon Silver 4216Xeon Gold 5218EPYC 2 7R32Xeon Gold 5220R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Gold 6226RXeon Gold 6226R4008001200160020001276.501281.911337.691384.901419.281425.671440.211501.501530.861620.231631.811671.411695.381. Xeon Silver 4216: Detected core count of 162. 2 x Xeon Silver 4216: Detected core count of 323. Xeon Gold 5218: Detected core count of 164. EPYC 2 7R32: Detected core count of 485. Xeon Gold 5220R: Detected core count of 186. 2 x Xeon Gold 5220R: Detected core count of 367. Xeon W-3275M: Detected core count of 288. 2 x Xeon Platinum 8280: Detected core count of 569. Xeon Platinum 8280: Detected core count of 2810. 2 x Xeon Platinum 8275CL: Detected core count of 4811. 2 x Xeon Platinum 8275CL NUMA off: Detected core count of 4812. 2 x Xeon Gold 6226R: Detected core count of 3213. Xeon Gold 6226R: Detected core count of 16
OpenBenchmarking.orgReal C/S Per Thread, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Thread - Test: BlowfishXeon Silver 42162 x Xeon Silver 4216Xeon Gold 5218EPYC 2 7R32Xeon Gold 5220R2 x Xeon Gold 5220RXeon W-3275M2 x Xeon Platinum 8280Xeon Platinum 82802 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Gold 6226RXeon Gold 6226R2004006008001000638.25640.95668.84692.45709.64712.83720.11750.75765.43810.11815.91835.70847.691. Xeon Silver 4216: Detected thread count of 322. 2 x Xeon Silver 4216: Detected thread count of 643. Xeon Gold 5218: Detected thread count of 324. EPYC 2 7R32: Detected thread count of 965. Xeon Gold 5220R: Detected thread count of 366. 2 x Xeon Gold 5220R: Detected thread count of 727. Xeon W-3275M: Detected thread count of 568. 2 x Xeon Platinum 8280: Detected thread count of 1129. Xeon Platinum 8280: Detected thread count of 5610. 2 x Xeon Platinum 8275CL: Detected thread count of 9611. 2 x Xeon Platinum 8275CL NUMA off: Detected thread count of 9612. 2 x Xeon Gold 6226R: Detected thread count of 6413. Xeon Gold 6226R: Detected thread count of 32
OpenBenchmarking.orgReal C/S Per GHz, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Clock - Test: BlowfishXeon Gold 5218Xeon Silver 4216Xeon Gold 5220RXeon Gold 6226RXeon W-3275MXeon Platinum 82802 x Xeon Silver 42162 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8280EPYC 2 7R325K10K15K20K25K5487.956382.506550.516955.388766.5210716.0012819.0613160.0013714.1019941.2820083.8521021.0023741.071. Xeon Gold 5218: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 2. Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 3. Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 4. Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 5. Xeon W-3275M: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.60 6. Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 7. 2 x Xeon Silver 4216: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.20 8. 2 x Xeon Gold 5220R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 9. 2 x Xeon Gold 6226R: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 10. 2 x Xeon Platinum 8275CL: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 11. 2 x Xeon Platinum 8275CL NUMA off: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.90 12. 2 x Xeon Platinum 8280: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 4.00 13. EPYC 2 7R32: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.80
OpenBenchmarking.orgReal C/S Per Channel, More Is BetterJohn The Ripper 1.9.0-jumbo-1Performance Per Memory Channel - Test: BlowfishXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon Platinum 8280EPYC 2 7R322 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 82803K6K9K12K15K3404.003567.174257.834521.006836.837144.008309.388554.008914.1714014.001. Xeon Silver 4216: Detected channel count of 62. Xeon Gold 5218: Detected channel count of 63. Xeon Gold 5220R: Detected channel count of 64. Xeon Gold 6226R: Detected channel count of 65. 2 x Xeon Silver 4216: Detected channel count of 66. Xeon Platinum 8280: Detected channel count of 67. EPYC 2 7R32: Detected channel count of 88. 2 x Xeon Gold 5220R: Detected channel count of 69. 2 x Xeon Gold 6226R: Detected channel count of 610. 2 x Xeon Platinum 8280: Detected channel count of 6
OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.9.0-jumbo-1Test: BlowfishXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon W-3275M2 x Xeon Silver 4216Xeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226REPYC 2 7R322 x Xeon Platinum 8275CL2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 828015K30K45K60K75KMin: 20383 / Avg: 20423.67 / Max: 20467Min: 21391 / Avg: 21403 / Max: 21410Min: 25414 / Avg: 25546.67 / Max: 25684Min: 27043 / Avg: 27126.33 / Max: 27245Min: 40071 / Avg: 40326.33 / Max: 40588Min: 40838 / Avg: 41020.67 / Max: 41112Min: 42674 / Avg: 42863.67 / Max: 43010Min: 51304 / Avg: 51324 / Max: 51347Min: 53299 / Avg: 53485.33 / Max: 53684Min: 66240 / Avg: 66474.67 / Max: 66880Min: 76870 / Avg: 77771.33 / Max: 78425Min: 78220 / Avg: 78327 / Max: 78480Min: 82930 / Avg: 84083.67 / Max: 853491. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2

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: u8s8f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CLEPYC 2 7R322 x Xeon Platinum 82803K6K9K12K15KSE +/- 13.03, N = 3SE +/- 2.59, N = 3SE +/- 11.33, N = 3SE +/- 9.11, N = 3SE +/- 109.20, N = 3SE +/- 41.02, N = 3SE +/- 11.11, N = 3SE +/- 3.52, N = 3SE +/- 10.40, N = 3SE +/- 0.59, N = 3SE +/- 4.78, N = 3SE +/- 25.92, N = 3SE +/- 7.97, N = 312300.3011821.309696.929374.606329.946183.115919.944896.414821.243252.483250.893076.763005.48MIN: 12254MIN: 11811.3MIN: 9656.48MIN: 9350.22MIN: 6190.58MIN: 5934.01MIN: 5895.99MIN: 4854.07MIN: 4760.44MIN: 3175.14MIN: 3176.52MIN: 3011.81MIN: 2934.551. (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_3d - Data Type: u8s8f32Xeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226R2 x Xeon Silver 4216Xeon W-3275MXeon Platinum 82802 x Xeon Gold 5220R2 x Xeon Gold 6226R2 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CLEPYC 2 7R322 x Xeon Platinum 82802K4K6K8K10KMin: 12276.3 / Avg: 12300.27 / Max: 12321.1Min: 11817.7 / Avg: 11821.27 / Max: 11826.3Min: 9679.29 / Avg: 9696.92 / Max: 9718.07Min: 9357.68 / Avg: 9374.6 / Max: 9388.89Min: 6214.71 / Avg: 6329.94 / Max: 6548.22Min: 6101.37 / Avg: 6183.11 / Max: 6230.15Min: 5908.76 / Avg: 5919.94 / Max: 5942.15Min: 4889.36 / Avg: 4896.41 / Max: 4900.09Min: 4810.42 / Avg: 4821.24 / Max: 4842.04Min: 3251.32 / Avg: 3252.48 / Max: 3253.21Min: 3241.39 / Avg: 3250.89 / Max: 3256.56Min: 3025.64 / Avg: 3076.76 / Max: 3109.75Min: 2989.81 / Avg: 3005.48 / Max: 3015.871. (CXX) g++ options: -O3 -march=native -std=c++11 -msse4.1 -fPIC -fopenmp -pie -lpthread -ldl

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 PingPongXeon Silver 4216Xeon Gold 5218Xeon Gold 5220RXeon Gold 6226RXeon Platinum 82802 x Xeon Silver 42162 x Xeon Gold 5220RXeon W-3275MEPYC 2 7R322 x Xeon Gold 6226R2 x Xeon Platinum 82802 x Xeon Platinum 8275CL NUMA off2 x Xeon Platinum 8275CL8M16M24M32M40M