Ice Lake Ubuntu 20.04.2 / 20.10 / 21.04 Linux Benchmarks

2 x Intel Xeon Platinum 8380 testing with a Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS) and ASPEED on Fedora 34 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 2105196-IB-2105195IB50
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 3 Tests
C++ Boost Tests 2 Tests
Chess Test Suite 2 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 15 Tests
CPU Massive 25 Tests
Creator Workloads 15 Tests
Cryptography 2 Tests
Database Test Suite 3 Tests
Encoding 7 Tests
Fortran Tests 5 Tests
Game Development 3 Tests
HPC - High Performance Computing 13 Tests
Java 2 Tests
Machine Learning 3 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 6 Tests
Multi-Core 31 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 6 Tests
OpenMPI Tests 9 Tests
Programmer / Developer System Benchmarks 6 Tests
Python Tests 2 Tests
Raytracing 2 Tests
Renderers 4 Tests
Scientific Computing 6 Tests
Server 6 Tests
Server CPU Tests 19 Tests
Single-Threaded 2 Tests
Video Encoding 7 Tests
Common Workstation Benchmarks 4 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
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Ubuntu 21.04
May 13 2021
  10 Hours, 46 Minutes
Ubuntu 20.04.2 LTS
May 17 2021
  12 Hours, 22 Minutes
Clear Linux 34630
May 18 2021
  7 Hours, 41 Minutes
Fedora Server 34
May 19 2021
  10 Hours, 54 Minutes
Invert Hiding All Results Option
  10 Hours, 26 Minutes

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

Ubuntu 21.04

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

OS: Ubuntu 21.04, Kernel: 5.11.0-17-generic (x86_64), Desktop: GNOME Shell 3.38.4, Display Server: X Server, Compiler: GCC 10.3.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-gcn/usr,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-build-config=bootstrap-lto-lean --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
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2)
Python Notes: Python 3.9.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Ubuntu 20.04.2 LTS

OS: Ubuntu 20.04, Kernel: 5.8.0-53-generic (x86_64), Desktop: GNOME Shell 3.36.7, Display Server: X Server 1.20.9, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,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
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2.20.04)
Python Notes: Python 3.8.5
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Clear Linux 34630

OS: Clear Linux OS 34630, Kernel: 5.10.19-1032.native (x86_64), Desktop: GNOME Shell 40.0, Display Server: X Server, Compiler: GCC 11.1.1 20210517 releases/gcc-11.1.0-132-g7d91dd2efb + Clang 11.1.0 + LLVM 11.1.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags -Wa,-mbranches-within-32B-boundaries" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries -fvisibility-inlines-hidden -Wl,--enable-new-dtags" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-languages=c,c++,fortran,go --enable-ld=default --enable-libstdcxx-pch --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=westmere --with-gcc-major-version-only --with-glibc-version=2.19 --with-gnu-ld --with-isl --with-ppl=yes --with-tune=skylake-avx512
Disk Notes: MQ-DEADLINE / relatime,rw,stripe=256 / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate performance - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment (build 1.8.0-u252-ga-b00)
Python Notes: Python 3.9.5
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Fedora Server 34

OS: Fedora 34, Kernel: 5.11.20-300.fc34.x86_64 (x86_64), Compiler: GCC 11.1.1 20210428, File-System: xfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver
Disk Notes: NONE / attr2,inode64,logbsize=32k,logbufs=8,noquota,relatime,rw,seclabel / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment 18.9 (build 11.0.11+9)
Python Notes: Python 3.9.5
Security Notes: SELinux + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Ice Lake Ubuntu 20.04.2 / 20.10 / 21.04 Linux Benchmarks ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionUbuntu 21.04Ubuntu 20.04.2 LTSClear Linux 34630Fedora Server 342 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads)Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS)Intel Device 0998504GB800GB INTEL SSDPF21Q800GBASPEEDVE2282 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFPUbuntu 21.045.11.0-17-generic (x86_64)GNOME Shell 3.38.4X ServerGCC 10.3.0ext41920x1080Ubuntu 20.045.8.0-53-generic (x86_64)GNOME Shell 3.36.7X Server 1.20.9GCC 9.3.0Clear Linux OS 346305.10.19-1032.native (x86_64)GNOME Shell 40.0X ServerGCC 11.1.1 20210517 releases/gcc-11.1.0-132-g7d91dd2efb + Clang 11.1.0 + LLVM 11.1.0Fedora 345.11.20-300.fc34.x86_64 (x86_64)GCC 11.1.1 20210428xfsOpenBenchmarking.orgKernel Details- Ubuntu 21.04: Transparent Huge Pages: madvise- Ubuntu 20.04.2 LTS: Transparent Huge Pages: madvise- Clear Linux 34630: Transparent Huge Pages: always- Fedora Server 34: Transparent Huge Pages: madviseCompiler Details- Ubuntu 21.04: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-gcn/usr,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-build-config=bootstrap-lto-lean --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 - Ubuntu 20.04.2 LTS: --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=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,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 - Clear Linux 34630: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-languages=c,c++,fortran,go --enable-ld=default --enable-libstdcxx-pch --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=westmere --with-gcc-major-version-only --with-glibc-version=2.19 --with-gnu-ld --with-isl --with-ppl=yes --with-tune=skylake-avx512 - Fedora Server 34: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver Disk Details- Ubuntu 21.04: NONE / errors=remount-ro,relatime,rw / Block Size: 4096- Ubuntu 20.04.2 LTS: NONE / errors=remount-ro,relatime,rw / Block Size: 4096- Clear Linux 34630: MQ-DEADLINE / relatime,rw,stripe=256 / Block Size: 4096- Fedora Server 34: NONE / attr2,inode64,logbsize=32k,logbufs=8,noquota,relatime,rw,seclabel / Block Size: 4096Processor Details- Ubuntu 21.04: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270- Ubuntu 20.04.2 LTS: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270- Clear Linux 34630: Scaling Governor: intel_pstate performance - CPU Microcode: 0xd000270- Fedora Server 34: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270Java Details- Ubuntu 21.04: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2)- Ubuntu 20.04.2 LTS: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2.20.04)- Clear Linux 34630: OpenJDK Runtime Environment (build 1.8.0-u252-ga-b00)- Fedora Server 34: OpenJDK Runtime Environment 18.9 (build 11.0.11+9)Python Details- Ubuntu 21.04: Python 3.9.4- Ubuntu 20.04.2 LTS: Python 3.8.5- Clear Linux 34630: Python 3.9.5- Fedora Server 34: Python 3.9.5Security Details- Ubuntu 21.04: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected - Ubuntu 20.04.2 LTS: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected - Clear Linux 34630: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected - Fedora Server 34: SELinux + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected Environment Details- Clear Linux 34630: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags -Wa,-mbranches-within-32B-boundaries" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries -fvisibility-inlines-hidden -Wl,--enable-new-dtags" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""

Ubuntu 21.04Ubuntu 20.04.2 LTSClear Linux 34630Fedora Server 34Result OverviewPhoronix Test Suite100%162%224%285%347%NAS Parallel BenchmarksApache Cassandralibgav1x265KvazaarDaCapo BenchmarkSVT-VP9SVT-AV1PJSIPSVT-HEVCTimed Apache Compilationlibavif avifencTungsten RendererTimed PHP CompilationTimed Linux Kernel CompilationJava Gradle BuildPHPBenchLiquid-DSPNode.js V8 Web Tooling BenchmarkOpenVKLIntel Open Image DenoisePyBenchRodiniaLAMMPS Molecular Dynamics SimulatorBlogBenchXcompact3d Incompact3dTNNPostgreSQL pgbenchoneDNNGROMACSCoremarkZstd CompressionTimed MrBayes AnalysisEmbreeOSPrayStockfishPOV-RayChia Blockchain VDFHelsingBlenderRELIONasmFishHigh Performance Conjugate GradientSecureMarkSysbenchNAMDTensorFlow Lite

Ice Lake Ubuntu 20.04.2 / 20.10 / 21.04 Linux Benchmarks npb: LU.Cdacapobench: Tradebeansonednn: Deconvolution Batch shapes_1d - f32 - CPUkvazaar: Bosphorus 4K - Very Fastx265: Bosphorus 1080ppjsip: OPTIONS, Statefulpjsip: INVITElibgav1: Summer Nature 4Kdacapobench: H2kvazaar: Bosphorus 4K - Mediumsvt-hevc: 10 - Bosphorus 1080px265: Bosphorus 4Ksvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080ptungsten: Non-Exponentialsvt-av1: Preset 8 - Bosphorus 4Kkvazaar: Bosphorus 4K - Ultra Fastsvt-hevc: 7 - Bosphorus 1080ptungsten: Water Causticbuild-apache: Time To Compileavifenc: 6compress-zstd: 3 - Compression Speedtungsten: Hairavifenc: 6, Losslesstensorflow-lite: Mobilenet Floattensorflow-lite: Mobilenet Quantbuild-php: Time To Compileopenvkl: vklBenchmarkVdbVolumeonednn: Convolution Batch Shapes Auto - f32 - CPUsvt-hevc: 1 - Bosphorus 1080pphpbench: PHP Benchmark Suiteliquid-dsp: 160 - 256 - 57compress-zstd: 19 - Decompression Speedopenvkl: vklBenchmarktensorflow-lite: NASNet Mobileblogbench: Writenwchem: C240 Buckyballnode-web-tooling: liquid-dsp: 32 - 256 - 57liquid-dsp: 16 - 256 - 57liquid-dsp: 128 - 256 - 57blogbench: Readpybench: Total For Average Test Timesospray: NASA Streamlines - SciVisembree: Pathtracer ISPC - Crowncompress-zstd: 19, Long Mode - Decompression Speedospray: Magnetic Reconnection - SciViscompress-zstd: 19, Long Mode - Compression Speedincompact3d: input.i3d 193 Cells Per Directionincompact3d: X3D-benchmarking input.i3donednn: IP Shapes 3D - f32 - CPUospray: San Miguel - SciViscompress-zstd: 19 - Compression Speedpjsip: OPTIONS, Statelessgromacs: MPI CPU - water_GMX50_barechia-vdf: Square Plain C++coremark: CoreMark Size 666 - Iterations Per Secondonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUtensorflow-lite: Inception V4onednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUrodinia: OpenMP LavaMDcompress-zstd: 19 - Compression Speedonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUmrbayes: Primate Phylogeny Analysisblender: Barbershop - CPU-Onlycompress-zstd: 19, Long Mode - Compression Speedstockfish: Total Timepovray: Trace Timeonednn: IP Shapes 3D - bf16bf16bf16 - CPUchia-vdf: Square Assembly Optimizedopenvkl: vklBenchmarkUnstructuredVolumeblender: Fishy Cat - CPU-Onlyblender: BMW27 - CPU-Onlyhelsing: 14 digittensorflow-lite: SqueezeNetsysbench: RAM / Memoryonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: IP Shapes 1D - f32 - CPUrelion: Basic - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUembree: Pathtracer - Asian Dragononednn: Recurrent Neural Network Training - u8s8f32 - CPUospray: NASA Streamlines - Path Traceronednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Training - f32 - CPUblender: Classroom - CPU-Onlyonednn: Recurrent Neural Network Inference - f32 - CPUblender: Pabellon Barcelona - CPU-Onlyhpcg: securemark: SecureMark-TLSonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPUlammps: 20k Atomsonednn: Deconvolution Batch shapes_3d - f32 - CPUospray: San Miguel - Path Traceronednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUembree: Pathtracer ISPC - Asian Dragontensorflow-lite: Inception ResNet V2rodinia: OpenMP HotSpot3Donednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUsysbench: CPUtnn: CPU - SqueezeNet v1.1namd: ATPase Simulation - 327,506 Atomsospray: XFrog Forest - Path Tracercompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 19 - Decompression Speedcompress-zstd: 8 - Decompression Speedcompress-zstd: 3 - Decompression Speedospray: XFrog Forest - SciViscompress-zstd: 8 - Compression Speedcassandra: Writestnn: CPU - MobileNet v2pgbench: 100 - 250 - Read Only - Average Latencypgbench: 100 - 250 - Read Onlykeydb: tungsten: Volumetric Causticbuild-linux-kernel: Time To Compileavifenc: 10, Losslessavifenc: 10asmfish: 1024 Hash Memory, 26 Depthoidn: Memorialdacapobench: Jythonjava-gradle-perf: Reactorlammps: Rhodopsin Proteinrodinia: OpenMP Leukocytenpb: EP.CUbuntu 21.04Ubuntu 20.04.2 LTSClear Linux 34630Fedora Server 34187807.081672928.751713.3828.333815253119.39107406.90260.8812.92204.865.2391229.80922.78172.0231.273935.18715.3836148.46.5253836.76441555.143074.540.207219158371.4108429.5771558730806333332727.666981319.3607681875.810.7916311666678267566673290766667225381099512567.03952719.4109.6347.211.3305505291.9939681.3790390.9182.3400829.0041389672347453.5673600.1931986870193.2705140.40682.20.231483170.019108.3146.01804935809.4761.82051147240177645745.7229.6182.46547808.312176.800.3628410.945450348.8380.25655581.2911684.27127.78671.541672.64672.28437.86688.3639.8231230272439.6773.035160.61533135.7250.84633010.42437.972106.6858569303105.6502.10597213870.87366.7290.2712810.3818.872096.7106863443.5470.266946634524178.0913.448124.8138.5225.05017271148157.595320366.36323.86662.0446259.88189606.141806032.650013.3328.503597184019.31112356.93234.9212.58196.975.1663128.13923.16157.7832.483335.98715.5014576.86.4564638.52445998.846915.540.723220770901.3962629.1472662331066666672740.766190927.963062182510.9416538666678329833333316633333222414598212566.88662782.7110.1943.611.7113234319.4878741.3743290.9184.0409548.8841394002261054.0292610.1909217165173.4758839.52283.00.218363178.773109.0744.81732503199.4511.80328143600173771446.1129.5482.63848816.512252.730.3537570.942653348.6040.24917082.6069667.64627.78678.041671.55771.92431.30488.1639.9669225195433.7413.011860.60293636.3000.85040010.42433.426106.1435576988104.3152.10694214228.62367.7990.2708310.3818.872128.7103051374.2030.263954214421171.5913.874525.9608.9675.36117161395049.875518382.98822.08460.9336011.42194386.0242829.7987240.9678.4210360529654.59413217.80592.9929.05456.472.4079060.27346.98315.8018.171222.24110.2377015.55.6220127.24632984.633677.231.104296055651.5807738.5190018438430666672253.681574734.9560092209.413.061973100000999833333397006666719084381015142.8673.83702447.012549.211.1637303289.5514731.3567910090.1438809.2011435672452833.3609450.1889066656663.2298138.69388.00.216048175.476102.2547.51820839268.9521.74481151167182881643.8328.2379.11846761.212670.820.3615210.921930358.3160.24772584.1301663.89628.57660.198660.05271.51434.95486.8339.4821229770433.4322.985720.61446835.6320.86206110.60433.798107.7726572684104.2782.08771212303.58364.9870.2714110.3818.872656.6312015354.2980.254990608503991.5813.865920.7365.8153.06617459684859.983590301.41331.15340.8155425.4044055.391698330.749813.2826.253733249119.00108786.95240.5411.65195.555.2873228.37023.20159.7031.445137.53216.0776020.48.7486736.68438365.841946.142.735222433961.8813629.1772117931011666672781.766780554.75201811.04163290000082783333333028000001979099115712564.66302793.4109.6345.712.5811322323.8790791.5126090.9182.5419838.4281318332391445.6625660.2034966790053.2896041.47682.10.222234167.322108.2745.91835049879.0891.84199148033181354946.0429.5579.61047685.712185.510.3679960.957911362.1290.25452681.5024679.93627.78676.539677.85573.35442.35088.8940.3877228894442.7003.048150.61223435.5860.85464210.42440.632106.0534574742104.7662.11134212714.19366.3680.2727410.312747.12698.63118.53057.518.872069.1108941401.8600.28488685813.086526.5069.3325.18616864452854.835088354.02727.20562.2112213.51OpenBenchmarking.org

NAS Parallel Benchmarks

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

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3440K80K120K160K200KSE +/- 301.93, N = 3SE +/- 1295.33, N = 3SE +/- 419.66, N = 3SE +/- 204.64, N = 3194386.02189606.14187807.0844055.39-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Clear Linux 34630: 3.23. Ubuntu 20.04.2 LTS: Open MPI 4.0.34. Ubuntu 21.04: Open MPI 4.1.0
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3430K60K90K120K150KMin: 193934.47 / Avg: 194386.02 / Max: 194959.01Min: 187043.01 / Avg: 189606.14 / Max: 191213.94Min: 187266.97 / Avg: 187807.08 / Max: 188633.52Min: 43841.72 / Avg: 44055.39 / Max: 44464.531. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Clear Linux 34630: 3.23. Ubuntu 20.04.2 LTS: Open MPI 4.0.34. Ubuntu 21.04: Open MPI 4.1.0

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS4K8K12K16K20KSE +/- 51.71, N = 4SE +/- 82.67, N = 4SE +/- 78.01, N = 4SE +/- 203.38, N = 44282167291698318060
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS3K6K9K12K15KMin: 4146 / Avg: 4281.5 / Max: 4391Min: 16552 / Avg: 16729.25 / Max: 16937Min: 16783 / Avg: 16983.25 / Max: 17138Min: 17625 / Avg: 18059.75 / Max: 18505

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS816243240SE +/- 0.09474, N = 15SE +/- 0.28978, N = 15SE +/- 0.34184, N = 12SE +/- 0.49102, N = 159.7987228.7517030.7498032.65000-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 6.28MIN: 15.46-O2 - MIN: 12.49MIN: 9.31. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS714212835Min: 9.13 / Avg: 9.8 / Max: 10.58Min: 27.21 / Avg: 28.75 / Max: 31.89Min: 29.35 / Avg: 30.75 / Max: 33.86Min: 31.03 / Avg: 32.65 / Max: 39.181. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34918273645SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 340.9613.3813.3313.28-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O2-O21. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Very FastClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34918273645Min: 40.79 / Avg: 40.96 / Max: 41.08Min: 13.36 / Avg: 13.38 / Max: 13.41Min: 13.28 / Avg: 13.33 / Max: 13.35Min: 13.24 / Avg: 13.28 / Max: 13.331. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3420406080100SE +/- 0.59, N = 3SE +/- 0.25, N = 3SE +/- 0.34, N = 3SE +/- 0.15, N = 378.4228.5028.3326.25-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lnuma-O3 -lnuma-O3 -lnuma-O21. (CXX) g++ options: -rdynamic -lpthread -lrt -ldl
OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 341530456075Min: 77.57 / Avg: 78.42 / Max: 79.56Min: 28.04 / Avg: 28.5 / Max: 28.91Min: 27.73 / Avg: 28.33 / Max: 28.9Min: 26 / Avg: 26.25 / Max: 26.521. (CXX) g++ options: -rdynamic -lpthread -lrt -ldl

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS2K4K6K8K10KSE +/- 1.53, N = 3SE +/- 6.89, N = 3SE +/- 13.59, N = 310360381537333597-lSDL2 -lopus -O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-lSDL2 -lavformat -lavcodec -lswscale -lavutil -lasound -O2-O2-lSDL2 -lavformat -lavcodec -lswscale -lavutil -lasound -O21. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread
OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatefulClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS2K4K6K8K10KMin: 10358 / Avg: 10360 / Max: 10363Min: 3802 / Avg: 3815.33 / Max: 3825Min: 3574 / Avg: 3596.67 / Max: 36211. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS11002200330044005500SE +/- 20.58, N = 3SE +/- 22.31, N = 15SE +/- 27.76, N = 5SE +/- 14.31, N = 35296253124911840-lSDL2 -lopus -O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-lSDL2 -lavformat -lavcodec -lswscale -lavutil -lasound -O2-O2-lSDL2 -lavformat -lavcodec -lswscale -lavutil -lasound -O21. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread
OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: INVITEClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS9001800270036004500Min: 5262 / Avg: 5295.67 / Max: 5333Min: 2326 / Avg: 2530.6 / Max: 2621Min: 2439 / Avg: 2490.8 / Max: 2588Min: 1812 / Avg: 1840.33 / Max: 18581. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread

libgav1

Libgav1 is an AV1 decoder developed by Google for AV1 profile 0/1 compliance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 4KClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 341224364860SE +/- 0.06, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.14, N = 354.5919.3919.3119.00-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O3-O3-O21. (CXX) g++ options: -lpthread -lrt
OpenBenchmarking.orgFPS, More Is Betterlibgav1 0.16.3Video Input: Summer Nature 4KClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 341122334455Min: 54.47 / Avg: 54.59 / Max: 54.68Min: 19.11 / Avg: 19.39 / Max: 19.59Min: 19.19 / Avg: 19.31 / Max: 19.42Min: 18.73 / Avg: 19 / Max: 19.161. (CXX) g++ options: -lpthread -lrt

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Clear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS2K4K6K8K10KSE +/- 53.70, N = 20SE +/- 133.93, N = 4SE +/- 85.70, N = 20SE +/- 115.71, N = 204132107401087811235
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Clear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS2K4K6K8K10KMin: 3505 / Avg: 4131.95 / Max: 4447Min: 10470 / Avg: 10739.5 / Max: 11108Min: 10402 / Avg: 10878.15 / Max: 12104Min: 10619 / Avg: 11234.65 / Max: 12182

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: MediumClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.0448121620SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 317.806.956.936.90-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O2-O21. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: MediumClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04510152025Min: 17.76 / Avg: 17.8 / Max: 17.86Min: 6.92 / Avg: 6.95 / Max: 6.97Min: 6.92 / Avg: 6.93 / Max: 6.94Min: 6.88 / Avg: 6.9 / Max: 6.911. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS130260390520650SE +/- 5.64, N = 3SE +/- 2.49, N = 3SE +/- 1.79, N = 15SE +/- 1.73, N = 3592.99260.88240.54234.92-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS100200300400500Min: 586.51 / Avg: 592.99 / Max: 604.23Min: 258.06 / Avg: 260.88 / Max: 265.84Min: 228.05 / Avg: 240.54 / Max: 249.48Min: 231.93 / Avg: 234.92 / Max: 237.911. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34714212835SE +/- 0.40, N = 15SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.08, N = 329.0512.9212.5811.65-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -lnuma-O3 -lnuma-O3 -lnuma-O21. (CXX) g++ options: -rdynamic -lpthread -lrt -ldl
OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34612182430Min: 27.84 / Avg: 29.05 / Max: 32.66Min: 12.75 / Avg: 12.92 / Max: 13.05Min: 12.45 / Avg: 12.58 / Max: 12.74Min: 11.56 / Avg: 11.65 / Max: 11.811. (CXX) g++ options: -rdynamic -lpthread -lrt -ldl

SVT-VP9

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34100200300400500SE +/- 2.91, N = 3SE +/- 2.00, N = 15SE +/- 1.73, N = 3SE +/- 2.71, N = 3456.47204.86196.97195.55-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O21. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3480160240320400Min: 452.65 / Avg: 456.47 / Max: 462.17Min: 194.94 / Avg: 204.86 / Max: 216.44Min: 193.67 / Avg: 196.97 / Max: 199.53Min: 190.19 / Avg: 195.55 / Max: 198.921. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Tungsten Renderer

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Non-ExponentialClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 341.18962.37923.56884.75845.948SE +/- 0.00072, N = 3SE +/- 0.02249, N = 3SE +/- 0.07466, N = 3SE +/- 0.00702, N = 32.407905.166315.239125.28732-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz-O21. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Non-ExponentialClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34246810Min: 2.41 / Avg: 2.41 / Max: 2.41Min: 5.13 / Avg: 5.17 / Max: 5.21Min: 5.15 / Avg: 5.24 / Max: 5.39Min: 5.28 / Avg: 5.29 / Max: 5.31. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1326395265SE +/- 0.42, N = 3SE +/- 0.34, N = 3SE +/- 0.29, N = 3SE +/- 0.34, N = 1560.2729.8128.3728.14-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8.7Encoder Mode: Preset 8 - Input: Bosphorus 4KClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1224364860Min: 59.85 / Avg: 60.27 / Max: 61.1Min: 29.23 / Avg: 29.81 / Max: 30.42Min: 27.85 / Avg: 28.37 / Max: 28.86Min: 25.33 / Avg: 28.14 / Max: 29.971. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie

Kvazaar

This is a test of Kvazaar as a CPU-based H.265 video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.041122334455SE +/- 0.34, N = 12SE +/- 0.31, N = 3SE +/- 0.24, N = 3SE +/- 0.11, N = 346.9823.2023.1622.78-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O2-O21. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.0Video Input: Bosphorus 4K - Video Preset: Ultra FastClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.041020304050Min: 45.44 / Avg: 46.98 / Max: 48.88Min: 22.78 / Avg: 23.2 / Max: 23.8Min: 22.68 / Avg: 23.16 / Max: 23.45Min: 22.55 / Avg: 22.78 / Max: 22.91. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -lpthread -lm -lrt

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS70140210280350SE +/- 2.34, N = 15SE +/- 1.72, N = 3SE +/- 1.49, N = 3SE +/- 1.68, N = 3315.80172.02159.70157.78-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS60120180240300Min: 300.6 / Avg: 315.8 / Max: 333.89Min: 169.06 / Avg: 172.02 / Max: 175.03Min: 157.32 / Avg: 159.7 / Max: 162.43Min: 155 / Avg: 157.78 / Max: 160.811. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

Tungsten Renderer

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Water CausticClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS816243240SE +/- 0.04, N = 3SE +/- 0.27, N = 3SE +/- 0.09, N = 3SE +/- 0.19, N = 318.1731.2731.4532.48-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz-O2-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz1. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Water CausticClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS714212835Min: 18.12 / Avg: 18.17 / Max: 18.25Min: 30.74 / Avg: 31.27 / Max: 31.59Min: 31.31 / Avg: 31.45 / Max: 31.61Min: 32.17 / Avg: 32.48 / Max: 32.831. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl

Timed Apache Compilation

This test times how long it takes to build the Apache HTTPD web server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To CompileClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34918273645SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.06, N = 322.2435.1935.9937.53
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To CompileClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34816243240Min: 22.22 / Avg: 22.24 / Max: 22.26Min: 35.1 / Avg: 35.19 / Max: 35.28Min: 35.73 / Avg: 35.99 / Max: 36.12Min: 37.45 / Avg: 37.53 / Max: 37.65

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6Clear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3448121620SE +/- 0.03, N = 3SE +/- 0.19, N = 4SE +/- 0.14, N = 15SE +/- 0.19, N = 1510.2415.3815.5016.081. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6Clear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3448121620Min: 10.17 / Avg: 10.24 / Max: 10.29Min: 14.97 / Avg: 15.38 / Max: 15.88Min: 14.62 / Avg: 15.5 / Max: 16.92Min: 15.28 / Avg: 16.08 / Max: 17.911. (CXX) g++ options: -O3 -fPIC -lm

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression SpeedClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS15003000450060007500SE +/- 61.86, N = 3SE +/- 53.61, N = 3SE +/- 54.04, N = 15SE +/- 7.13, N = 37166.76148.46020.44620.1-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression SpeedClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS12002400360048006000Min: 7053.2 / Avg: 7166.7 / Max: 7266.1Min: 6062.3 / Avg: 6148.43 / Max: 6246.8Min: 5505.4 / Avg: 6020.39 / Max: 6300.1Min: 4606 / Avg: 4620.13 / Max: 4628.81. (CC) gcc options: -O3 -pthread -lz

Tungsten Renderer

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: HairClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34246810SE +/- 0.02021, N = 3SE +/- 0.08029, N = 15SE +/- 0.07388, N = 3SE +/- 0.09718, N = 45.622016.456466.525388.74867-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz-O21. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: HairClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 343691215Min: 5.59 / Avg: 5.62 / Max: 5.66Min: 5.89 / Avg: 6.46 / Max: 7.05Min: 6.43 / Avg: 6.53 / Max: 6.67Min: 8.51 / Avg: 8.75 / Max: 8.951. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS918273645SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.47, N = 3SE +/- 0.38, N = 327.2536.6836.7638.521. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS816243240Min: 27.03 / Avg: 27.25 / Max: 27.37Min: 36.42 / Avg: 36.68 / Max: 36.97Min: 36.22 / Avg: 36.76 / Max: 37.7Min: 38.05 / Avg: 38.52 / Max: 39.281. (CXX) g++ options: -O3 -fPIC -lm

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS10K20K30K40K50KSE +/- 68.45, N = 3SE +/- 353.50, N = 3SE +/- 612.15, N = 15SE +/- 630.61, N = 1532984.638365.841555.145998.8
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS8K16K24K32K40KMin: 32910 / Avg: 32984.6 / Max: 33121.3Min: 37962.1 / Avg: 38365.8 / Max: 39070.3Min: 38052.1 / Avg: 41555.14 / Max: 45423.8Min: 41194.9 / Avg: 45998.8 / Max: 51599.7

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS10K20K30K40K50KSE +/- 60.78, N = 3SE +/- 623.47, N = 13SE +/- 601.06, N = 15SE +/- 706.39, N = 1533677.241946.143074.546915.5
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS8K16K24K32K40KMin: 33583.3 / Avg: 33677.17 / Max: 33791Min: 37471.9 / Avg: 41946.08 / Max: 44866.6Min: 39098.2 / Avg: 43074.54 / Max: 46567.3Min: 42405.3 / Avg: 46915.49 / Max: 50414.4

Timed PHP Compilation

This test times how long it takes to build PHP 7. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To CompileClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 341020304050SE +/- 0.10, N = 3SE +/- 0.52, N = 3SE +/- 0.02, N = 3SE +/- 0.25, N = 331.1040.2140.7242.74
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To CompileClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34918273645Min: 30.94 / Avg: 31.1 / Max: 31.28Min: 39.61 / Avg: 40.21 / Max: 41.23Min: 40.68 / Avg: 40.72 / Max: 40.76Min: 42.23 / Avg: 42.74 / Max: 43.03

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkVdbVolumeClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.046M12M18M24M30MSE +/- 324102.01, N = 3SE +/- 179975.75, N = 3SE +/- 187763.97, N = 3SE +/- 58341.71, N = 329605565222433962207709021915837MIN: 1030715 / MAX: 177194520MIN: 1023291 / MAX: 182220264MIN: 1030467 / MAX: 169859376MIN: 1062215 / MAX: 153704880
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkVdbVolumeClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.045M10M15M20M25MMin: 29264695 / Avg: 29605565.33 / Max: 30253475Min: 21885261 / Avg: 22243396.33 / Max: 22453739Min: 21708111 / Avg: 22077090 / Max: 22322051Min: 21807522 / Avg: 21915837.33 / Max: 22007573

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 340.42330.84661.26991.69322.1165SE +/- 0.00638, N = 3SE +/- 0.00589, N = 3SE +/- 0.01755, N = 3SE +/- 0.01180, N = 31.396261.410841.580771.88136MIN: 1.26MIN: 1.27-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 1.38-O2 - MIN: 1.721. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 34246810Min: 1.39 / Avg: 1.4 / Max: 1.41Min: 1.4 / Avg: 1.41 / Max: 1.42Min: 1.56 / Avg: 1.58 / Max: 1.62Min: 1.86 / Avg: 1.88 / Max: 1.91. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS918273645SE +/- 0.24, N = 3SE +/- 0.09, N = 3SE +/- 0.31, N = 4SE +/- 0.33, N = 338.5129.5729.1729.14-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS816243240Min: 38.14 / Avg: 38.51 / Max: 38.95Min: 29.41 / Avg: 29.57 / Max: 29.7Min: 28.25 / Avg: 29.17 / Max: 29.67Min: 28.5 / Avg: 29.14 / Max: 29.571. (CC) gcc options: -O3 -fPIE -fPIC -O2 -pie -rdynamic -lpthread -lrt

PHPBench

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

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04200K400K600K800K1000KSE +/- 1642.74, N = 3SE +/- 5472.66, N = 3SE +/- 3881.00, N = 3SE +/- 3328.66, N = 3900184726623721179715587
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04160K320K480K640K800KMin: 897566 / Avg: 900183.67 / Max: 903212Min: 715971 / Avg: 726623.33 / Max: 734128Min: 713612 / Avg: 721179.33 / Max: 726459Min: 709499 / Avg: 715586.67 / Max: 720964

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 160 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04800M1600M2400M3200M4000MSE +/- 13074699.91, N = 3SE +/- 11545032.60, N = 3SE +/- 13741098.13, N = 3SE +/- 15087448.79, N = 33843066667310666666731011666673080633333-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 160 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04700M1400M2100M2800M3500MMin: 3818900000 / Avg: 3843066666.67 / Max: 3863800000Min: 3083600000 / Avg: 3106666666.67 / Max: 3119100000Min: 3073700000 / Avg: 3101166666.67 / Max: 3115700000Min: 3050700000 / Avg: 3080633333.33 / Max: 30989000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedFedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04Clear Linux 346306001200180024003000SE +/- 35.16, N = 3SE +/- 9.85, N = 3SE +/- 12.48, N = 3SE +/- 4.09, N = 32781.72740.72727.62253.6-llzma-llzma-llzma-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedFedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04Clear Linux 346305001000150020002500Min: 2711.5 / Avg: 2781.7 / Max: 2820.4Min: 2721 / Avg: 2740.7 / Max: 2750.6Min: 2703.7 / Avg: 2727.6 / Max: 2745.8Min: 2246.9 / Avg: 2253.6 / Max: 22611. (CC) gcc options: -O3 -pthread -lz

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS2004006008001000SE +/- 2.08, N = 3SE +/- 2.40, N = 3SE +/- 2.67, N = 3SE +/- 5.81, N = 3815669667661MIN: 1 / MAX: 3198MIN: 1 / MAX: 2821MIN: 1 / MAX: 2865MIN: 1 / MAX: 2845
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS140280420560700Min: 811 / Avg: 815 / Max: 818Min: 666 / Avg: 669.33 / Max: 674Min: 662 / Avg: 667.33 / Max: 670Min: 650 / Avg: 660.67 / Max: 670

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS20K40K60K80K100KSE +/- 102.89, N = 3SE +/- 755.90, N = 15SE +/- 768.19, N = 15SE +/- 1128.79, N = 474734.980554.781319.390927.9
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS16K32K48K64K80KMin: 74602.8 / Avg: 74734.9 / Max: 74937.6Min: 75590.6 / Avg: 80554.73 / Max: 85844.7Min: 77704.6 / Avg: 81319.34 / Max: 87869.3Min: 89451.7 / Avg: 90927.88 / Max: 94284.6

BlogBench

BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: WriteUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 3414K28K42K56K70KSE +/- 770.50, N = 3SE +/- 717.36, N = 3SE +/- 74.61, N = 3SE +/- 835.62, N = 363062607685600952018-O2-O2-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O21. (CC) gcc options: -pthread
OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: WriteUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 3411K22K33K44K55KMin: 61633 / Avg: 63062 / Max: 64276Min: 59667 / Avg: 60767.67 / Max: 62115Min: 55869 / Avg: 56008.67 / Max: 56124Min: 50384 / Avg: 52018.33 / Max: 531381. (CC) gcc options: -pthread

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 3463050010001500200025001825.01875.82209.4-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lz1. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lcomex -lm -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

Node.js V8 Web Tooling Benchmark

Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.043691215SE +/- 0.01, N = 3SE +/- 0.08, N = 3SE +/- 0.12, N = 15SE +/- 0.07, N = 313.0611.0410.9410.791. Clear Linux 34630: Nodejs v14.17.02. Fedora Server 34: Nodejs v14.16.13. Ubuntu 20.04.2 LTS: Nodejs v10.19.04. Ubuntu 21.04: Nodejs v12.21.0
OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.0448121620Min: 13.03 / Avg: 13.06 / Max: 13.08Min: 10.89 / Avg: 11.04 / Max: 11.17Min: 9.79 / Avg: 10.94 / Max: 11.49Min: 10.66 / Avg: 10.79 / Max: 10.891. Clear Linux 34630: Nodejs v14.17.02. Fedora Server 34: Nodejs v14.16.13. Ubuntu 20.04.2 LTS: Nodejs v10.19.04. Ubuntu 21.04: Nodejs v12.21.0

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04400M800M1200M1600M2000MSE +/- 16877598.57, N = 3SE +/- 5634516.64, N = 3SE +/- 2608320.02, N = 3SE +/- 768837.51, N = 31973100000165386666716329000001631166667-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 32 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04300M600M900M1200M1500MMin: 1953500000 / Avg: 1973100000 / Max: 2006700000Min: 1642600000 / Avg: 1653866666.67 / Max: 1659700000Min: 1627800000 / Avg: 1632900000 / Max: 1636400000Min: 1630300000 / Avg: 1631166666.67 / Max: 16327000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04200M400M600M800M1000MSE +/- 2729061.70, N = 3SE +/- 5004752.19, N = 3SE +/- 1422279.08, N = 3SE +/- 702242.44, N = 3999833333832983333827833333826756667-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04200M400M600M800M1000MMin: 994400000 / Avg: 999833333.33 / Max: 1003000000Min: 823060000 / Avg: 832983333.33 / Max: 839080000Min: 825910000 / Avg: 827833333.33 / Max: 830610000Min: 825730000 / Avg: 826756666.67 / Max: 8281000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04900M1800M2700M3600M4500MSE +/- 36096414.10, N = 3SE +/- 8772368.23, N = 3SE +/- 6501025.56, N = 3SE +/- 9583724.63, N = 33970066667331663333333028000003290766667-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 128 - Buffer Length: 256 - Filter Length: 57Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04700M1400M2100M2800M3500MMin: 3899000000 / Avg: 3970066666.67 / Max: 4016600000Min: 3299400000 / Avg: 3316633333.33 / Max: 3328100000Min: 3289800000 / Avg: 3302800000 / Max: 3309500000Min: 3271600000 / Avg: 3290766666.67 / Max: 33005000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

BlogBench

BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadUbuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34Clear Linux 34630500K1000K1500K2000K2500KSE +/- 30037.66, N = 3SE +/- 21799.22, N = 9SE +/- 22175.10, N = 9SE +/- 12916.51, N = 32253810222414519790991908438-O2-O2-O2-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -pthread
OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadUbuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34Clear Linux 34630400K800K1200K1600K2000KMin: 2197358 / Avg: 2253810 / Max: 2299831Min: 2149752 / Avg: 2224145.44 / Max: 2319913Min: 1876615 / Avg: 1979098.89 / Max: 2049154Min: 1886633 / Avg: 1908438.33 / Max: 19313371. (CC) gcc options: -pthread

PyBench

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 342004006008001000SE +/- 0.88, N = 3SE +/- 1.86, N = 3SE +/- 1.00, N = 3SE +/- 0.58, N = 398299510151157
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 342004006008001000Min: 980 / Avg: 981.67 / Max: 983Min: 993 / Avg: 995.33 / Max: 999Min: 1013 / Avg: 1015 / Max: 1016Min: 1156 / Avg: 1157 / Max: 1158

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: SciVisClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04306090120150SE +/- 0.00, N = 3142.86125.00125.00125.00MIN: 41.67MIN: 32.26 / MAX: 142.86MIN: 33.33 / MAX: 142.86MIN: 31.25 / MAX: 142.86
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04306090120150Min: 142.86 / Avg: 142.86 / Max: 142.86

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 341632486480SE +/- 0.68, N = 3SE +/- 0.30, N = 3SE +/- 0.70, N = 3SE +/- 0.40, N = 373.8467.0466.8964.66MIN: 64.62 / MAX: 94.03MIN: 59.72 / MAX: 87.99MIN: 59.4 / MAX: 88.24MIN: 58.41 / MAX: 86.67
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: CrownClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 341428425670Min: 73.05 / Avg: 73.84 / Max: 75.19Min: 66.54 / Avg: 67.04 / Max: 67.57Min: 65.49 / Avg: 66.89 / Max: 67.64Min: 64 / Avg: 64.66 / Max: 65.39

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedFedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04Clear Linux 346306001200180024003000SE +/- 3.95, N = 15SE +/- 5.43, N = 15SE +/- 1.06, N = 3SE +/- 7.42, N = 32793.42782.72719.42447.0-llzma-llzma-llzma-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedFedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04Clear Linux 346305001000150020002500Min: 2757 / Avg: 2793.4 / Max: 2806.6Min: 2739.3 / Avg: 2782.72 / Max: 2805.3Min: 2718 / Avg: 2719.43 / Max: 2721.5Min: 2432.2 / Avg: 2446.97 / Max: 2455.61. (CC) gcc options: -O3 -pthread -lz

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04306090120150SE +/- 0.00, N = 3SE +/- 0.93, N = 12SE +/- 1.01, N = 15SE +/- 1.01, N = 15125.00110.19109.63109.63MIN: 90.91MIN: 21.74 / MAX: 111.11MIN: 20 / MAX: 111.11MIN: 20.83 / MAX: 125
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0420406080100Min: 125 / Avg: 125 / Max: 125Min: 100 / Avg: 110.19 / Max: 111.11Min: 100 / Avg: 109.63 / Max: 111.11Min: 100 / Avg: 109.63 / Max: 111.11

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression SpeedClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1122334455SE +/- 0.18, N = 3SE +/- 0.33, N = 3SE +/- 0.35, N = 15SE +/- 0.39, N = 1549.247.245.743.6-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression SpeedClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1020304050Min: 48.9 / Avg: 49.23 / Max: 49.5Min: 46.9 / Avg: 47.23 / Max: 47.9Min: 43 / Avg: 45.65 / Max: 48.5Min: 41.9 / Avg: 43.6 / Max: 48.41. (CC) gcc options: -O3 -pthread -lz

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 343691215SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 311.1611.3311.7112.58-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3448121620Min: 11.15 / Avg: 11.16 / Max: 11.17Min: 11.27 / Avg: 11.33 / Max: 11.4Min: 11.67 / Avg: 11.71 / Max: 11.78Min: 12.47 / Avg: 12.58 / Max: 12.661. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3470140210280350SE +/- 0.46, N = 3SE +/- 1.18, N = 3SE +/- 3.10, N = 3SE +/- 0.15, N = 3289.55291.99319.49323.88-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: X3D-benchmarking input.i3dClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3460120180240300Min: 288.69 / Avg: 289.55 / Max: 290.28Min: 290.54 / Avg: 291.99 / Max: 294.33Min: 315.61 / Avg: 319.49 / Max: 325.63Min: 323.72 / Avg: 323.88 / Max: 324.171. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 340.34030.68061.02091.36121.7015SE +/- 0.00192, N = 3SE +/- 0.01216, N = 3SE +/- 0.01131, N = 3SE +/- 0.00594, N = 31.356791.374321.379031.51260-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 1.31MIN: 1.32MIN: 1.33-O2 - MIN: 1.461. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34246810Min: 1.35 / Avg: 1.36 / Max: 1.36Min: 1.36 / Avg: 1.37 / Max: 1.4Min: 1.36 / Avg: 1.38 / Max: 1.4Min: 1.5 / Avg: 1.51 / Max: 1.521. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -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: SciVisClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.0420406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3100.0090.9190.9190.91MIN: 52.63 / MAX: 111.11MIN: 55.56 / MAX: 100MIN: 55.56 / MAX: 100MIN: 55.56 / MAX: 100
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.0420406080100Min: 100 / Avg: 100 / Max: 100Min: 90.91 / Avg: 90.91 / Max: 90.91Min: 90.91 / Avg: 90.91 / Max: 90.91Min: 90.91 / Avg: 90.91 / Max: 90.91

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression SpeedClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0420406080100SE +/- 0.32, N = 3SE +/- 0.68, N = 3SE +/- 0.72, N = 3SE +/- 0.81, N = 390.184.082.582.3-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression SpeedClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0420406080100Min: 89.7 / Avg: 90.07 / Max: 90.7Min: 83.3 / Avg: 84.03 / Max: 85.4Min: 81.5 / Avg: 82.5 / Max: 83.9Min: 81.2 / Avg: 82.33 / Max: 83.91. (CC) gcc options: -O3 -pthread -lz

PJSIP

PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.049K18K27K36K45KSE +/- 270.37, N = 3SE +/- 181.15, N = 3SE +/- 352.27, N = 3SE +/- 533.56, N = 343880419834095440082-lSDL2 -lopus -O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-lSDL2 -lavformat -lavcodec -lswscale -lavutil -lasound -O2-lSDL2 -lavformat -lavcodec -lswscale -lavutil -lasound -O21. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread
OpenBenchmarking.orgResponses Per Second, More Is BetterPJSIP 2.11Method: OPTIONS, StatelessClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.048K16K24K32K40KMin: 43369 / Avg: 43879.67 / Max: 44289Min: 41751 / Avg: 41983 / Max: 42340Min: 40572 / Avg: 40954.33 / Max: 41658Min: 39055 / Avg: 40082.33 / Max: 408461. (CC) gcc options: -lstdc++ -lssl -lcrypto -luuid -lm -lrt -lpthread

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bareClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 343691215SE +/- 0.028, N = 3SE +/- 0.058, N = 3SE +/- 0.033, N = 3SE +/- 0.014, N = 39.2019.0048.8848.428-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O3-O3-O21. (CXX) g++ options: -pthread
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021.2Implementation: MPI CPU - Input: water_GMX50_bareClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 343691215Min: 9.15 / Avg: 9.2 / Max: 9.24Min: 8.9 / Avg: 9 / Max: 9.1Min: 8.82 / Avg: 8.88 / Max: 8.94Min: 8.4 / Avg: 8.43 / Max: 8.451. (CXX) g++ options: -pthread

Chia Blockchain VDF

Chia is a blockchain and smart transaction platform based on proofs of space and time rather than proofs of work with other cryptocurrencies. This test profile is benchmarking the CPU performance for Chia VDF performance using the Chia VDF benchmark. The Chia VDF is for the Chia Verifiable Delay Function (Proof of Time). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Plain C++Clear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3430K60K90K120K150KSE +/- 33.33, N = 3SE +/- 585.95, N = 3SE +/- 88.19, N = 3SE +/- 589.73, N = 31435671394001389671318331. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Plain C++Clear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3420K40K60K80K100KMin: 143500 / Avg: 143566.67 / Max: 143600Min: 138300 / Avg: 139400 / Max: 140300Min: 138800 / Avg: 138966.67 / Max: 139100Min: 131100 / Avg: 131833.33 / Max: 1330001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS500K1000K1500K2000K2500KSE +/- 6737.83, N = 3SE +/- 5456.02, N = 3SE +/- 2872.35, N = 3SE +/- 24473.66, N = 52452833.362391445.662347453.572261054.03-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -O2 -lrt" -lrt
OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS400K800K1200K1600K2000KMin: 2445361.45 / Avg: 2452833.36 / Max: 2466281.31Min: 2382370.46 / Avg: 2391445.66 / Max: 2401230.63Min: 2342777.66 / Avg: 2347453.57 / Max: 2352681.69Min: 2168388.95 / Avg: 2261054.03 / Max: 2297777.621. (CC) gcc options: -O2 -lrt" -lrt

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 340.04580.09160.13740.18320.229SE +/- 0.000520, N = 3SE +/- 0.001679, N = 14SE +/- 0.001836, N = 13SE +/- 0.001612, N = 150.1889060.1909210.1931980.203496-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.18MIN: 0.18MIN: 0.18-O2 - MIN: 0.181. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3412345Min: 0.19 / Avg: 0.19 / Max: 0.19Min: 0.19 / Avg: 0.19 / Max: 0.21Min: 0.19 / Avg: 0.19 / Max: 0.21Min: 0.19 / Avg: 0.2 / Max: 0.211. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Clear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS150K300K450K600K750KSE +/- 6494.21, N = 3SE +/- 3314.01, N = 3SE +/- 7233.85, N = 5SE +/- 2002.95, N = 3665666679005687019716517
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Clear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS120K240K360K480K600KMin: 655870 / Avg: 665666 / Max: 677950Min: 675501 / Avg: 679004.67 / Max: 685629Min: 665700 / Avg: 687018.6 / Max: 702860Min: 714314 / Avg: 716517 / Max: 720516

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS0.78211.56422.34633.12843.9105SE +/- 0.00331, N = 3SE +/- 0.00929, N = 3SE +/- 0.01091, N = 3SE +/- 0.00414, N = 33.229813.270513.289603.47588-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 3.02MIN: 3.09-O2 - MIN: 3.12MIN: 3.311. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS246810Min: 3.23 / Avg: 3.23 / Max: 3.24Min: 3.26 / Avg: 3.27 / Max: 3.29Min: 3.28 / Avg: 3.29 / Max: 3.31Min: 3.47 / Avg: 3.48 / Max: 3.481. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34918273645SE +/- 0.46, N = 4SE +/- 0.07, N = 3SE +/- 0.15, N = 3SE +/- 0.15, N = 338.6939.5240.4141.481. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34918273645Min: 38.19 / Avg: 38.69 / Max: 40.08Min: 39.42 / Avg: 39.52 / Max: 39.65Min: 40.17 / Avg: 40.41 / Max: 40.68Min: 41.32 / Avg: 41.48 / Max: 41.781. (CXX) g++ options: -O2 -lOpenCL

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3420406080100SE +/- 0.66, N = 3SE +/- 0.70, N = 8SE +/- 1.03, N = 15SE +/- 0.74, N = 788.083.082.282.1-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression SpeedClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3420406080100Min: 86.7 / Avg: 88 / Max: 88.8Min: 79 / Avg: 82.98 / Max: 84.6Min: 73 / Avg: 82.25 / Max: 89.8Min: 80 / Avg: 82.11 / Max: 85.61. (CC) gcc options: -O3 -pthread -lz

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.040.05210.10420.15630.20840.2605SE +/- 0.001684, N = 3SE +/- 0.001876, N = 8SE +/- 0.002070, N = 7SE +/- 0.002322, N = 60.2160480.2183630.2222340.231483-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.19MIN: 0.19-O2 - MIN: 0.19MIN: 0.21. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0412345Min: 0.21 / Avg: 0.22 / Max: 0.22Min: 0.21 / Avg: 0.22 / Max: 0.23Min: 0.22 / Avg: 0.22 / Max: 0.23Min: 0.23 / Avg: 0.23 / Max: 0.241. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Timed MrBayes Analysis

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisFedora Server 34Ubuntu 21.04Clear Linux 34630Ubuntu 20.04.2 LTS4080120160200SE +/- 0.50, N = 3SE +/- 0.83, N = 3SE +/- 1.06, N = 3SE +/- 0.79, N = 3167.32170.02175.48178.77-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msha -maes -mavx -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisFedora Server 34Ubuntu 21.04Clear Linux 34630Ubuntu 20.04.2 LTS306090120150Min: 166.69 / Avg: 167.32 / Max: 168.3Min: 168.36 / Avg: 170.02 / Max: 171.01Min: 173.89 / Avg: 175.48 / Max: 177.49Min: 177.23 / Avg: 178.77 / Max: 179.841. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msha -maes -mavx -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CPU-OnlyClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS20406080100SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.26, N = 3SE +/- 0.27, N = 3102.25108.27108.31109.07
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Barbershop - Compute: CPU-OnlyClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS20406080100Min: 102.01 / Avg: 102.25 / Max: 102.38Min: 107.89 / Avg: 108.27 / Max: 108.62Min: 107.89 / Avg: 108.31 / Max: 108.77Min: 108.53 / Avg: 109.07 / Max: 109.38

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression SpeedClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1122334455SE +/- 0.64, N = 3SE +/- 0.54, N = 15SE +/- 0.55, N = 3SE +/- 0.61, N = 1547.546.045.944.8-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression SpeedClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1020304050Min: 46.2 / Avg: 47.47 / Max: 48.3Min: 42.8 / Avg: 46 / Max: 48.7Min: 45 / Avg: 45.9 / Max: 46.9Min: 41.7 / Avg: 44.78 / Max: 47.91. (CC) gcc options: -O3 -pthread -lz

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 512 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total TimeFedora Server 34Clear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTS40M80M120M160M200MSE +/- 1764312.05, N = 15SE +/- 1887131.91, N = 3SE +/- 2129761.70, N = 15SE +/- 1908422.73, N = 4183504987182083926180493580173250319-pipe -fexceptions -fstack-protector -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver
OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total TimeFedora Server 34Clear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTS30M60M90M120M150MMin: 173083841 / Avg: 183504986.8 / Max: 200284302Min: 178309680 / Avg: 182083926 / Max: 183981095Min: 171815407 / Avg: 180493580.2 / Max: 203693488Min: 168713134 / Avg: 173250319 / Max: 1780491631. (CXX) g++ options: -lgcov -m64 -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -msse4.1 -mssse3 -msse2 -mbmi2 -flto -flto=jobserver

POV-Ray

This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.043691215SE +/- 0.016, N = 3SE +/- 0.090, N = 3SE +/- 0.084, N = 3SE +/- 0.097, N = 58.9529.0899.4519.476-fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake -R/usr/lib -lSDL -lpthread -lSM -lICE-R/usr/lib-lSM -lICE -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lpthread-lSM -lICE -lIlmImf -lIlmImf-2_5 -lImath-2_5 -lHalf-2_5 -lIex-2_5 -lIexMath-2_5 -lIlmThread-2_5 -lIlmThread1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace TimeClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.043691215Min: 8.92 / Avg: 8.95 / Max: 8.98Min: 8.93 / Avg: 9.09 / Max: 9.24Min: 9.29 / Avg: 9.45 / Max: 9.58Min: 9.26 / Avg: 9.48 / Max: 9.821. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 340.41440.82881.24321.65762.072SE +/- 0.00330, N = 3SE +/- 0.00660, N = 3SE +/- 0.01169, N = 14SE +/- 0.01185, N = 31.744811.803281.820511.84199-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 1.61MIN: 1.67MIN: 1.67-O2 - MIN: 1.691. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34246810Min: 1.74 / Avg: 1.74 / Max: 1.75Min: 1.8 / Avg: 1.8 / Max: 1.82Min: 1.8 / Avg: 1.82 / Max: 1.97Min: 1.83 / Avg: 1.84 / Max: 1.871. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Chia Blockchain VDF

Chia is a blockchain and smart transaction platform based on proofs of space and time rather than proofs of work with other cryptocurrencies. This test profile is benchmarking the CPU performance for Chia VDF performance using the Chia VDF benchmark. The Chia VDF is for the Chia Verifiable Delay Function (Proof of Time). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Assembly OptimizedClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS30K60K90K120K150KSE +/- 240.37, N = 3SE +/- 1369.10, N = 3SE +/- 1568.63, N = 5SE +/- 680.69, N = 31511671480331472401436001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread
OpenBenchmarking.orgIPS, More Is BetterChia Blockchain VDF 1.0.1Test: Square Assembly OptimizedClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS30K60K90K120K150KMin: 150700 / Avg: 151166.67 / Max: 151500Min: 145500 / Avg: 148033.33 / Max: 150200Min: 141700 / Avg: 147240 / Max: 149800Min: 142300 / Avg: 143600 / Max: 1446001. (CXX) g++ options: -flto -no-pie -lgmpxx -lgmp -lboost_system -pthread

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkUnstructuredVolumeClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS400K800K1200K1600K2000KSE +/- 13533.55, N = 3SE +/- 7558.12, N = 3SE +/- 7183.20, N = 3SE +/- 17974.50, N = 31828816181354917764571737714MIN: 25882 / MAX: 5901670MIN: 19659 / MAX: 5912498MIN: 24424 / MAX: 5906521MIN: 21419 / MAX: 5849413
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkUnstructuredVolumeClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS300K600K900K1200K1500KMin: 1801893 / Avg: 1828816 / Max: 1844693Min: 1799940 / Avg: 1813548.67 / Max: 1826052Min: 1762100 / Avg: 1776457.33 / Max: 1784078Min: 1719734 / Avg: 1737714 / Max: 1773663

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CPU-OnlyClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1020304050SE +/- 0.01, N = 3SE +/- 0.15, N = 3SE +/- 0.08, N = 3SE +/- 0.17, N = 343.8345.7246.0446.11
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Fishy Cat - Compute: CPU-OnlyClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS918273645Min: 43.81 / Avg: 43.83 / Max: 43.85Min: 45.53 / Avg: 45.72 / Max: 46.02Min: 45.95 / Avg: 46.04 / Max: 46.2Min: 45.77 / Avg: 46.11 / Max: 46.3

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-OnlyClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04714212835SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.04, N = 328.2329.5429.5529.61
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: BMW27 - Compute: CPU-OnlyClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04714212835Min: 28.12 / Avg: 28.23 / Max: 28.42Min: 29.51 / Avg: 29.54 / Max: 29.61Min: 29.26 / Avg: 29.55 / Max: 29.74Min: 29.52 / Avg: 29.61 / Max: 29.66

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS20406080100SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.73, N = 15SE +/- 0.11, N = 379.1279.6182.4782.641. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS1632486480Min: 78.98 / Avg: 79.12 / Max: 79.29Min: 79.37 / Avg: 79.61 / Max: 79.89Min: 79.23 / Avg: 82.46 / Max: 86.7Min: 82.45 / Avg: 82.64 / Max: 82.821. (CC) gcc options: -O2 -pthread

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS10K20K30K40K50KSE +/- 60.24, N = 3SE +/- 31.53, N = 3SE +/- 35.00, N = 3SE +/- 510.23, N = 1546761.247685.747808.348816.5
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS8K16K24K32K40KMin: 46654.5 / Avg: 46761.23 / Max: 46863Min: 47637.1 / Avg: 47685.7 / Max: 47744.8Min: 47739.8 / Avg: 47808.27 / Max: 47855.1Min: 47058.1 / Avg: 48816.53 / Max: 53639.6

Sysbench

This is a benchmark of Sysbench with the built-in CPU and memory sub-tests. Sysbench is a scriptable multi-threaded benchmark tool based on LuaJIT. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / MemoryClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.043K6K9K12K15KSE +/- 126.52, N = 3SE +/- 138.52, N = 15SE +/- 101.00, N = 15SE +/- 145.91, N = 1512670.8212252.7312185.5112176.80-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgMiB/sec, More Is BetterSysbench 1.0.20Test: RAM / MemoryClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.042K4K6K8K10KMin: 12420.62 / Avg: 12670.82 / Max: 12828.65Min: 10845.88 / Avg: 12252.73 / Max: 13282.15Min: 11422.45 / Avg: 12185.51 / Max: 12791.22Min: 11022.83 / Avg: 12176.8 / Max: 12925.511. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUUbuntu 20.04.2 LTSClear Linux 34630Ubuntu 21.04Fedora Server 340.08280.16560.24840.33120.414SE +/- 0.003468, N = 3SE +/- 0.001144, N = 3SE +/- 0.004351, N = 3SE +/- 0.004582, N = 30.3537570.3615210.3628410.367996MIN: 0.31-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.32MIN: 0.32-O2 - MIN: 0.321. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUUbuntu 20.04.2 LTSClear Linux 34630Ubuntu 21.04Fedora Server 3412345Min: 0.35 / Avg: 0.35 / Max: 0.36Min: 0.36 / Avg: 0.36 / Max: 0.36Min: 0.36 / Avg: 0.36 / Max: 0.37Min: 0.36 / Avg: 0.37 / Max: 0.381. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 340.21550.4310.64650.8621.0775SE +/- 0.004580, N = 3SE +/- 0.011491, N = 3SE +/- 0.010679, N = 4SE +/- 0.011811, N = 30.9219300.9426530.9454500.957911-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.84MIN: 0.85MIN: 0.86-O2 - MIN: 0.871. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34246810Min: 0.92 / Avg: 0.92 / Max: 0.93Min: 0.93 / Avg: 0.94 / Max: 0.96Min: 0.92 / Avg: 0.95 / Max: 0.97Min: 0.94 / Avg: 0.96 / Max: 0.981. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

RELION

RELION - REgularised LIkelihood OptimisatioN - is a stand-alone computer program for Maximum A Posteriori refinement of (multiple) 3D reconstructions or 2D class averages in cryo-electron microscopy (cryo-EM). It is developed in the research group of Sjors Scheres at the MRC Laboratory of Molecular Biology. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 3.1.1Test: Basic - Device: CPUUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 3480160240320400SE +/- 0.74, N = 3SE +/- 1.61, N = 3SE +/- 0.17, N = 3SE +/- 0.25, N = 3348.60348.84358.32362.13-O3 -lmpi_cxx-O3 -lmpi_cxx-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2 -lmpi_cxx1. (CXX) g++ options: -fopenmp -std=c++0x -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi
OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 3.1.1Test: Basic - Device: CPUUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 3460120180240300Min: 347.85 / Avg: 348.6 / Max: 350.08Min: 346.86 / Avg: 348.84 / Max: 352.03Min: 358 / Avg: 358.32 / Max: 358.56Min: 361.83 / Avg: 362.13 / Max: 362.631. (CXX) g++ options: -fopenmp -std=c++0x -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -pthread -lmpi

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.040.05770.11540.17310.23080.2885SE +/- 0.001913, N = 3SE +/- 0.003241, N = 3SE +/- 0.003140, N = 4SE +/- 0.002948, N = 30.2477250.2491700.2545260.256555-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.22MIN: 0.22-O2 - MIN: 0.23MIN: 0.231. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0412345Min: 0.24 / Avg: 0.25 / Max: 0.25Min: 0.24 / Avg: 0.25 / Max: 0.26Min: 0.25 / Avg: 0.25 / Max: 0.26Min: 0.25 / Avg: 0.26 / Max: 0.261. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian DragonClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0420406080100SE +/- 1.14, N = 3SE +/- 0.68, N = 9SE +/- 0.76, N = 3SE +/- 0.58, N = 384.1382.6181.5081.29MIN: 68.3 / MAX: 93.51MIN: 67 / MAX: 93.51MIN: 65.61 / MAX: 92.76MIN: 67.06 / MAX: 91.79
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer - Model: Asian DragonClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.041632486480Min: 81.87 / Avg: 84.13 / Max: 85.49Min: 79.42 / Avg: 82.61 / Max: 84.87Min: 80.5 / Avg: 81.5 / Max: 82.99Min: 80.18 / Avg: 81.29 / Max: 82.11

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04150300450600750SE +/- 3.59, N = 3SE +/- 1.50, N = 3SE +/- 2.47, N = 3SE +/- 5.64, N = 15663.90667.65679.94684.27-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 627.72MIN: 642.94-O2 - MIN: 650MIN: 645.41. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04120240360480600Min: 659.93 / Avg: 663.9 / Max: 671.05Min: 665.94 / Avg: 667.65 / Max: 670.63Min: 677.12 / Avg: 679.94 / Max: 684.85Min: 669.34 / Avg: 684.27 / Max: 729.621. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -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 TracerClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04714212835SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 328.5727.7827.7827.78MIN: 18.87 / MAX: 30.3MIN: 16.39 / MAX: 29.41MIN: 16.67 / MAX: 29.41MIN: 17.24 / MAX: 29.41
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04612182430Min: 28.57 / Avg: 28.57 / Max: 28.57Min: 27.78 / Avg: 27.78 / Max: 27.78Min: 27.78 / Avg: 27.78 / Max: 27.78Min: 27.78 / Avg: 27.78 / Max: 27.78

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS150300450600750SE +/- 0.41, N = 3SE +/- 0.16, N = 3SE +/- 0.57, N = 3SE +/- 4.59, N = 15660.20671.54676.54678.04-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 628.82MIN: 647.67-O2 - MIN: 652.01MIN: 638.981. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS120240360480600Min: 659.39 / Avg: 660.2 / Max: 660.7Min: 671.26 / Avg: 671.54 / Max: 671.82Min: 675.52 / Avg: 676.54 / Max: 677.49Min: 661.97 / Avg: 678.04 / Max: 721.851. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34150300450600750SE +/- 2.16, N = 3SE +/- 5.68, N = 8SE +/- 1.73, N = 3SE +/- 0.75, N = 3660.05671.56672.65677.86-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 626.35MIN: 639.68MIN: 646.15-O2 - MIN: 653.31. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34120240360480600Min: 656.74 / Avg: 660.05 / Max: 664.12Min: 663.28 / Avg: 671.56 / Max: 711.14Min: 669.81 / Avg: 672.65 / Max: 675.79Min: 676.53 / Avg: 677.86 / Max: 679.141. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 341632486480SE +/- 0.10, N = 3SE +/- 0.10, N = 3SE +/- 0.03, N = 3SE +/- 0.69, N = 371.5171.9272.2873.35
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Classroom - Compute: CPU-OnlyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 341428425670Min: 71.32 / Avg: 71.51 / Max: 71.63Min: 71.77 / Avg: 71.92 / Max: 72.12Min: 72.22 / Avg: 72.28 / Max: 72.34Min: 72.5 / Avg: 73.35 / Max: 74.72

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUUbuntu 20.04.2 LTSClear Linux 34630Ubuntu 21.04Fedora Server 34100200300400500SE +/- 1.48, N = 3SE +/- 1.15, N = 3SE +/- 0.63, N = 3SE +/- 1.32, N = 3431.30434.95437.87442.35MIN: 414.55-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 417.43MIN: 422.67-O2 - MIN: 425.261. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUUbuntu 20.04.2 LTSClear Linux 34630Ubuntu 21.04Fedora Server 3480160240320400Min: 429.46 / Avg: 431.3 / Max: 434.23Min: 432.88 / Avg: 434.95 / Max: 436.87Min: 436.65 / Avg: 437.87 / Max: 438.74Min: 440.49 / Avg: 442.35 / Max: 444.91. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: CPU-OnlyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3420406080100SE +/- 0.08, N = 3SE +/- 0.21, N = 3SE +/- 0.24, N = 3SE +/- 0.12, N = 386.8388.1688.3688.89
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.92Blend File: Pabellon Barcelona - Compute: CPU-OnlyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3420406080100Min: 86.68 / Avg: 86.83 / Max: 86.95Min: 87.74 / Avg: 88.16 / Max: 88.39Min: 87.89 / Avg: 88.36 / Max: 88.65Min: 88.65 / Avg: 88.89 / Max: 89.05

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 3.1Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630918273645SE +/- 0.08, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 340.3939.9739.8239.48-lmpi_cxx-lmpi_cxx-lmpi_cxx1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630816243240Min: 40.24 / Avg: 40.39 / Max: 40.48Min: 39.61 / Avg: 39.97 / Max: 40.25Min: 39.77 / Avg: 39.82 / Max: 39.87Min: 39.47 / Avg: 39.48 / Max: 39.51. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi

SecureMark

SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSUbuntu 21.04Clear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTS50K100K150K200K250KSE +/- 178.37, N = 3SE +/- 121.72, N = 3SE +/- 359.29, N = 3SE +/- 124.15, N = 32302722297702288942251951. (CC) gcc options: -pedantic -O3
OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSUbuntu 21.04Clear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTS40K80K120K160K200KMin: 229915.7 / Avg: 230272.23 / Max: 230461.28Min: 229616.69 / Avg: 229769.97 / Max: 230010.41Min: 228234.53 / Avg: 228894.3 / Max: 229470.75Min: 224948.22 / Avg: 225194.61 / Max: 225344.381. (CC) gcc options: -pedantic -O3

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34100200300400500SE +/- 2.18, N = 3SE +/- 1.79, N = 3SE +/- 0.53, N = 3SE +/- 1.22, N = 3433.43433.74439.68442.70-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 414.35MIN: 416.67MIN: 424.75-O2 - MIN: 423.411. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3480160240320400Min: 430.52 / Avg: 433.43 / Max: 437.69Min: 431.61 / Avg: 433.74 / Max: 437.3Min: 438.65 / Avg: 439.68 / Max: 440.44Min: 441.29 / Avg: 442.7 / Max: 445.131. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 340.68581.37162.05742.74323.429SE +/- 0.00391, N = 3SE +/- 0.03357, N = 3SE +/- 0.02675, N = 12SE +/- 0.02287, N = 142.985723.011863.035163.04815-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 2.85MIN: 2.84MIN: 2.86-O2 - MIN: 2.871. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34246810Min: 2.98 / Avg: 2.99 / Max: 2.99Min: 2.98 / Avg: 3.01 / Max: 3.08Min: 2.99 / Avg: 3.04 / Max: 3.33Min: 3.01 / Avg: 3.05 / Max: 3.341. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPUUbuntu 20.04.2 LTSFedora Server 34Clear Linux 34630Ubuntu 21.040.13840.27680.41520.55360.692SE +/- 0.002882, N = 3SE +/- 0.006364, N = 3SE +/- 0.001093, N = 3SE +/- 0.004554, N = 30.6029360.6122340.6144680.615331MIN: 0.56-O2 - MIN: 0.56-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.56MIN: 0.571. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPUUbuntu 20.04.2 LTSFedora Server 34Clear Linux 34630Ubuntu 21.04246810Min: 0.6 / Avg: 0.6 / Max: 0.61Min: 0.6 / Avg: 0.61 / Max: 0.62Min: 0.61 / Avg: 0.61 / Max: 0.62Min: 0.61 / Avg: 0.62 / Max: 0.621. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 34816243240SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 336.3035.7335.6335.59-O3-O3-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O21. (CXX) g++ options: -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 34816243240Min: 36.21 / Avg: 36.3 / Max: 36.35Min: 35.67 / Avg: 35.73 / Max: 35.77Min: 35.58 / Avg: 35.63 / Max: 35.74Min: 35.52 / Avg: 35.59 / Max: 35.671. (CXX) g++ options: -pthread -lm

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUUbuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34Clear Linux 346300.1940.3880.5820.7760.97SE +/- 0.008374, N = 6SE +/- 0.008602, N = 6SE +/- 0.008032, N = 6SE +/- 0.000743, N = 30.8463300.8504000.8546420.862061MIN: 0.8MIN: 0.8-O2 - MIN: 0.8-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 0.811. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUUbuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34Clear Linux 34630246810Min: 0.84 / Avg: 0.85 / Max: 0.89Min: 0.84 / Avg: 0.85 / Max: 0.89Min: 0.84 / Avg: 0.85 / Max: 0.89Min: 0.86 / Avg: 0.86 / Max: 0.861. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -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: Path TracerClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.043691215SE +/- 0.04, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 310.6010.4210.4210.42MIN: 7.58 / MAX: 10.87MIN: 7.81 / MAX: 10.53MIN: 7.94 / MAX: 10.53MIN: 7.87 / MAX: 10.64
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerClear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTSUbuntu 21.043691215Min: 10.53 / Avg: 10.6 / Max: 10.64Min: 10.42 / Avg: 10.42 / Max: 10.42Min: 10.42 / Avg: 10.42 / Max: 10.42Min: 10.42 / Avg: 10.42 / Max: 10.42

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUUbuntu 20.04.2 LTSClear Linux 34630Ubuntu 21.04Fedora Server 34100200300400500SE +/- 1.72, N = 3SE +/- 1.50, N = 3SE +/- 0.96, N = 3SE +/- 0.36, N = 3433.43433.80437.97440.63MIN: 414.41-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 414.44MIN: 422.42-O2 - MIN: 424.841. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUUbuntu 20.04.2 LTSClear Linux 34630Ubuntu 21.04Fedora Server 3480160240320400Min: 431.44 / Avg: 433.43 / Max: 436.86Min: 431.41 / Avg: 433.8 / Max: 436.58Min: 436.63 / Avg: 437.97 / Max: 439.83Min: 439.93 / Avg: 440.63 / Max: 441.141. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3420406080100SE +/- 1.45, N = 3SE +/- 0.84, N = 10SE +/- 0.91, N = 3SE +/- 1.03, N = 6107.77106.69106.14106.05MIN: 85.87 / MAX: 112.96MIN: 85.33 / MAX: 112.78MIN: 84.21 / MAX: 111.87MIN: 82.33 / MAX: 111.97
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.13Binary: Pathtracer ISPC - Model: Asian DragonClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3420406080100Min: 104.86 / Avg: 107.77 / Max: 109.28Min: 100.97 / Avg: 106.69 / Max: 108.8Min: 104.82 / Avg: 106.14 / Max: 107.9Min: 101.13 / Avg: 106.05 / Max: 108.06

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Ubuntu 21.04Clear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTS120K240K360K480K600KSE +/- 790.57, N = 3SE +/- 4180.58, N = 3SE +/- 888.88, N = 3SE +/- 5213.69, N = 3569303572684574742576988
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Ubuntu 21.04Clear Linux 34630Fedora Server 34Ubuntu 20.04.2 LTS100K200K300K400K500KMin: 567962 / Avg: 569303.33 / Max: 570699Min: 568140 / Avg: 572683.67 / Max: 581034Min: 573085 / Avg: 574742.33 / Max: 576128Min: 570309 / Avg: 576987.67 / Max: 587262

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0420406080100SE +/- 0.11, N = 3SE +/- 0.14, N = 3SE +/- 0.20, N = 3SE +/- 0.57, N = 3104.28104.32104.77105.651. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP HotSpot3DClear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0420406080100Min: 104.1 / Avg: 104.28 / Max: 104.48Min: 104.11 / Avg: 104.31 / Max: 104.59Min: 104.42 / Avg: 104.77 / Max: 105.09Min: 104.66 / Avg: 105.65 / Max: 106.631. (CXX) g++ options: -O2 -lOpenCL

oneDNN

This is a test of the Intel oneDNN 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. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 340.47510.95021.42531.90042.3755SE +/- 0.00152, N = 3SE +/- 0.01823, N = 3SE +/- 0.01729, N = 3SE +/- 0.01824, N = 32.087712.105972.106942.11134-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 2.03MIN: 2.03MIN: 2.03-O2 - MIN: 2.031. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.1.2Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34246810Min: 2.09 / Avg: 2.09 / Max: 2.09Min: 2.08 / Avg: 2.11 / Max: 2.14Min: 2.09 / Avg: 2.11 / Max: 2.14Min: 2.09 / Avg: 2.11 / Max: 2.151. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Sysbench

This is a benchmark of Sysbench with the built-in CPU and memory sub-tests. Sysbench is a scriptable multi-threaded benchmark tool based on LuaJIT. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPUUbuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34Clear Linux 3463050K100K150K200K250KSE +/- 252.80, N = 3SE +/- 275.74, N = 3SE +/- 276.88, N = 3SE +/- 23.74, N = 3214228.62213870.87212714.19212303.58-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm
OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 1.0.20Test: CPUUbuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34Clear Linux 3463040K80K120K160K200KMin: 213723.05 / Avg: 214228.62 / Max: 214485.5Min: 213320.06 / Avg: 213870.87 / Max: 214169.83Min: 212160.43 / Avg: 212714.19 / Max: 212994.22Min: 212264.19 / Avg: 212303.58 / Max: 212346.241. (CC) gcc options: -pthread -O2 -funroll-loops -rdynamic -ldl -laio -lm

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1Clear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS80160240320400SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.53, N = 3364.99366.37366.73367.80-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 364.69 / MAX: 366.31-O2 - MIN: 366.07 / MAX: 367.44-O3 - MIN: 365.95 / MAX: 373.2-O3 - MIN: 367 / MAX: 370.111. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1Clear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS70140210280350Min: 364.93 / Avg: 364.99 / Max: 365.06Min: 366.23 / Avg: 366.37 / Max: 366.5Min: 366.65 / Avg: 366.73 / Max: 366.88Min: 367.17 / Avg: 367.8 / Max: 368.851. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -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 2.14ATPase Simulation - 327,506 AtomsUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 340.06140.12280.18420.24560.307SE +/- 0.00044, N = 3SE +/- 0.00049, N = 3SE +/- 0.00032, N = 3SE +/- 0.00056, N = 30.270830.271280.271410.27274
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsUbuntu 20.04.2 LTSUbuntu 21.04Clear Linux 34630Fedora Server 3412345Min: 0.27 / Avg: 0.27 / Max: 0.27Min: 0.27 / Avg: 0.27 / Max: 0.27Min: 0.27 / Avg: 0.27 / Max: 0.27Min: 0.27 / Avg: 0.27 / Max: 0.27

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 343691215SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.00, N = 310.3810.3810.3810.31MIN: 8.13 / MAX: 10.75MIN: 9.17 / MAX: 10.53MIN: 7.58 / MAX: 10.75MIN: 8.77 / MAX: 10.42
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 343691215Min: 10.31 / Avg: 10.38 / Max: 10.42Min: 10.31 / Avg: 10.38 / Max: 10.42Min: 10.31 / Avg: 10.38 / Max: 10.42Min: 10.31 / Avg: 10.31 / Max: 10.31

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression SpeedFedora Server 346001200180024003000SE +/- 16.65, N = 32747.11. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression SpeedFedora Server 346001200180024003000SE +/- 2.07, N = 72698.61. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Decompression SpeedFedora Server 347001400210028003500SE +/- 8.89, N = 143118.51. (CC) gcc options: -O3 -pthread -lz -llzma

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Decompression SpeedFedora Server 347001400210028003500SE +/- 8.55, N = 23057.51. (CC) gcc options: -O3 -pthread -lz -llzma

OSPray

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

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisFedora Server 34Clear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 318.8718.8718.8718.87MIN: 12.05 / MAX: 19.23MIN: 13.16 / MAX: 19.61MIN: 12.82 / MAX: 19.23MIN: 13.33 / MAX: 19.23
OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisFedora Server 34Clear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04510152025Min: 18.87 / Avg: 18.87 / Max: 18.87Min: 18.87 / Avg: 18.87 / Max: 18.87Min: 18.87 / Avg: 18.87 / Max: 18.87Min: 18.87 / Avg: 18.87 / Max: 18.87

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression SpeedClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 346001200180024003000SE +/- 58.63, N = 12SE +/- 25.03, N = 15SE +/- 24.65, N = 3SE +/- 30.33, N = 152656.62128.72096.72069.1-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 8 - Compression SpeedClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 345001000150020002500Min: 2334.5 / Avg: 2656.63 / Max: 2921.4Min: 1981.7 / Avg: 2128.71 / Max: 2367.8Min: 2071.8 / Avg: 2096.7 / Max: 2146Min: 1822.7 / Avg: 2069.11 / Max: 2317.71. (CC) gcc options: -O3 -pthread -lz

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: WritesClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS70K140K210K280K350KSE +/- 2265.50, N = 3SE +/- 2127.47, N = 15SE +/- 2263.61, N = 15SE +/- 1368.32, N = 15312015108941106863103051
OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: WritesClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS50K100K150K200K250KMin: 309114 / Avg: 312015.33 / Max: 316480Min: 97223 / Avg: 108941.4 / Max: 122300Min: 90540 / Avg: 106863.2 / Max: 119298Min: 95165 / Avg: 103050.73 / Max: 116380

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.04100200300400500SE +/- 0.15, N = 3SE +/- 0.44, N = 3SE +/- 6.88, N = 15SE +/- 9.40, N = 15354.30374.20401.86443.55-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake - MIN: 353.49 / MAX: 357.29-O3 - MIN: 371.48 / MAX: 419.18-O2 - MIN: 368.12 / MAX: 804.92-O3 - MIN: 367.81 / MAX: 808.751. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2Clear Linux 34630Ubuntu 20.04.2 LTSFedora Server 34Ubuntu 21.0480160240320400Min: 354 / Avg: 354.3 / Max: 354.47Min: 373.34 / Avg: 374.2 / Max: 374.78Min: 369.7 / Avg: 401.86 / Max: 457.57Min: 393.52 / Avg: 443.55 / Max: 497.971. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -rdynamic -ldl

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average LatencyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 340.06390.12780.19170.25560.3195SE +/- 0.005, N = 15SE +/- 0.003, N = 4SE +/- 0.003, N = 15SE +/- 0.006, N = 130.2540.2630.2660.284-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O2-O21. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average LatencyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 3412345Min: 0.23 / Avg: 0.25 / Max: 0.29Min: 0.26 / Avg: 0.26 / Max: 0.27Min: 0.25 / Avg: 0.27 / Max: 0.29Min: 0.26 / Avg: 0.28 / Max: 0.341. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read OnlyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34200K400K600K800K1000KSE +/- 19682.88, N = 15SE +/- 10643.51, N = 4SE +/- 10755.01, N = 15SE +/- 17861.91, N = 13990608954214946634886858-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O2-O21. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read OnlyClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 34200K400K600K800K1000KMin: 864101.92 / Avg: 990607.63 / Max: 1107203.18Min: 922367.58 / Avg: 954213.54 / Max: 966977.37Min: 861584.42 / Avg: 946633.68 / Max: 989623.06Min: 739211.23 / Avg: 886858.17 / Max: 951479.731. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

KeyDB

A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Ubuntu 21.04Clear Linux 34630Ubuntu 20.04.2 LTS110K220K330K440K550KSE +/- 4128.37, N = 3SE +/- 2367.62, N = 3SE +/- 18099.02, N = 15524178.09503991.58421171.591. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Ubuntu 21.04Clear Linux 34630Ubuntu 20.04.2 LTS90K180K270K360K450KMin: 516045.54 / Avg: 524178.09 / Max: 529479.93Min: 499256.52 / Avg: 503991.58 / Max: 506395.83Min: 317166.44 / Avg: 421171.59 / Max: 510934.691. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Tungsten Renderer

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Volumetric CausticFedora Server 34Ubuntu 21.04Clear Linux 34630Ubuntu 20.04.2 LTS48121620SE +/- 0.74, N = 12SE +/- 0.44, N = 15SE +/- 0.48, N = 15SE +/- 0.39, N = 1513.0913.4513.8713.87-O2-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O3 -lIlmImf -lIlmThread -lImath -lHalf -lIex -lz1. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl
OpenBenchmarking.orgSeconds, Fewer Is BetterTungsten Renderer 0.2.2Scene: Volumetric CausticFedora Server 34Ubuntu 21.04Clear Linux 34630Ubuntu 20.04.2 LTS48121620Min: 10.5 / Avg: 13.09 / Max: 18.86Min: 11.14 / Avg: 13.45 / Max: 16.22Min: 11.16 / Avg: 13.87 / Max: 15.79Min: 11.2 / Avg: 13.87 / Max: 15.921. (CXX) g++ options: -std=c++0x -march=core2 -msse2 -msse3 -mssse3 -mno-sse4.1 -mno-sse4.2 -mno-sse4a -mno-avx -mno-fma -mno-bmi2 -mno-avx2 -mno-xop -mno-fma4 -mno-avx512f -mno-avx512vl -mno-avx512pf -mno-avx512er -mno-avx512cd -mno-avx512dq -mno-avx512bw -mno-avx512ifma -mno-avx512vbmi -fstrict-aliasing -rdynamic -ljpeg -lpthread -ldl

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34612182430SE +/- 0.17, N = 9SE +/- 0.35, N = 13SE +/- 0.36, N = 12SE +/- 0.46, N = 1320.7424.8125.9626.51
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 34612182430Min: 20.43 / Avg: 20.74 / Max: 22.04Min: 24.19 / Avg: 24.81 / Max: 28.88Min: 25.18 / Avg: 25.96 / Max: 29.78Min: 25.71 / Avg: 26.51 / Max: 31.88

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, LosslessClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 343691215SE +/- 0.013, N = 3SE +/- 0.146, N = 15SE +/- 0.118, N = 12SE +/- 0.100, N = 155.8158.5228.9679.3321. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10, LosslessClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 343691215Min: 5.8 / Avg: 5.82 / Max: 5.84Min: 7.83 / Avg: 8.52 / Max: 9.73Min: 8.25 / Avg: 8.97 / Max: 9.44Min: 8.2 / Avg: 9.33 / Max: 9.791. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10Clear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1.20622.41243.61864.82486.031SE +/- 0.030, N = 15SE +/- 0.074, N = 15SE +/- 0.070, N = 15SE +/- 0.099, N = 153.0665.0505.1865.3611. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10Clear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS246810Min: 2.95 / Avg: 3.07 / Max: 3.39Min: 4.78 / Avg: 5.05 / Max: 5.48Min: 4.86 / Avg: 5.19 / Max: 5.53Min: 4.88 / Avg: 5.36 / Max: 6.171. (CXX) g++ options: -O3 -fPIC -lm

asmFish

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

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3440M80M120M160M200MSE +/- 2090545.83, N = 3SE +/- 1805964.50, N = 3SE +/- 3374554.91, N = 12SE +/- 1128044.08, N = 3174596848172711481171613950168644528
OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 DepthClear Linux 34630Ubuntu 21.04Ubuntu 20.04.2 LTSFedora Server 3430M60M90M120M150MMin: 171387534 / Avg: 174596847.67 / Max: 178522361Min: 169883036 / Avg: 172711480.67 / Max: 176071069Min: 164700846 / Avg: 171613949.58 / Max: 208012828Min: 167016977 / Avg: 168644528.33 / Max: 170811359

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1326395265SE +/- 0.85, N = 3SE +/- 2.57, N = 12SE +/- 2.83, N = 12SE +/- 2.40, N = 1259.9857.5954.8349.87
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialClear Linux 34630Ubuntu 21.04Fedora Server 34Ubuntu 20.04.2 LTS1224364860Min: 58.28 / Avg: 59.98 / Max: 60.94Min: 40.92 / Avg: 57.59 / Max: 64.18Min: 38.92 / Avg: 54.83 / Max: 63.84Min: 39.17 / Avg: 49.87 / Max: 64.16

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS12002400360048006000SE +/- 14.30, N = 4SE +/- 150.49, N = 20SE +/- 153.92, N = 20SE +/- 194.90, N = 203590508853205518
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS10002000300040005000Min: 3565 / Avg: 3589.75 / Max: 3625Min: 4375 / Avg: 5088.05 / Max: 6473Min: 4359 / Avg: 5320.05 / Max: 6488Min: 4400 / Avg: 5518.05 / Max: 6968

Java Gradle Build

This test runs Java software project builds using the Gradle build system. It is intended to give developers an idea as to the build performance for development activities and build servers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle BuildGradle Build: ReactorClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS80160240320400SE +/- 8.63, N = 6SE +/- 8.29, N = 7SE +/- 5.06, N = 9SE +/- 10.71, N = 9301.41354.03366.36382.99
OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle BuildGradle Build: ReactorClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS70140210280350Min: 266.9 / Avg: 301.41 / Max: 328.67Min: 333.35 / Avg: 354.03 / Max: 386.94Min: 341.71 / Avg: 366.36 / Max: 386.86Min: 319.9 / Avg: 382.99 / Max: 426.9

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS714212835SE +/- 0.47, N = 15SE +/- 0.30, N = 4SE +/- 0.02, N = 3SE +/- 0.64, N = 1231.1527.2123.8722.08-O3 -pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake-O2-O3-O31. (CXX) g++ options: -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinClear Linux 34630Fedora Server 34Ubuntu 21.04Ubuntu 20.04.2 LTS714212835Min: 27.94 / Avg: 31.15 / Max: 33.48Min: 26.78 / Avg: 27.2 / Max: 28.1Min: 23.83 / Avg: 23.87 / Max: 23.9Min: 19.16 / Avg: 22.08 / Max: 25.281. (CXX) g++ options: -pthread -lm

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 341428425670SE +/- 0.36, N = 15SE +/- 1.30, N = 15SE +/- 0.67, N = 3SE +/- 0.49, N = 340.8260.9362.0462.211. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteClear Linux 34630Ubuntu 20.04.2 LTSUbuntu 21.04Fedora Server 341224364860Min: 38.69 / Avg: 40.82 / Max: 43.11Min: 53.86 / Avg: 60.93 / Max: 72.21Min: 60.71 / Avg: 62.04 / Max: 62.82Min: 61.24 / Avg: 62.21 / Max: 62.811. (CXX) g++ options: -O2 -lOpenCL

NAS Parallel Benchmarks

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

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CUbuntu 21.04Ubuntu 20.04.2 LTSClear Linux 34630Fedora Server 3413002600390052006500SE +/- 51.84, N = 15SE +/- 81.23, N = 15SE +/- 165.45, N = 12SE +/- 13.65, N = 36259.886011.425425.402213.51-lopen-rte -lopen-pal -lhwloc -ldl -levent_core -levent_pthreads -lutil -lm -lrt -lz-pipe -fexceptions -fstack-protector -m64 -ffat-lto-objects -fno-trapping-math -mtune=skylake1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 21.04: Open MPI 4.1.03. Ubuntu 20.04.2 LTS: Open MPI 4.0.34. Clear Linux 34630: 3.2
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CUbuntu 21.04Ubuntu 20.04.2 LTSClear Linux 34630Fedora Server 3411002200330044005500Min: 5918.12 / Avg: 6259.88 / Max: 6635.86Min: 5432.62 / Avg: 6011.42 / Max: 6369.67Min: 3686.05 / Avg: 5425.4 / Max: 5740.07Min: 2193.42 / Avg: 2213.51 / Max: 2239.561. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Ubuntu 21.04: Open MPI 4.1.03. Ubuntu 20.04.2 LTS: Open MPI 4.0.34. Clear Linux 34630: 3.2

126 Results Shown

NAS Parallel Benchmarks
DaCapo Benchmark
oneDNN
Kvazaar
x265
PJSIP:
  OPTIONS, Stateful
  INVITE
libgav1
DaCapo Benchmark
Kvazaar
SVT-HEVC
x265
SVT-VP9
Tungsten Renderer
SVT-AV1
Kvazaar
SVT-HEVC
Tungsten Renderer
Timed Apache Compilation
libavif avifenc
Zstd Compression
Tungsten Renderer
libavif avifenc
TensorFlow Lite:
  Mobilenet Float
  Mobilenet Quant
Timed PHP Compilation
OpenVKL
oneDNN
SVT-HEVC
PHPBench
Liquid-DSP
Zstd Compression
OpenVKL
TensorFlow Lite
BlogBench
NWChem
Node.js V8 Web Tooling Benchmark
Liquid-DSP:
  32 - 256 - 57
  16 - 256 - 57
  128 - 256 - 57
BlogBench
PyBench
OSPray
Embree
Zstd Compression
OSPray
Zstd Compression
Xcompact3d Incompact3d:
  input.i3d 193 Cells Per Direction
  X3D-benchmarking input.i3d
oneDNN
OSPray
Zstd Compression
PJSIP
GROMACS
Chia Blockchain VDF
Coremark
oneDNN
TensorFlow Lite
oneDNN
Rodinia
Zstd Compression
oneDNN
Timed MrBayes Analysis
Blender
Zstd Compression
Stockfish
POV-Ray
oneDNN
Chia Blockchain VDF
OpenVKL
Blender:
  Fishy Cat - CPU-Only
  BMW27 - CPU-Only
Helsing
TensorFlow Lite
Sysbench
oneDNN:
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  IP Shapes 1D - f32 - CPU
RELION
oneDNN
Embree
oneDNN
OSPray
oneDNN:
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
  Recurrent Neural Network Training - f32 - CPU
Blender
oneDNN
Blender
High Performance Conjugate Gradient
SecureMark
oneDNN:
  Recurrent Neural Network Inference - u8s8f32 - CPU
  IP Shapes 1D - bf16bf16bf16 - CPU
  Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPU
LAMMPS Molecular Dynamics Simulator
oneDNN
OSPray
oneDNN
Embree
TensorFlow Lite
Rodinia
oneDNN
Sysbench
TNN
NAMD
OSPray
Zstd Compression:
  19, Long Mode - Decompression Speed
  19 - Decompression Speed
  8 - Decompression Speed
  3 - Decompression Speed
OSPray
Zstd Compression
Apache Cassandra
TNN
PostgreSQL pgbench:
  100 - 250 - Read Only - Average Latency
  100 - 250 - Read Only
KeyDB
Tungsten Renderer
Timed Linux Kernel Compilation
libavif avifenc:
  10, Lossless
  10
asmFish
Intel Open Image Denoise
DaCapo Benchmark
Java Gradle Build
LAMMPS Molecular Dynamics Simulator
Rodinia
NAS Parallel Benchmarks

Ubuntu 21.04

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

OS: Ubuntu 21.04, Kernel: 5.11.0-17-generic (x86_64), Desktop: GNOME Shell 3.38.4, Display Server: X Server, Compiler: GCC 10.3.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-gDeRY6/gcc-10-10.3.0/debian/tmp-gcn/usr,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-build-config=bootstrap-lto-lean --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
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2)
Python Notes: Python 3.9.4
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 13 May 2021 20:13 by user phoronix.

Ubuntu 20.04.2 LTS

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

OS: Ubuntu 20.04, Kernel: 5.8.0-53-generic (x86_64), Desktop: GNOME Shell 3.36.7, Display Server: X Server 1.20.9, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/usr,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
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment (build 11.0.11+9-Ubuntu-0ubuntu2.20.04)
Python Notes: Python 3.8.5
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 17 May 2021 12:06 by user phoronix.

Clear Linux 34630

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

OS: Clear Linux OS 34630, Kernel: 5.10.19-1032.native (x86_64), Desktop: GNOME Shell 40.0, Display Server: X Server, Compiler: GCC 11.1.1 20210517 releases/gcc-11.1.0-132-g7d91dd2efb + Clang 11.1.0 + LLVM 11.1.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: FFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,--enable-new-dtags -Wa,-mbranches-within-32B-boundaries" CXXFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries -fvisibility-inlines-hidden -Wl,--enable-new-dtags" FCFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -malign-data=abi -fno-semantic-interposition -ftree-vectorize -ftree-loop-vectorize -Wl,-sort-common -Wl,--enable-new-dtags" CFLAGS="-g -O3 -feliminate-unused-debug-types -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=32 -Wformat -Wformat-security -m64 -fasynchronous-unwind-tables -Wp,-D_REENTRANT -ftree-loop-distribute-patterns -Wl,-z -Wl,now -Wl,-z -Wl,relro -fno-semantic-interposition -ffat-lto-objects -fno-trapping-math -Wl,-sort-common -Wl,--enable-new-dtags -mtune=skylake -Wa,-mbranches-within-32B-boundaries" THEANO_FLAGS="floatX=float32,openmp=true,gcc.cxxflags="-ftree-vectorize -mavx""
Compiler Notes: --build=x86_64-generic-linux --disable-libmpx --disable-libunwind-exceptions --disable-multiarch --disable-vtable-verify --disable-werror --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-clocale=gnu --enable-default-pie --enable-gnu-indirect-function --enable-languages=c,c++,fortran,go --enable-ld=default --enable-libstdcxx-pch --enable-lto --enable-multilib --enable-plugin --enable-shared --enable-threads=posix --exec-prefix=/usr --includedir=/usr/include --target=x86_64-generic-linux --with-arch=westmere --with-gcc-major-version-only --with-glibc-version=2.19 --with-gnu-ld --with-isl --with-ppl=yes --with-tune=skylake-avx512
Disk Notes: MQ-DEADLINE / relatime,rw,stripe=256 / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate performance - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment (build 1.8.0-u252-ga-b00)
Python Notes: Python 3.9.5
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 18 May 2021 22:01 by user phoronix.

Fedora Server 34

Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP

OS: Fedora 34, Kernel: 5.11.20-300.fc34.x86_64 (x86_64), Compiler: GCC 11.1.1 20210428, File-System: xfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver
Disk Notes: NONE / attr2,inode64,logbsize=32k,logbufs=8,noquota,relatime,rw,seclabel / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0xd000270
Java Notes: OpenJDK Runtime Environment 18.9 (build 11.0.11+9)
Python Notes: Python 3.9.5
Security Notes: SELinux + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 19 May 2021 07:16 by user .