12400 march

Intel Core i5-12400 testing with a MSI PRO Z690-A WIFI DDR4(MS-7D25) v1.0 (Dasharo coreboot+UEFI v1.0.0 BIOS) and MSI Intel ADL-S GT1 14GB on Ubuntu 22.04 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 2303066-NE-12400MARC37
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:

Audio Encoding 2 Tests
AV1 3 Tests
Timed Code Compilation 2 Tests
C/C++ Compiler Tests 11 Tests
CPU Massive 14 Tests
Creator Workloads 19 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 2 Tests
Database Test Suite 5 Tests
Encoding 10 Tests
Game Development 2 Tests
HPC - High Performance Computing 9 Tests
Imaging 3 Tests
Common Kernel Benchmarks 3 Tests
Machine Learning 5 Tests
Multi-Core 19 Tests
NVIDIA GPU Compute 2 Tests
Intel oneAPI 4 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 3 Tests
Python Tests 8 Tests
Server 6 Tests
Server CPU Tests 8 Tests
Single-Threaded 2 Tests
Video Encoding 8 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
a
March 03 2023
  2 Days, 44 Minutes
b
March 05 2023
  11 Hours, 15 Minutes
c
March 05 2023
  11 Hours, 18 Minutes
Invert Hiding All Results Option
  23 Hours, 46 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):

a

Processor: Intel Core i5-12400 @ 5.60GHz (6 Cores / 12 Threads), Motherboard: MSI PRO Z690-A WIFI DDR4(MS-7D25) v1.0 (Dasharo coreboot+UEFI v1.0.0 BIOS), Chipset: Intel Device 7aa7, Memory: 16GB, Disk: Western Digital WD_BLACK SN750 SE 500GB, Graphics: MSI Intel ADL-S GT1 14GB (1450MHz), Audio: Realtek ALC897, Monitor: DELL S2409W, Network: Intel I225-V + Intel Device 7af0

OS: Ubuntu 22.04, Kernel: 5.15.0-47-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, Vulkan: 1.2.204, Compiler: GCC 11.2.0, 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-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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

b

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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

c

OS: Ubuntu 22.04, Kernel: 5.15.0-47-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, Vulkan: 1.2.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 1920x1080

12400 marchOpenBenchmarking.orgPhoronix Test SuiteIntel Core i5-12400 @ 5.60GHz (6 Cores / 12 Threads)MSI PRO Z690-A WIFI DDR4(MS-7D25) v1.0 (Dasharo coreboot+UEFI v1.0.0 BIOS)Intel Device 7aa716GBWestern Digital WD_BLACK SN750 SE 500GBMSI Intel ADL-S GT1 14GB (1450MHz)Realtek ALC897DELL S2409WIntel I225-V + Intel Device 7af0Ubuntu 22.045.15.0-47-generic (x86_64)GNOME Shell 42.2X Server + Wayland4.6 Mesa 22.0.11.2.204GCC 11.2.0GCC 11.3.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilersFile-SystemScreen Resolution12400 March BenchmarksSystem Logs- Transparent Huge Pages: madvise- a: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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 - b: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --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 - c: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --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 - Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9 - Python 3.10.6- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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

abcResult OverviewPhoronix Test Suite100%105%110%114%119%RocksDBFLAC Audio EncodingPostgreSQLKeyDBEnCodecTimed FFmpeg CompilationVP9 libvpx EncodingNeural Magic DeepSparseSVT-AV1nekRSJPEG XL Decoding libjxlAOM AV1miniBUDEClickHouseStress-NGuvg266Cpuminer-OptJPEG XL libjxlKvazaarCockroachDBEmbreeXmrignginxFFmpegVVenCTimed Linux Kernel CompilationOpenVINOOpenVKLBRL-CADStargate Digital Audio WorkstationOpenRadiossoneDNNBlenderGROMACSTensorFlowOpenEMSlibavif avifencScikit-LearnZstd CompressionUnvanquished

12400 marchrocksdb: Rand Fill Syncpgbench: 100 - 800 - Read Write - Average Latencypgbench: 100 - 800 - Read Writeonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUsvt-av1: Preset 12 - Bosphorus 4Kpgbench: 100 - 1000 - Read Writepgbench: 100 - 1000 - Read Write - Average Latencydeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamopenvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16 - CPUencode-flac: WAV To FLACencodec: 1.5 kbpsvpxenc: Speed 0 - Bosphorus 1080pcockroach: MoVR - 128keydb: SET - 100keydb: GET - 100deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUaom-av1: Speed 0 Two-Pass - Bosphorus 1080pcpuminer-opt: Blake-2 Scpuminer-opt: x25xencodec: 6 kbpscpuminer-opt: LBC, LBRY Creditsaom-av1: Speed 0 Two-Pass - Bosphorus 4Kencodec: 3 kbpsffmpeg: libx264 - Video On Demandffmpeg: libx264 - Video On Demandrocksdb: Update Randvpxenc: Speed 5 - Bosphorus 4Kkeydb: LPUSH - 50stress-ng: Glibc C String Functionspgbench: 1 - 1000 - Read Write - Average Latencydeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streampgbench: 1 - 1000 - Read Writekeydb: HMSET - 100keydb: SADD - 100pgbench: 1 - 500 - Read Only - Average Latencypgbench: 1 - 500 - Read Onlykvazaar: Bosphorus 4K - Very Fastclickhouse: 100M Rows Hits Dataset, Third Runcockroach: MoVR - 256kvazaar: Bosphorus 4K - Ultra Fastpgbench: 1 - 800 - Read Write - Average Latencypgbench: 1 - 800 - Read Writekeydb: LPUSH - 100pgbench: 100 - 800 - Read Only - Average Latencypgbench: 100 - 800 - Read Onlypgbench: 1 - 500 - Read Write - Average Latencypgbench: 100 - 250 - Read Onlypgbench: 100 - 250 - Read Only - Average Latencyonednn: IP Shapes 1D - f32 - CPUjpegxl-decode: Allminibude: OpenMP - BM1encodec: 24 kbpsminibude: OpenMP - BM1pgbench: 1 - 500 - Read Writepgbench: 1 - 800 - Read Only - Average Latencypgbench: 1 - 800 - Read Onlycpuminer-opt: Garlicoinclickhouse: 100M Rows Hits Dataset, Second Runpgbench: 100 - 1000 - Read Onlydeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamuvg266: Bosphorus 1080p - Mediumpgbench: 100 - 1000 - Read Only - Average Latencyuvg266: Bosphorus 1080p - Slowcockroach: KV, 10% Reads - 512rocksdb: Rand Readbuild-ffmpeg: Time To Compilevvenc: Bosphorus 4K - Fasterdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamcockroach: KV, 60% Reads - 128vvenc: Bosphorus 1080p - Fasterdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamrocksdb: Seq Fillpgbench: 1 - 100 - Read Write - Average Latencystress-ng: Forkingpgbench: 1 - 100 - Read Writesvt-av1: Preset 4 - Bosphorus 4Kpgbench: 1 - 100 - Read Onlykvazaar: Bosphorus 4K - Super Fastpgbench: 1 - 100 - Read Only - Average Latencydeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streampgbench: 1 - 250 - Read Write - Average Latencypgbench: 1 - 250 - Read Writecockroach: KV, 50% Reads - 128deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamffmpeg: libx265 - Video On Demandffmpeg: libx265 - Video On Demanduvg266: Bosphorus 4K - Ultra Fastuvg266: Bosphorus 1080p - Very Fastnekrs: TurboPipe Periodiconednn: IP Shapes 3D - u8s8f32 - CPUkvazaar: Bosphorus 4K - Slowcockroach: KV, 95% Reads - 128stress-ng: Pthreadembree: Pathtracer - Crowndeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamuvg266: Bosphorus 4K - Slowffmpeg: libx264 - Livecockroach: KV, 50% Reads - 256openvino: Person Detection FP16 - CPUffmpeg: libx264 - Liveaom-av1: Speed 6 Two-Pass - Bosphorus 1080popenvino: Face Detection FP16-INT8 - CPUcompress-zstd: 19 - Compression Speeduvg266: Bosphorus 4K - Very Fastkeydb: HMSET - 50deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamuvg266: Bosphorus 4K - Super Fastopenvino: Age Gender Recognition Retail 0013 FP16 - CPUkvazaar: Bosphorus 1080p - Slowpgbench: 1 - 250 - Read Only - Average Latencynginx: 20onednn: Recurrent Neural Network Training - f32 - CPUpgbench: 1 - 250 - Read Onlyjpegxl: JPEG - 90embree: Pathtracer ISPC - Crowndeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamopenvino: Face Detection FP16-INT8 - CPUjpegxl: PNG - 80avifenc: 2cockroach: KV, 60% Reads - 256deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamjpegxl: JPEG - 80cockroach: MoVR - 512avifenc: 6, Losslessaom-av1: Speed 4 Two-Pass - Bosphorus 4Kopenvino: Age Gender Recognition Retail 0013 FP16 - CPUjpegxl: PNG - 90cockroach: KV, 60% Reads - 1024pgbench: 1 - 1000 - Read Onlypgbench: 1 - 1000 - Read Only - Average Latencyonednn: IP Shapes 3D - f32 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamxmrig: Monero - 1Mtensorflow: CPU - 16 - GoogLeNetclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachecpuminer-opt: Ringcoincpuminer-opt: Skeincoinvvenc: Bosphorus 4K - Fastkvazaar: Bosphorus 4K - Mediumonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUcockroach: KV, 60% Reads - 512onednn: Recurrent Neural Network Inference - f32 - CPUpgbench: 100 - 100 - Read Only - Average Latencyopenvino: Machine Translation EN To DE FP16 - CPUsvt-av1: Preset 13 - Bosphorus 4Kcockroach: KV, 10% Reads - 128jpegxl-decode: 1pgbench: 100 - 100 - Read Onlyffmpeg: libx265 - Livebuild-linux-kernel: allmodconfigopenvino: Machine Translation EN To DE FP16 - CPUffmpeg: libx265 - Livecockroach: KV, 95% Reads - 256onednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUcpuminer-opt: Magiopenvkl: vklBenchmark Scalarjpegxl: JPEG - 100jpegxl: PNG - 100openradioss: Rubber O-Ring Seal Installationdeepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Streamrocksdb: Read Rand Write Randcockroach: KV, 10% Reads - 256vvenc: Bosphorus 1080p - Fastcompress-zstd: 3 - Compression Speedcpuminer-opt: Myriad-Groestluvg266: Bosphorus 4K - Mediumvpxenc: Speed 5 - Bosphorus 1080pdeepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Streamstress-ng: Glibc Qsort Data Sortingopenvino: Person Detection FP32 - CPUdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamsvt-av1: Preset 8 - Bosphorus 4Kkvazaar: Bosphorus 1080p - Ultra Fastnginx: 100onednn: Recurrent Neural Network Training - u8s8f32 - CPUstargate: 480000 - 512blender: BMW27 - CPU-Onlydeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamopenvino: Person Detection FP16 - CPUopenvino: Face Detection FP16 - CPUavifenc: 6stargate: 96000 - 1024aom-av1: Speed 6 Two-Pass - Bosphorus 4Kopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamaom-av1: Speed 8 Realtime - Bosphorus 1080popenvino: Weld Porosity Detection FP16-INT8 - CPUcompress-zstd: 8, Long Mode - Compression Speedcockroach: KV, 50% Reads - 512stress-ng: NUMAopenradioss: INIVOL and Fluid Structure Interaction Drop Containerstargate: 480000 - 1024build-linux-kernel: defconfigonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUcpuminer-opt: Deepcoinsvt-av1: Preset 4 - Bosphorus 1080pembree: Pathtracer - Asian Dragonrocksdb: Read While Writingnginx: 200openvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUtensorflow: CPU - 16 - AlexNetopenvino: Person Detection FP32 - CPUffmpeg: libx264 - Platformcompress-zstd: 19, Long Mode - Compression Speedffmpeg: libx264 - Platformtensorflow: CPU - 256 - AlexNetnginx: 1000stargate: 192000 - 1024openvkl: vklBenchmark ISPCaom-av1: Speed 9 Realtime - Bosphorus 4Kstress-ng: Mallocopenvino: Face Detection FP16 - CPUscikit-learn: TSNE MNIST Datasetscikit-learn: MNIST Datasetembree: Pathtracer ISPC - Asian Dragonuvg266: Bosphorus 1080p - Ultra Fastavifenc: 0minibude: OpenMP - BM2minibude: OpenMP - BM2cockroach: MoVR - 1024onednn: Recurrent Neural Network Inference - u8s8f32 - CPUaom-av1: Speed 6 Realtime - Bosphorus 4Kaom-av1: Speed 8 Realtime - Bosphorus 4Kcpuminer-opt: scrypttensorflow: CPU - 64 - AlexNetstress-ng: Memory Copyingvpxenc: Speed 0 - Bosphorus 4Kbrl-cad: VGR Performance Metrictensorflow: CPU - 32 - AlexNetstargate: 44100 - 1024stargate: 96000 - 512openradioss: Bird Strike on Windshieldstress-ng: SENDFILEopenradioss: Cell Phone Drop Testopenvino: Vehicle Detection FP16-INT8 - CPUembree: Pathtracer ISPC - Asian Dragon Objffmpeg: libx265 - Uploadtensorflow: CPU - 64 - GoogLeNetffmpeg: libx265 - Uploadopenvino: Vehicle Detection FP16-INT8 - CPUdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Streamdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Streamtensorflow: CPU - 32 - GoogLeNetopenems: pyEMS Couplercockroach: KV, 10% Reads - 1024onednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUuvg266: Bosphorus 1080p - Super Fastsvt-av1: Preset 8 - Bosphorus 1080ptensorflow: CPU - 256 - GoogLeNetxmrig: Wownero - 1Mdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamtensorflow: CPU - 512 - AlexNetdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamffmpeg: libx264 - Uploadpgbench: 100 - 500 - Read Only - Average Latencycompress-zstd: 19, Long Mode - Decompression Speedpgbench: 100 - 500 - Read Onlyffmpeg: libx264 - Uploadcompress-zstd: 8 - Compression Speedsvt-av1: Preset 12 - Bosphorus 1080pstress-ng: Mutexsvt-av1: Preset 13 - Bosphorus 1080pblender: Pabellon Barcelona - CPU-Onlyopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUkvazaar: Bosphorus 1080p - Very Faststargate: 192000 - 512stress-ng: System V Message Passingaom-av1: Speed 10 Realtime - Bosphorus 4Kblender: Barbershop - CPU-Onlyffmpeg: libx265 - Platformffmpeg: libx265 - Platformonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUcompress-zstd: 3, Long Mode - Compression Speedcompress-zstd: 3 - Decompression Speednginx: 500gromacs: MPI CPU - water_GMX50_barestress-ng: Matrix Mathstargate: 44100 - 512openradioss: Bumper Beamcockroach: KV, 95% Reads - 512cpuminer-opt: Triple SHA-256, Onecoinblender: Classroom - CPU-Onlycompress-zstd: 19 - Decompression Speedcompress-zstd: 12 - Compression Speedkvazaar: Bosphorus 1080p - Super Fastcockroach: KV, 50% Reads - 1024unvanquished: 1920 x 1080 - Highstress-ng: Zlibunvanquished: 1920 x 1080 - Ultracockroach: KV, 95% Reads - 1024compress-zstd: 12 - Decompression Speedtensorflow: CPU - 512 - GoogLeNetkvazaar: Bosphorus 1080p - Mediumcompress-zstd: 8, Long Mode - Decompression Speedopenems: openEMS MSL_NotchFilterscikit-learn: Sparse Rand Projections, 100 Iterationscpuminer-opt: Quad SHA-256, Pyriteblender: Fishy Cat - CPU-Onlyonednn: Convolution Batch Shapes Auto - f32 - CPUstress-ng: Semaphoresstress-ng: Vector Mathcompress-zstd: 8 - Decompression Speedstress-ng: Cryptostress-ng: Hashstress-ng: Pollcompress-zstd: 3, Long Mode - Decompression Speedembree: Pathtracer - Asian Dragon Objstress-ng: Function Callunvanquished: 1920 x 1080 - Mediumstress-ng: CPU Stressstress-ng: x86_64 RdRandaom-av1: Speed 4 Two-Pass - Bosphorus 1080procksdb: Rand Fillstress-ng: Context Switchingstress-ng: Socket Activitystress-ng: CPU Cachestress-ng: IO_uringstress-ng: Atomicstress-ng: MEMFDstress-ng: Futexstress-ng: MMAPpgbench: 100 - 500 - Read Write - Average Latencypgbench: 100 - 500 - Read Writepgbench: 100 - 250 - Read Write - Average Latencypgbench: 100 - 250 - Read Writepgbench: 100 - 100 - Read Write - Average Latencypgbench: 100 - 100 - Read Writekeydb: LPOP - 100keydb: SADD - 50keydb: LPOP - 50keydb: SET - 50keydb: GET - 50onednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUavifenc: 10, Losslessaom-av1: Speed 10 Realtime - Bosphorus 1080paom-av1: Speed 9 Realtime - Bosphorus 1080paom-av1: Speed 6 Realtime - Bosphorus 1080pabc1351188.70642522437.9688.4264070246.31370.834142.337020.142149.640127.97143.2117.32436.91512.18176.41069422.991283175.5822.0196136.27391.526650.50282207344.3038.516184380.1538.663155.2748.8062994412.731323318.971220179.933797.31474.098840.468038.336626.0851263361525.211148270.291.39335916516.45133.73181.627.832891.941277976117.593.0312639501561.2693945260.6344.09853217.33244.48944.5589.7803212.6063070181836.59130.8424154051.147219.551424.114.14121.5325625.04161411061.4686.065129.735120255.720.15523.14011502792258.42754276.343872.18561770620.380.16265.1167681.41836716123.046.100943.25175.1316.9962.33298446555562.504626.3452512.6128698.018.487591.194532.89274.1121.0530043.81.17239.8631.697.2112.613.25511218.215.640113.985560.2732.530.54162961.454329.654620717.799.3366530.8351552.798.0588.24634782.85.64347.93177.413.4615.121.067.9432151.82892053.45810.3722530.17892901.838.89112.081560.09478532.6966.531.2021935029.42438.210.185184.3795.5659270.550.0554141042.021605.96521.68120.1749613.54334.97331.49680.680.69236.8732.7332154665517827.67.9771567.8112374.5832.7530.5731101.423438.1815.567064.232727.441123.2667993.524341.922.624567172.015.4650182.98233384.141.759.1752.0716699.118.41188.1731.870.95125.27077.9821139.13712.70198.432781.2244.13862.762.743784117.65318.27506392.107.7379.6958151851561096.58276.7814.4460.811.15145.957.0051.90107.9141663.171.39866312057.995226456.992257.1726.91490.14611.058880.01195.068243.6629.747177.42406.1355.3345.87110.8788.602301.695.7613293177.952.7051221.930960378.02177365.84141.00348.499.570621.2041.05119.1311.4720.326949.178940.1739.7226050.82.0002067.0981.45343.375359.8371.27078.0798117.88184.65465.415413.501.5731444.4317872187.09202.6348.1144254720.23398.861602.176331.0174.791.2726778216877.6058.761913.7144.40170.613.59719869.31592.651922.480.85245892.792.578599187.4943281.778157493.251516.377.193.4830338.7170.8717.35100.640249.91758.944.6534.461727.998.88142.72948250242.5116.59251189258.3241733.971732.414276.391548492.99823762.131619.48.710249787.62248.217781.8681909.4513.318464571897936.078703.5483.601965.75146.1158.571579624.8874.70113.076447260.802414824.55241141301994.001345962.361403220.971249454.751481042.428.938858.697481.2518716.571175.80188.81147.691239217.46236792403.1969.5653652273.83558.962550.842116.97358.906831.54126.7819.29741.02911.57161.51146000.381382055.3820.4372146.71711.415380.54302980367.2841.175192000.1638.375145.9551.9062454012.021252473.621273002.924001.53778.043638.429837.635126.5668250360477.281137811.751.32937630217.22130.95173.629.112861.2652801010733.942.9032755401536.8554079210.6133.9326226.31254.50743.52810.183252.6543014671908.99129.6424042652.802518.937125.054.15922.3424787.34315211259.3535.893126.342819591.820.85923.7431511361260.58152494.813842.25762423821.040.1664.2317702.47535615644.646.699344.31170.9517.4964.1306997000002.452786.5251610.2130862.388.718988.782833.77784.2220.8730151.31.18241.9531.547.3912.313.57518247.55.738614.35661.3933.270.52964372.014234.914721717.899.5395522.759541.218.1987.04134070.85.68888.06177.313.7245.221.048.0932001.22843523.51710.557526.95952943.739.2114.011583.58484102.7426.641.1992534456.32398.620.182182.5396.589148.750.8354983842.211581.69521.87119.65490244270.75327.09690.690.69233.5532.2883156281217639.68.0851588.6110904.6432.3330.9632102.533394.5715.371765.033627.782121.8268835.784289.212.593242169.975.4027185.08763363.571.779.2772.0489319.158.32190.231.530.94123.96798.0658139.06719.91199.832901.1246.58855.122.722846117.84318.40636428.97.8099.7857153115761654.77279.314.3161.241.16146.927.0651.56108.6942015.721.38709912058.455241054.322239.7226.70890.81911.143279.81193.626245.4719.819177.42394.3354.9346.2111.6588.452295.125.7213205877.82.6974381.918443375.59176973.24141.91347.839.629621.1140.8119.62656910711.4920.387449.031740.3239.8626158.42.0115467.4681.39943.65362.5371.96948.0647118.22183.88985.437913.431.5811451.7316288188.02203.6348.9134273551.98399.081601.736344.8274.731.2720578225295.7958.661918.1144.36170.763.59364866.51596.251890.450.85146046.482.570597187.5343180.177950491.931512.377.293.5230358.1171.1717.62100.540176.21761.744.734.521725.898.72142.60348310242.416.61241190334.2741733.471733.214261.231549710.05823948.191618.78.716849807.05248.117780.7681907.9713.318663811941869.469385.8389.51723.13145.0141.651723742.3850.26103.714482168.072367323.64242301336505.381920122.881585238.251253729.881373626.388.133738.372391.640766.438177.46171.53145.583193124.50864254000.3468.8624518221.3671.630641.865820.606248.520726.79149.2119.60938.70612.69161.51049912.881273333.8822.1491135.42781.40930.54295250368.4140.12197100.1640.903146.4551.7366415712.251317870.381288214.513967.36576.883139.010139.595725.2518252343680.411193659.251.36836548517.13137.07173.828.852990.051268967773.123.0252645001495.6314114680.6083.95246225.63254.30645.30910.1723342.5523135031895.32134.7523244550.816219.677324.964.30222.14257094297599559.4535.857125.312219566.620.59223.93811461566252.00354204.653972.2563801120.920.15763.1436697.8923581573347.503544.55170.0317.4164.13306969000002.435326.4551080.3132268.728.630190.158533.26364.221.4229377.91.15235.8032.357.2812.413.5506124.095.772914.245687.2333.220.53764314.454253.174656197.969.482519.6558549.428.2286.43434301.15.75868.09173.913.7275.171.048.0931560.12839733.52110.4342520.94942952.338.53112.441586.86486702.7256.61.1822834933.32410.160.184185.4895.0939128.150.4854445942.671591.75821.54118.3649764.24291.98326.58680.690.7235.0932.2818156820117585.48.021583.6111004.6232.3830.97102.733413.8715.382564.989427.504123.3368707.494303.622.60295169.985.4023185.10373402.111.759.2682.052669.218.35189.3231.680.94123.95478.0666140.47717.7200.432572.9245854.212.717116118.79518.23476368.987.759.7544153254661396.93278.214.3760.691.16145.677.0252.00108.8342016.881.39274612158.385200128.822256.426.86490.83811.112179.41195.07245.2269.809176.12388.5155.3146.08111.6689.082285.435.7313236877.442.6874491.923352376.26178115.64141.1346.349.583721.2440.83118.9011.5420.263949.330540.0839.9526199.12.010567.3681.02143.375388.2373.23628.0373118.5183.69245.443713.451.5731447.2317863187.76203.6347.2194254389.29397.306604.276357.3175.041.2674448192747.7258.531921.0944.53170.113.60705866.11598.451737.260.84945978.842.575363188.0643303.977940492.051515.177.393.330286.7170.7716.14100.740255.31762.244.6234.51728.698.77142.50148320242.7516.58931188758.8441779.981734.314271.881548182.16823174.051620.28.712449784.19248.217782.2181907.3913.3111024391906264.579379.7372.12856.89162.6552.991689762.3849.7382.83603641.632600519.602510213036451225640.514794651422030.121690102.759.33678.44040.9599426.439168.56201.29169.96OpenBenchmarking.org

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Random Fill Syncabc7001400210028003500SE +/- 15.17, N = 31351123931931. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Random Fill Syncabc6001200180024003000Min: 1335 / Avg: 1350.67 / Max: 13811. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latencyabc50100150200250SE +/- 3.52, N = 9188.71217.46124.511. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latencyabc4080120160200Min: 167.29 / Avg: 188.71 / Max: 201.211. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Writeabc14002800420056007000SE +/- 83.04, N = 94252367964251. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Writeabc11002200330044005500Min: 3975.91 / Avg: 4251.74 / Max: 4782.141. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUabc9001800270036004500SE +/- 4.94, N = 32437.962403.194000.34MIN: 2279.1MIN: 2251.07MIN: 3992.831. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUabc7001400210028003500Min: 2429.83 / Avg: 2437.96 / Max: 2446.881. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4Kabc20406080100SE +/- 0.17, N = 388.4369.5768.861. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4Kabc20406080100Min: 88.24 / Avg: 88.43 / Max: 88.761. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Writeabc10002000300040005000SE +/- 59.66, N = 124070365245181. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Writeabc8001600240032004000Min: 3636.91 / Avg: 4069.76 / Max: 4346.481. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latencyabc60120180240300SE +/- 3.72, N = 12246.31273.84221.361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latencyabc50100150200250Min: 230.07 / Avg: 246.31 / Max: 274.961. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Streamabc1632486480SE +/- 0.07, N = 370.8358.9671.63
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Streamabc1428425670Min: 70.7 / Avg: 70.83 / Max: 70.93

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Streamabc1122334455SE +/- 0.04, N = 342.3450.8441.87
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Streamabc1020304050Min: 42.28 / Avg: 42.34 / Max: 42.42

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Streamabc510152025SE +/- 0.06, N = 320.1416.9720.61
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Streamabc510152025Min: 20.02 / Avg: 20.14 / Max: 20.21

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Streamabc1326395265SE +/- 0.15, N = 349.6458.9148.52
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Streamabc1224364860Min: 49.48 / Avg: 49.64 / Max: 49.95

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUabc714212835SE +/- 0.37, N = 1427.9731.5426.79MIN: 19.8 / MAX: 50.33MIN: 19.58 / MAX: 49.19MIN: 21.9 / MAX: 44.221. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUabc714212835Min: 26.4 / Avg: 27.97 / Max: 31.471. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUabc306090120150SE +/- 1.79, N = 14143.21126.78149.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUabc306090120150Min: 127.03 / Avg: 143.21 / Max: 151.441. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

FLAC Audio Encoding

This test times how long it takes to encode a sample WAV file to FLAC audio format ten times using the --best preset settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACabc510152025SE +/- 0.15, N = 2517.3219.3019.611. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACabc510152025Min: 16.55 / Avg: 17.32 / Max: 19.251. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 1.5 kbpsabc918273645SE +/- 0.20, N = 336.9241.0338.71
OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 1.5 kbpsabc918273645Min: 36.69 / Avg: 36.91 / Max: 37.32

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 0 - Input: Bosphorus 1080pabc3691215SE +/- 0.13, N = 312.1811.5712.691. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 0 - Input: Bosphorus 1080pabc48121620Min: 12.03 / Avg: 12.18 / Max: 12.451. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 128abc4080120160200SE +/- 1.32, N = 11176.4161.5161.5
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 128abc306090120150Min: 164.3 / Avg: 176.45 / Max: 181.8

KeyDB

A benchmark of KeyDB as an open-source, multi-threaded fork of the Redis server now developed by Snapchat. The developers self-describe the KeyDB in-memory data structure store as a "faster drop in alternative to Redis". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SET - Parallel Connections: 100abc200K400K600K800K1000KSE +/- 12223.15, N = 41069422.991146000.381049912.881. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SET - Parallel Connections: 100abc200K400K600K800K1000KMin: 1041362.94 / Avg: 1069422.99 / Max: 1098490.751. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: GET - Parallel Connections: 100abc300K600K900K1200K1500KSE +/- 15206.41, N = 31283175.581382055.381273333.881. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: GET - Parallel Connections: 100abc200K400K600K800K1000KMin: 1263487.62 / Avg: 1283175.58 / Max: 1313094.251. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streamabc510152025SE +/- 0.30, N = 322.0220.4422.15
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streamabc510152025Min: 21.41 / Avg: 22.02 / Max: 22.36

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streamabc306090120150SE +/- 1.91, N = 3136.27146.72135.43
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streamabc306090120150Min: 134.17 / Avg: 136.27 / Max: 140.08

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUabc0.34350.6871.03051.3741.7175SE +/- 0.01138, N = 151.526651.415381.40930MIN: 1.39MIN: 1.39MIN: 1.391. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUabc246810Min: 1.41 / Avg: 1.53 / Max: 1.561. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pabc0.12150.2430.36450.4860.6075SE +/- 0.00, N = 150.500.540.541. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080pabc246810Min: 0.48 / Avg: 0.5 / Max: 0.541. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Blake-2 Sabc60K120K180K240K300KSE +/- 791.71, N = 32822073029802952501. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Blake-2 Sabc50K100K150K200K250KMin: 281400 / Avg: 282206.67 / Max: 2837901. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xabc80160240320400SE +/- 0.59, N = 3344.30367.28368.411. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xabc70140210280350Min: 343.24 / Avg: 344.3 / Max: 345.271. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 6 kbpsabc918273645SE +/- 0.47, N = 338.5241.1840.12
OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 6 kbpsabc918273645Min: 37.98 / Avg: 38.52 / Max: 39.46

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY Creditsabc4K8K12K16K20KSE +/- 186.26, N = 51843819200197101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY Creditsabc3K6K9K12K15KMin: 18010 / Avg: 18438 / Max: 191201. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4Kabc0.0360.0720.1080.1440.18SE +/- 0.00, N = 70.150.160.161. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4Kabc12345Min: 0.15 / Avg: 0.15 / Max: 0.161. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 3 kbpsabc918273645SE +/- 0.60, N = 1538.6638.3840.90
OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 3 kbpsabc918273645Min: 36.94 / Avg: 38.66 / Max: 45.67

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On Demandabc306090120150SE +/- 2.11, N = 3155.27145.95146.451. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On Demandabc306090120150Min: 152.55 / Avg: 155.27 / Max: 159.431. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On Demandabc1224364860SE +/- 0.66, N = 348.8051.9051.731. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On Demandabc1020304050Min: 47.51 / Avg: 48.8 / Max: 49.651. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Update Randomabc140K280K420K560K700KSE +/- 6561.15, N = 56299446245406641571. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Update Randomabc120K240K360K480K600KMin: 619005 / Avg: 629944 / Max: 6512301. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 5 - Input: Bosphorus 4Kabc3691215SE +/- 0.07, N = 312.7312.0212.251. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 5 - Input: Bosphorus 4Kabc48121620Min: 12.58 / Avg: 12.73 / Max: 12.821. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

KeyDB

A benchmark of KeyDB as an open-source, multi-threaded fork of the Redis server now developed by Snapchat. The developers self-describe the KeyDB in-memory data structure store as a "faster drop in alternative to Redis". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPUSH - Parallel Connections: 50abc300K600K900K1200K1500KSE +/- 14149.76, N = 41323318.971252473.621317870.381. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPUSH - Parallel Connections: 50abc200K400K600K800K1000KMin: 1289224.62 / Avg: 1323318.97 / Max: 1357625.881. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Glibc C String Functionsabc300K600K900K1200K1500KSE +/- 9938.23, N = 91220179.931273002.921288214.511. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Glibc C String Functionsabc200K400K600K800K1000KMin: 1189467.16 / Avg: 1220179.93 / Max: 1292156.431. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Write - Average Latencyabc9001800270036004500SE +/- 35.63, N = 33797.314001.543967.371. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Write - Average Latencyabc7001400210028003500Min: 3726.37 / Avg: 3797.31 / Max: 3838.51. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamabc20406080100SE +/- 0.23, N = 374.1078.0476.88
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamabc1530456075Min: 73.78 / Avg: 74.1 / Max: 74.55

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamabc918273645SE +/- 0.13, N = 340.4738.4339.01
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamabc816243240Min: 40.21 / Avg: 40.47 / Max: 40.65

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamabc918273645SE +/- 0.34, N = 338.3437.6439.60
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamabc816243240Min: 37.84 / Avg: 38.34 / Max: 38.98

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamabc612182430SE +/- 0.23, N = 326.0926.5725.25
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Streamabc612182430Min: 25.65 / Avg: 26.09 / Max: 26.42

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Writeabc60120180240300SE +/- 2.49, N = 32632502521. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Writeabc50100150200250Min: 260.52 / Avg: 263.39 / Max: 268.361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

KeyDB

A benchmark of KeyDB as an open-source, multi-threaded fork of the Redis server now developed by Snapchat. The developers self-describe the KeyDB in-memory data structure store as a "faster drop in alternative to Redis". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: HMSET - Parallel Connections: 100abc80K160K240K320K400KSE +/- 3780.33, N = 12361525.21360477.28343680.411. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: HMSET - Parallel Connections: 100abc60K120K180K240K300KMin: 346353.94 / Avg: 361525.21 / Max: 390301.781. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SADD - Parallel Connections: 100abc300K600K900K1200K1500KSE +/- 8944.31, N = 31148270.291137811.751193659.251. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SADD - Parallel Connections: 100abc200K400K600K800K1000KMin: 1138952.12 / Avg: 1148270.29 / Max: 1166153.621. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Only - Average Latencyabc0.31340.62680.94021.25361.567SE +/- 0.016, N = 41.3931.3291.3681. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Only - Average Latencyabc246810Min: 1.36 / Avg: 1.39 / Max: 1.431. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Onlyabc80K160K240K320K400KSE +/- 3941.00, N = 43591653763023654851. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Onlyabc70K140K210K280K350KMin: 349031.88 / Avg: 359165.38 / Max: 366991.61. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 4K - Video Preset: Very Fastabc48121620SE +/- 0.12, N = 316.4517.2217.131. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Very Fastabc48121620Min: 16.28 / Avg: 16.45 / Max: 16.671. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third Runabc306090120150SE +/- 0.37, N = 9133.73130.95137.07MIN: 6.54 / MAX: 8571.43MIN: 7.08 / MAX: 6000MIN: 7.03 / MAX: 7500
OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third Runabc306090120150Min: 132.26 / Avg: 133.73 / Max: 135.39

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 256abc4080120160200SE +/- 0.00, N = 3181.6173.6173.8
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 256abc306090120150Min: 181.6 / Avg: 181.6 / Max: 181.6

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 4K - Video Preset: Ultra Fastabc714212835SE +/- 0.21, N = 327.8329.1128.851. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Ultra Fastabc612182430Min: 27.62 / Avg: 27.83 / Max: 28.251. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Write - Average Latencyabc6001200180024003000SE +/- 26.55, N = 122891.942861.272990.051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Write - Average Latencyabc5001000150020002500Min: 2685.17 / Avg: 2891.94 / Max: 2981.891. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Writeabc60120180240300SE +/- 2.63, N = 122772802681. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Writeabc50100150200250Min: 268.29 / Avg: 276.9 / Max: 297.931. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

KeyDB

A benchmark of KeyDB as an open-source, multi-threaded fork of the Redis server now developed by Snapchat. The developers self-describe the KeyDB in-memory data structure store as a "faster drop in alternative to Redis". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPUSH - Parallel Connections: 100abc200K400K600K800K1000KSE +/- 10077.27, N = 15976117.591010733.94967773.121. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPUSH - Parallel Connections: 100abc200K400K600K800K1000KMin: 929609.94 / Avg: 976117.59 / Max: 1056881.381. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latencyabc0.6821.3642.0462.7283.41SE +/- 0.027, N = 33.0312.9033.0251. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latencyabc246810Min: 2.99 / Avg: 3.03 / Max: 3.081. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Onlyabc60K120K180K240K300KSE +/- 2360.08, N = 32639502755402645001. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Onlyabc50K100K150K200K250KMin: 259818.92 / Avg: 263950.43 / Max: 267993.041. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Write - Average Latencyabc30060090012001500SE +/- 15.49, N = 121561.271536.861495.631. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Write - Average Latencyabc30060090012001500Min: 1468.9 / Avg: 1561.27 / Max: 1644.691. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Onlyabc90K180K270K360K450KSE +/- 2783.61, N = 33945264079214114681. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Onlyabc70K140K210K280K350KMin: 389543.06 / Avg: 394526.4 / Max: 399167.541. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latencyabc0.14270.28540.42810.57080.7135SE +/- 0.005, N = 30.6340.6130.6081. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latencyabc246810Min: 0.63 / Avg: 0.63 / Max: 0.641. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUabc0.92221.84442.76663.68884.611SE +/- 0.02631, N = 154.098533.932603.95246MIN: 3.8MIN: 3.81MIN: 3.821. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUabc246810Min: 3.96 / Avg: 4.1 / Max: 4.311. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

JPEG XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: Allabc50100150200250SE +/- 0.07, N = 3217.33226.31225.63
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: Allabc4080120160200Min: 217.2 / Avg: 217.33 / Max: 217.43

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1abc60120180240300SE +/- 2.17, N = 7244.49254.51254.311. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1abc50100150200250Min: 231.81 / Avg: 244.49 / Max: 249.281. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

EnCodec

EnCodec is a Facebook/Meta developed AI means of compressing audio files using High Fidelity Neural Audio Compression. EnCodec is designed to provide codec compression at 6 kbps using their novel AI-powered compression technique. The test profile uses a lengthy JFK speech as the audio input for benchmarking and the performance measurement is measuring the time to encode the EnCodec file from WAV. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 24 kbpsabc1020304050SE +/- 0.52, N = 1544.5643.5345.31
OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 24 kbpsabc918273645Min: 42.8 / Avg: 44.56 / Max: 49

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1abc3691215SE +/- 0.087, N = 79.78010.18010.1721. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1abc3691215Min: 9.27 / Avg: 9.78 / Max: 9.971. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Writeabc70140210280350SE +/- 3.20, N = 123213253341. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 500 - Mode: Read Writeabc60120180240300Min: 304.01 / Avg: 320.6 / Max: 340.391. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Only - Average Latencyabc0.59721.19441.79162.38882.986SE +/- 0.012, N = 32.6062.6542.5521. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Only - Average Latencyabc246810Min: 2.59 / Avg: 2.61 / Max: 2.631. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Onlyabc70K140K210K280K350KSE +/- 1379.07, N = 33070183014673135031. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 800 - Mode: Read Onlyabc50K100K150K200K250KMin: 304598.61 / Avg: 307017.54 / Max: 309374.631. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Garlicoinabc400800120016002000SE +/- 21.45, N = 41836.591908.991895.321. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Garlicoinabc30060090012001500Min: 1789.19 / Avg: 1836.59 / Max: 1887.71. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second Runabc306090120150SE +/- 0.66, N = 9130.84129.64134.75MIN: 6.51 / MAX: 8571.43MIN: 7.07 / MAX: 6000MIN: 7.04 / MAX: 8571.43
OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second Runabc306090120150Min: 127.89 / Avg: 130.84 / Max: 133.42

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Onlyabc50K100K150K200K250KSE +/- 1961.46, N = 32415402404262324451. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Onlyabc40K80K120K160K200KMin: 237993.43 / Avg: 241540.44 / Max: 244765.171. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Streamabc1224364860SE +/- 0.32, N = 351.1552.8050.82
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Streamabc1122334455Min: 50.79 / Avg: 51.15 / Max: 51.78

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Streamabc510152025SE +/- 0.12, N = 319.5518.9419.68
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Streamabc510152025Min: 19.31 / Avg: 19.55 / Max: 19.69

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Mediumabc612182430SE +/- 0.14, N = 324.1125.0524.96
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Mediumabc612182430Min: 23.93 / Avg: 24.11 / Max: 24.38

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latencyabc0.9681.9362.9043.8724.84SE +/- 0.034, N = 34.1414.1594.3021. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latencyabc246810Min: 4.09 / Avg: 4.14 / Max: 4.21. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Slowabc510152025SE +/- 0.02, N = 321.5322.3422.14
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Slowabc510152025Min: 21.51 / Avg: 21.53 / Max: 21.56

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 512abc6K12K18K24K30KSE +/- 297.75, N = 325625.024787.325709.0
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 512abc4K8K12K16K20KMin: 25327.1 / Avg: 25625 / Max: 26220.5

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Random Readabc9M18M27M36M45MSE +/- 414703.47, N = 64161411043152112429759951. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Random Readabc7M14M21M28M35MMin: 40937598 / Avg: 41614110.33 / Max: 435619061. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Timed FFmpeg Compilation

This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 6.0Time To Compileabc1428425670SE +/- 0.68, N = 361.4759.3559.45
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 6.0Time To Compileabc1224364860Min: 60.19 / Avg: 61.47 / Max: 62.53

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under The Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 4K - Video Preset: Fasterabc246810SE +/- 0.017, N = 36.0655.8935.8571. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 4K - Video Preset: Fasterabc246810Min: 6.05 / Avg: 6.07 / Max: 6.11. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streamabc306090120150SE +/- 1.18, N = 12129.74126.34125.31
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streamabc20406080100Min: 126.78 / Avg: 129.74 / Max: 142

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 128abc4K8K12K16K20KSE +/- 104.08, N = 320255.719591.819566.6
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 128abc4K8K12K16K20KMin: 20126.8 / Avg: 20255.7 / Max: 20461.7

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under The Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: Fasterabc510152025SE +/- 0.07, N = 320.1620.8620.591. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: Fasterabc510152025Min: 20.01 / Avg: 20.15 / Max: 20.241. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streamabc612182430SE +/- 0.20, N = 1223.1423.7423.94
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streamabc612182430Min: 21.11 / Avg: 23.14 / Max: 23.66

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Sequential Fillabc300K600K900K1200K1500KSE +/- 3413.48, N = 31502792151136114615661. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Sequential Fillabc300K600K900K1200K1500KMin: 1496121 / Avg: 1502792 / Max: 15073841. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Write - Average Latencyabc60120180240300SE +/- 1.27, N = 3258.43260.58252.001. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Write - Average Latencyabc50100150200250Min: 257.04 / Avg: 258.43 / Max: 260.971. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Forkingabc12K24K36K48K60KSE +/- 302.19, N = 354276.3452494.8154204.651. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Forkingabc9K18K27K36K45KMin: 53863.04 / Avg: 54276.34 / Max: 54864.891. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Writeabc90180270360450SE +/- 1.90, N = 33873843971. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Writeabc70140210280350Min: 383.19 / Avg: 386.98 / Max: 389.051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

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 1.4Encoder Mode: Preset 4 - Input: Bosphorus 4Kabc0.50781.01561.52342.03122.539SE +/- 0.007, N = 32.1852.2572.2501. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 4 - Input: Bosphorus 4Kabc246810Min: 2.17 / Avg: 2.18 / Max: 2.21. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Onlyabc140K280K420K560K700KSE +/- 4488.67, N = 36177066242386380111. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Onlyabc110K220K330K440K550KMin: 610548.36 / Avg: 617705.62 / Max: 625977.291. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 4K - Video Preset: Super Fastabc510152025SE +/- 0.11, N = 320.3821.0420.921. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Super Fastabc510152025Min: 20.18 / Avg: 20.38 / Max: 20.561. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average Latencyabc0.03650.0730.10950.1460.1825SE +/- 0.001, N = 30.1620.1600.1571. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average Latencyabc12345Min: 0.16 / Avg: 0.16 / Max: 0.161. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamabc1530456075SE +/- 0.58, N = 1265.1264.2363.14
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamabc1326395265Min: 64.12 / Avg: 65.12 / Max: 71.45

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Write - Average Latencyabc150300450600750SE +/- 4.83, N = 3681.42702.48697.891. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Write - Average Latencyabc120240360480600Min: 672.25 / Avg: 681.42 / Max: 688.621. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Writeabc80160240320400SE +/- 2.61, N = 33673563581. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Writeabc70140210280350Min: 363.05 / Avg: 366.92 / Max: 371.891. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 128abc3K6K9K12K15KSE +/- 199.13, N = 316123.015644.615733.0
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 128abc3K6K9K12K15KMin: 15725.3 / Avg: 16123 / Max: 16340.1

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamabc1122334455SE +/- 0.38, N = 1246.1046.7047.50
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamabc1020304050Min: 41.98 / Avg: 46.1 / Max: 46.78

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demandabc1020304050SE +/- 0.04, N = 343.2544.3144.551. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demandabc918273645Min: 43.21 / Avg: 43.25 / Max: 43.331. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demandabc4080120160200SE +/- 0.16, N = 3175.13170.95170.031. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demandabc306090120150Min: 174.82 / Avg: 175.13 / Max: 175.311. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Ultra Fastabc48121620SE +/- 0.04, N = 316.9917.4917.41
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Ultra Fastabc48121620Min: 16.92 / Avg: 16.99 / Max: 17.06

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Very Fastabc1428425670SE +/- 0.17, N = 362.3364.1064.13
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Very Fastabc1326395265Min: 62.01 / Avg: 62.33 / Max: 62.6

nekRS

nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFLOP/s, More Is BetternekRS 22.0Input: TurboPipe Periodicabc7000M14000M21000M28000M35000MSE +/- 533119138.01, N = 92984465555630699700000306969000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -lmpi_cxx -lmpi
OpenBenchmarking.orgFLOP/s, More Is BetternekRS 22.0Input: TurboPipe Periodicabc5000M10000M15000M20000M25000MMin: 25924900000 / Avg: 29844655555.56 / Max: 306764000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -lmpi_cxx -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUabc0.56351.1271.69052.2542.8175SE +/- 0.02403, N = 152.504622.452782.43532MIN: 2.34MIN: 2.41MIN: 2.391. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUabc246810Min: 2.4 / Avg: 2.5 / Max: 2.681. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 4K - Video Preset: Slowabc246810SE +/- 0.03, N = 36.346.526.451. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Slowabc3691215Min: 6.31 / Avg: 6.34 / Max: 6.391. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 128abc11K22K33K44K55KSE +/- 345.86, N = 352512.651610.251080.3
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 128abc9K18K27K36K45KMin: 51862.1 / Avg: 52512.57 / Max: 53041.6

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Pthreadabc30K60K90K120K150KSE +/- 1404.99, N = 4128698.01130862.38132268.721. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Pthreadabc20K40K60K80K100KMin: 125530.65 / Avg: 128698.01 / Max: 132355.941. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.0Binary: Pathtracer - Model: Crownabc246810SE +/- 0.0433, N = 38.48758.71898.6301MIN: 8.32 / MAX: 8.82MIN: 8.66 / MAX: 8.9MIN: 8.56 / MAX: 8.78
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0Binary: Pathtracer - Model: Crownabc3691215Min: 8.44 / Avg: 8.49 / Max: 8.57

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamabc20406080100SE +/- 0.97, N = 391.1988.7890.16
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamabc20406080100Min: 89.33 / Avg: 91.19 / Max: 92.58

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamabc816243240SE +/- 0.35, N = 332.8933.7833.26
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamabc714212835Min: 32.39 / Avg: 32.89 / Max: 33.57

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Slowabc0.94951.8992.84853.7984.7475SE +/- 0.05, N = 34.114.224.20
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Slowabc246810Min: 4.06 / Avg: 4.11 / Max: 4.2

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Liveabc510152025SE +/- 0.01, N = 321.0520.8721.421. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Liveabc510152025Min: 21.03 / Avg: 21.05 / Max: 21.081. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 256abc6K12K18K24K30KSE +/- 375.67, N = 330043.830151.329377.9
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 256abc5K10K15K20K25KMin: 29415.9 / Avg: 30043.83 / Max: 30715.1

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUabc0.26550.5310.79651.0621.3275SE +/- 0.01, N = 31.171.181.151. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUabc246810Min: 1.16 / Avg: 1.17 / Max: 1.181. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Liveabc50100150200250SE +/- 0.14, N = 3239.86241.95235.801. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Liveabc4080120160200Min: 239.6 / Avg: 239.86 / Max: 240.091. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pabc816243240SE +/- 0.42, N = 331.6931.5432.351. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080pabc714212835Min: 31.17 / Avg: 31.69 / Max: 32.531. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUabc246810SE +/- 0.04, N = 37.217.397.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUabc3691215Min: 7.16 / Avg: 7.21 / Max: 7.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression Speedabc3691215SE +/- 0.10, N = 1512.612.312.41. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Compression Speedabc48121620Min: 12.1 / Avg: 12.57 / Max: 13.41. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very Fastabc3691215SE +/- 0.08, N = 313.2513.5713.50
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Very Fastabc48121620Min: 13.16 / Avg: 13.25 / Max: 13.4

KeyDB

A benchmark of KeyDB as an open-source, multi-threaded fork of the Redis server now developed by Snapchat. The developers self-describe the KeyDB in-memory data structure store as a "faster drop in alternative to Redis". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: HMSET - Parallel Connections: 50abc110K220K330K440K550KSE +/- 2335.08, N = 3511218.21518247.50506124.091. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: HMSET - Parallel Connections: 50abc90K180K270K360K450KMin: 507737.94 / Avg: 511218.21 / Max: 515655.311. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamabc1.29892.59783.89675.19566.4945SE +/- 0.0204, N = 35.64015.73865.7729
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamabc246810Min: 5.61 / Avg: 5.64 / Max: 5.68

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Super Fastabc48121620SE +/- 0.08, N = 313.9814.3014.24
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Super Fastabc48121620Min: 13.87 / Avg: 13.98 / Max: 14.13

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUabc12002400360048006000SE +/- 27.53, N = 35560.275661.395687.231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUabc10002000300040005000Min: 5525.69 / Avg: 5560.27 / Max: 5614.671. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 1080p - Video Preset: Slowabc816243240SE +/- 0.14, N = 332.5333.2733.221. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 1080p - Video Preset: Slowabc714212835Min: 32.38 / Avg: 32.53 / Max: 32.81. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Only - Average Latencyabc0.12170.24340.36510.48680.6085SE +/- 0.008, N = 30.5410.5290.5371. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Only - Average Latencyabc246810Min: 0.53 / Avg: 0.54 / Max: 0.561. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 20abc14K28K42K56K70KSE +/- 804.71, N = 362961.4564372.0164314.451. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 20abc11K22K33K44K55KMin: 62072.31 / Avg: 62961.45 / Max: 64567.811. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUabc9001800270036004500SE +/- 15.35, N = 34329.654234.914253.17MIN: 4088.59MIN: 4027.35MIN: 4043.251. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUabc8001600240032004000Min: 4301.53 / Avg: 4329.65 / Max: 4354.371. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Onlyabc100K200K300K400K500KSE +/- 6603.78, N = 34620714721714656191. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 250 - Mode: Read Onlyabc80K160K240K320K400KMin: 449544.53 / Avg: 462071.31 / Max: 471959.541. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 90abc246810SE +/- 0.01, N = 37.797.897.961. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 90abc3691215Min: 7.78 / Avg: 7.79 / Max: 7.81. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.0Binary: Pathtracer ISPC - Model: Crownabc3691215SE +/- 0.0598, N = 39.33669.53959.4820MIN: 9.13 / MAX: 9.88MIN: 9.33 / MAX: 9.85MIN: 9.26 / MAX: 9.81
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0Binary: Pathtracer ISPC - Model: Crownabc3691215Min: 9.27 / Avg: 9.34 / Max: 9.46

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamabc110220330440550SE +/- 1.89, N = 3530.84522.76519.66
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamabc90180270360450Min: 528.49 / Avg: 530.84 / Max: 534.57

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUabc120240360480600SE +/- 2.23, N = 3552.79541.21549.42MIN: 501.54 / MAX: 634.83MIN: 503.72 / MAX: 633.37MIN: 502.92 / MAX: 645.961. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUabc100200300400500Min: 548.34 / Avg: 552.79 / Max: 555.311. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 80abc246810SE +/- 0.00, N = 38.058.198.221. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 80abc3691215Min: 8.05 / Avg: 8.05 / Max: 8.061. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

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 0.11Encoder Speed: 2abc20406080100SE +/- 0.11, N = 388.2587.0486.431. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 2abc20406080100Min: 88.02 / Avg: 88.25 / Max: 88.391. (CXX) g++ options: -O3 -fPIC -lm

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 256abc7K14K21K28K35KSE +/- 405.43, N = 334782.834070.834301.1
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 256abc6K12K18K24K30KMin: 34151.6 / Avg: 34782.8 / Max: 35539.2

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamabc1.29572.59143.88715.18286.4785SE +/- 0.0208, N = 35.64345.68885.7586
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamabc246810Min: 5.6 / Avg: 5.64 / Max: 5.68

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 80abc246810SE +/- 0.01, N = 37.938.068.091. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 80abc3691215Min: 7.91 / Avg: 7.93 / Max: 7.951. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 512abc4080120160200SE +/- 0.00, N = 3177.4177.3173.9
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 512abc306090120150Min: 177.4 / Avg: 177.4 / Max: 177.4

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 0.11Encoder Speed: 6, Losslessabc48121620SE +/- 0.14, N = 313.4613.7213.731. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, Losslessabc48121620Min: 13.32 / Avg: 13.46 / Max: 13.741. (CXX) g++ options: -O3 -fPIC -lm

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4Kabc1.17452.3493.52354.6985.8725SE +/- 0.00, N = 35.125.225.171. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4Kabc246810Min: 5.12 / Avg: 5.12 / Max: 5.121. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUabc0.23850.4770.71550.9541.1925SE +/- 0.01, N = 31.061.041.04MIN: 0.92 / MAX: 68.14MIN: 0.92 / MAX: 7.17MIN: 0.92 / MAX: 4.881. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUabc246810Min: 1.05 / Avg: 1.06 / Max: 1.071. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 90abc246810SE +/- 0.01, N = 37.948.098.091. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 90abc3691215Min: 7.93 / Avg: 7.94 / Max: 7.951. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 1024abc7K14K21K28K35KSE +/- 47.74, N = 332151.832001.231560.1
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 1024abc6K12K18K24K30KMin: 32061.4 / Avg: 32151.83 / Max: 32223.6

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Onlyabc60K120K180K240K300KSE +/- 2823.16, N = 32892052843522839731. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Onlyabc50K100K150K200K250KMin: 284517.95 / Avg: 289205.4 / Max: 294275.231. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Only - Average Latencyabc0.79221.58442.37663.16883.961SE +/- 0.034, N = 33.4583.5173.5211. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1000 - Mode: Read Only - Average Latencyabc246810Min: 3.4 / Avg: 3.46 / Max: 3.521. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUabc3691215SE +/- 0.00, N = 310.3710.5610.43MIN: 10.1MIN: 10.29MIN: 10.171. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUabc3691215Min: 10.37 / Avg: 10.37 / Max: 10.381. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamabc110220330440550SE +/- 1.42, N = 3530.18526.96520.95
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamabc90180270360450Min: 527.43 / Avg: 530.18 / Max: 532.19

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1Mabc6001200180024003000SE +/- 10.62, N = 32901.82943.72952.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1Mabc5001000150020002500Min: 2888 / Avg: 2901.83 / Max: 2922.71. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: GoogLeNetabc918273645SE +/- 0.03, N = 338.8939.2038.53
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: GoogLeNetabc816243240Min: 38.86 / Avg: 38.89 / Max: 38.95

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold Cacheabc306090120150SE +/- 1.04, N = 9112.08114.01112.44MIN: 6.36 / MAX: 6666.67MIN: 6.89 / MAX: 5000MIN: 6.91 / MAX: 5454.55
OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold Cacheabc20406080100Min: 107.25 / Avg: 112.08 / Max: 117.76

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Ringcoinabc30060090012001500SE +/- 14.19, N = 31560.091583.581586.861. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Ringcoinabc30060090012001500Min: 1531.71 / Avg: 1560.09 / Max: 1574.911. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Skeincoinabc10K20K30K40K50KSE +/- 360.94, N = 34785348410486701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Skeincoinabc8K16K24K32K40KMin: 47420 / Avg: 47853.33 / Max: 485701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under The Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 4K - Video Preset: Fastabc0.6171.2341.8512.4683.085SE +/- 0.016, N = 32.6962.7422.7251. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 4K - Video Preset: Fastabc246810Min: 2.67 / Avg: 2.7 / Max: 2.721. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 4K - Video Preset: Mediumabc246810SE +/- 0.02, N = 36.536.646.601. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 4K - Video Preset: Mediumabc3691215Min: 6.51 / Avg: 6.53 / Max: 6.571. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUabc0.27050.5410.81151.0821.3525SE +/- 0.00770, N = 31.202191.199251.18228MIN: 1.13MIN: 1.13MIN: 1.121. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUabc246810Min: 1.19 / Avg: 1.2 / Max: 1.221. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 512abc8K16K24K32K40KSE +/- 342.69, N = 335029.434456.334933.3
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 512abc6K12K18K24K30KMin: 34448.8 / Avg: 35029.37 / Max: 35635.1

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUabc5001000150020002500SE +/- 3.98, N = 32438.212398.622410.16MIN: 2279.8MIN: 2251.99MIN: 2254.611. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUabc400800120016002000Min: 2430.27 / Avg: 2438.21 / Max: 2442.741. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latencyabc0.04160.08320.12480.16640.208SE +/- 0.001, N = 30.1850.1820.1841. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latencyabc12345Min: 0.18 / Avg: 0.18 / Max: 0.191. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPUabc4080120160200SE +/- 0.73, N = 3184.37182.53185.48MIN: 164.97 / MAX: 226.63MIN: 165.34 / MAX: 243.94MIN: 166.33 / MAX: 257.371. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPUabc306090120150Min: 182.95 / Avg: 184.37 / Max: 185.351. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

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 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4Kabc20406080100SE +/- 0.27, N = 395.5796.5895.091. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4Kabc20406080100Min: 95.24 / Avg: 95.57 / Max: 96.111. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 128abc2K4K6K8K10KSE +/- 11.01, N = 39270.59148.79128.1
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 128abc16003200480064008000Min: 9248.5 / Avg: 9270.5 / Max: 9282.3

JPEG XL Decoding libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 1abc1122334455SE +/- 0.34, N = 350.0550.8350.48
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 1abc1020304050Min: 49.71 / Avg: 50.05 / Max: 50.74

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Onlyabc120K240K360K480K600KSE +/- 1426.41, N = 35414105498385444591. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Onlyabc100K200K300K400K500KMin: 538641.13 / Avg: 541409.69 / Max: 5433901. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Liveabc1020304050SE +/- 0.15, N = 342.0242.2142.671. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Liveabc918273645Min: 41.77 / Avg: 42.02 / Max: 42.291. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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 or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigabc30060090012001500SE +/- 2.13, N = 31605.971581.701591.76
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigabc30060090012001500Min: 1602.93 / Avg: 1605.97 / Max: 1610.07

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPUabc510152025SE +/- 0.09, N = 321.6821.8721.541. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPUabc510152025Min: 21.56 / Avg: 21.68 / Max: 21.861. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Liveabc306090120150SE +/- 0.43, N = 3120.17119.65118.361. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Liveabc20406080100Min: 119.42 / Avg: 120.17 / Max: 120.91. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 256abc11K22K33K44K55KSE +/- 168.75, N = 349613.549024.049764.2
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 256abc9K18K27K36K45KMin: 49286.3 / Avg: 49613.47 / Max: 49848.8

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUabc9001800270036004500SE +/- 5.86, N = 34334.974270.754291.98MIN: 4122.67MIN: 4043.43MIN: 4067.511. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUabc8001600240032004000Min: 4323.57 / Avg: 4334.97 / Max: 4343.021. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Magiabc70140210280350SE +/- 2.70, N = 3331.49327.09326.581. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Magiabc60120180240300Min: 327.08 / Avg: 331.49 / Max: 336.381. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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 1.3.1Benchmark: vklBenchmark Scalarabc1530456075SE +/- 0.33, N = 3686968MIN: 7 / MAX: 1212MIN: 7 / MAX: 1201MIN: 7 / MAX: 1202
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark Scalarabc1326395265Min: 68 / Avg: 68.33 / Max: 69

JPEG XL libjxl

The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 100abc0.15530.31060.46590.62120.7765SE +/- 0.00, N = 30.680.690.691. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 100abc246810Min: 0.68 / Avg: 0.68 / Max: 0.691. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 100abc0.15750.3150.47250.630.7875SE +/- 0.00, N = 30.690.690.701. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic
OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 100abc246810Min: 0.69 / Avg: 0.69 / Max: 0.71. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -latomic

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal Installationabc50100150200250SE +/- 1.23, N = 3236.87233.55235.09
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal Installationabc4080120160200Min: 235.04 / Avg: 236.87 / Max: 239.21

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Streamabc816243240SE +/- 0.33, N = 1232.7332.2932.28
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Streamabc714212835Min: 32.15 / Avg: 32.73 / Max: 36.21

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Read Random Write Randomabc300K600K900K1200K1500KSE +/- 6061.26, N = 31546655156281215682011. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Read Random Write Randomabc300K600K900K1200K1500KMin: 1539339 / Avg: 1546655 / Max: 15586841. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 256abc4K8K12K16K20KSE +/- 126.23, N = 317827.617639.617585.4
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 256abc3K6K9K12K15KMin: 17656.5 / Avg: 17827.57 / Max: 18073.9

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under The Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: Fastabc246810SE +/- 0.014, N = 37.9778.0858.0201. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto
OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: Fastabc3691215Min: 7.95 / Avg: 7.98 / Max: 7.991. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Compression Speedabc30060090012001500SE +/- 1.71, N = 31567.81588.61583.61. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Compression Speedabc30060090012001500Min: 1564.7 / Avg: 1567.83 / Max: 1570.61. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Myriad-Groestlabc2K4K6K8K10KSE +/- 68.39, N = 31123711090111001. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Myriad-Groestlabc2K4K6K8K10KMin: 11100 / Avg: 11236.67 / Max: 113101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Mediumabc1.0442.0883.1324.1765.22SE +/- 0.01, N = 34.584.644.62
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 4K - Video Preset: Mediumabc246810Min: 4.57 / Avg: 4.58 / Max: 4.59

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 5 - Input: Bosphorus 1080pabc816243240SE +/- 0.02, N = 332.7532.3332.381. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 5 - Input: Bosphorus 1080pabc714212835Min: 32.73 / Avg: 32.75 / Max: 32.81. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Streamabc714212835SE +/- 0.28, N = 1230.5730.9630.97
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Streamabc714212835Min: 27.61 / Avg: 30.57 / Max: 31.09

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Glibc Qsort Data Sortingabc20406080100SE +/- 0.60, N = 3101.42102.53102.731. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Glibc Qsort Data Sortingabc20406080100Min: 100.47 / Avg: 101.42 / Max: 102.531. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUabc7001400210028003500SE +/- 29.12, N = 33438.183394.573413.87MIN: 3203.29 / MAX: 4311.69MIN: 3178.76 / MAX: 3780.74MIN: 3197.21 / MAX: 4433.431. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUabc6001200180024003000Min: 3381.98 / Avg: 3438.18 / Max: 3479.531. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamabc48121620SE +/- 0.17, N = 315.5715.3715.38
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamabc48121620Min: 15.38 / Avg: 15.57 / Max: 15.91

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamabc1530456075SE +/- 0.70, N = 364.2365.0364.99
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Streamabc1326395265Min: 62.83 / Avg: 64.23 / Max: 64.98

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 1.4Encoder Mode: Preset 8 - Input: Bosphorus 4Kabc714212835SE +/- 0.22, N = 327.4427.7827.501. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 8 - Input: Bosphorus 4Kabc612182430Min: 27.18 / Avg: 27.44 / Max: 27.871. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 1080p - Video Preset: Ultra Fastabc306090120150SE +/- 0.14, N = 3123.26121.82123.331. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 1080p - Video Preset: Ultra Fastabc20406080100Min: 123.04 / Avg: 123.26 / Max: 123.531. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100abc15K30K45K60K75KSE +/- 335.89, N = 367993.5268835.7868707.491. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100abc12K24K36K48K60KMin: 67610.77 / Avg: 67993.52 / Max: 68663.021. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUabc9001800270036004500SE +/- 2.56, N = 34341.924289.214303.62MIN: 4124.68MIN: 4045.19MIN: 4079.691. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUabc8001600240032004000Min: 4337.19 / Avg: 4341.92 / Max: 4345.971. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 480000 - Buffer Size: 512abc0.59051.1811.77152.3622.9525SE +/- 0.002482, N = 32.6245672.5932422.6029501. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 480000 - Buffer Size: 512abc246810Min: 2.62 / Avg: 2.62 / Max: 2.631. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: BMW27 - Compute: CPU-Onlyabc4080120160200SE +/- 1.22, N = 3172.01169.97169.98
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: BMW27 - Compute: CPU-Onlyabc306090120150Min: 169.56 / Avg: 172.01 / Max: 173.26

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamabc1.22962.45923.68884.91846.148SE +/- 0.0157, N = 35.46505.40275.4023
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamabc246810Min: 5.44 / Avg: 5.46 / Max: 5.49

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamabc4080120160200SE +/- 0.52, N = 3182.98185.09185.10
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Streamabc306090120150Min: 182.1 / Avg: 182.98 / Max: 183.91

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUabc7001400210028003500SE +/- 16.56, N = 33384.143363.573402.11MIN: 3150.48 / MAX: 4797.77MIN: 3197.76 / MAX: 3701.88MIN: 3157.9 / MAX: 5910.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUabc6001200180024003000Min: 3352.44 / Avg: 3384.14 / Max: 3408.281. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUabc0.39830.79661.19491.59321.9915SE +/- 0.02, N = 31.751.771.751. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUabc246810Min: 1.72 / Avg: 1.75 / Max: 1.771. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

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 0.11Encoder Speed: 6abc3691215SE +/- 0.071, N = 39.1759.2779.2681. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6abc3691215Min: 9.1 / Avg: 9.17 / Max: 9.321. (CXX) g++ options: -O3 -fPIC -lm

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 96000 - Buffer Size: 1024abc0.46610.93221.39831.86442.3305SE +/- 0.001510, N = 32.0716692.0489312.0526601. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 96000 - Buffer Size: 1024abc246810Min: 2.07 / Avg: 2.07 / Max: 2.071. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4Kabc3691215SE +/- 0.03, N = 39.119.159.211. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4Kabc3691215Min: 9.05 / Avg: 9.11 / Max: 9.151. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUabc246810SE +/- 0.04, N = 38.418.328.35MIN: 7.49 / MAX: 16.85MIN: 7.49 / MAX: 14.34MIN: 7.49 / MAX: 38.941. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUabc3691215Min: 8.32 / Avg: 8.41 / Max: 8.451. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUabc4080120160200SE +/- 0.88, N = 3188.17190.20189.321. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUabc306090120150Min: 186.82 / Avg: 188.17 / Max: 189.831. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUabc714212835SE +/- 0.15, N = 331.8731.5331.68MIN: 28.47 / MAX: 39.41MIN: 28.51 / MAX: 38.58MIN: 28.34 / MAX: 39.141. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUabc714212835Min: 31.59 / Avg: 31.87 / Max: 32.11. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUabc0.21380.42760.64140.85521.069SE +/- 0.00, N = 30.950.940.94MIN: 0.85 / MAX: 7.32MIN: 0.85 / MAX: 6.61MIN: 0.84 / MAX: 7.11. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUabc246810Min: 0.94 / Avg: 0.95 / Max: 0.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamabc306090120150SE +/- 0.41, N = 3125.27123.97123.95
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamabc20406080100Min: 124.8 / Avg: 125.27 / Max: 126.1

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamabc246810SE +/- 0.0263, N = 37.98218.06588.0666
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Streamabc3691215Min: 7.93 / Avg: 7.98 / Max: 8.01

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pabc306090120150SE +/- 1.01, N = 11139.13139.06140.471. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080pabc306090120150Min: 132.22 / Avg: 139.13 / Max: 143.281. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUabc160320480640800SE +/- 3.68, N = 3712.70719.91717.701. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUabc130260390520650Min: 708.79 / Avg: 712.7 / Max: 720.051. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Compression Speedabc4080120160200SE +/- 0.15, N = 3198.4199.8200.41. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Compression Speedabc4080120160200Min: 198.2 / Avg: 198.4 / Max: 198.71. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 512abc7K14K21K28K35KSE +/- 338.62, N = 332781.232901.132572.9
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 512abc6K12K18K24K30KMin: 32158.4 / Avg: 32781.2 / Max: 33323

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: NUMAabc50100150200250SE +/- 0.70, N = 3244.13246.58245.001. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: NUMAabc4080120160200Min: 243.34 / Avg: 244.13 / Max: 245.531. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop Containerabc2004006008001000SE +/- 5.89, N = 3862.76855.12854.21
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop Containerabc150300450600750Min: 855.77 / Avg: 862.76 / Max: 874.47

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 480000 - Buffer Size: 1024abc0.61741.23481.85222.46963.087SE +/- 0.003205, N = 32.7437842.7228462.7171161. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 480000 - Buffer Size: 1024abc246810Min: 2.74 / Avg: 2.74 / Max: 2.751. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

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 or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigabc306090120150SE +/- 1.00, N = 8117.65117.84118.80
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigabc20406080100Min: 116.29 / Avg: 117.65 / Max: 124.62

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUabc510152025SE +/- 0.03, N = 318.2818.4118.23MIN: 17.89MIN: 18.02MIN: 17.931. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUabc510152025Min: 18.24 / Avg: 18.28 / Max: 18.341. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Deepcoinabc14002800420056007000SE +/- 12.35, N = 36392.106428.906368.981. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Deepcoinabc11002200330044005500Min: 6368.91 / Avg: 6392.1 / Max: 6411.061. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

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 1.4Encoder Mode: Preset 4 - Input: Bosphorus 1080pabc246810SE +/- 0.018, N = 37.7377.8097.7501. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 4 - Input: Bosphorus 1080pabc3691215Min: 7.72 / Avg: 7.74 / Max: 7.771. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.0Binary: Pathtracer - Model: Asian Dragonabc3691215SE +/- 0.0065, N = 39.69589.78579.7544MIN: 9.6 / MAX: 9.92MIN: 9.73 / MAX: 9.93MIN: 9.7 / MAX: 9.87
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0Binary: Pathtracer - Model: Asian Dragonabc3691215Min: 9.69 / Avg: 9.7 / Max: 9.71

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Read While Writingabc300K600K900K1200K1500KSE +/- 18612.51, N = 31518515153115715325461. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Read While Writingabc300K600K900K1200K1500KMin: 1496779 / Avg: 1518514.67 / Max: 15555541. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200abc13K26K39K52K65KSE +/- 149.19, N = 361096.5861654.7761396.931. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200abc11K22K33K44K55KMin: 60868.81 / Avg: 61096.58 / Max: 61377.381. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUabc60120180240300SE +/- 1.32, N = 3276.78279.30278.201. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUabc50100150200250Min: 274.18 / Avg: 276.78 / Max: 278.461. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUabc48121620SE +/- 0.07, N = 314.4414.3114.37MIN: 12.82 / MAX: 32.76MIN: 12.84 / MAX: 26.05MIN: 12.83 / MAX: 26.081. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUabc48121620Min: 14.35 / Avg: 14.44 / Max: 14.581. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: AlexNetabc1428425670SE +/- 0.10, N = 360.8161.2460.69
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 16 - Model: AlexNetabc1224364860Min: 60.6 / Avg: 60.81 / Max: 60.94

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUabc0.2610.5220.7831.0441.305SE +/- 0.01, N = 31.151.161.161. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUabc246810Min: 1.14 / Avg: 1.15 / Max: 1.171. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Platformabc306090120150SE +/- 0.75, N = 3145.95146.92145.671. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Platformabc306090120150Min: 144.96 / Avg: 145.95 / Max: 147.421. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Compression Speedabc246810SE +/- 0.01, N = 37.007.067.021. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Compression Speedabc3691215Min: 6.98 / Avg: 7 / Max: 7.011. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Platformabc1224364860SE +/- 0.27, N = 351.9051.5652.001. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Platformabc1020304050Min: 51.38 / Avg: 51.9 / Max: 52.261. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: AlexNetabc20406080100SE +/- 0.09, N = 3107.91108.69108.83
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: AlexNetabc20406080100Min: 107.74 / Avg: 107.91 / Max: 108

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000abc9K18K27K36K45KSE +/- 203.43, N = 341663.1742015.7242016.881. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000abc7K14K21K28K35KMin: 41282.98 / Avg: 41663.17 / Max: 41978.751. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 192000 - Buffer Size: 1024abc0.31470.62940.94411.25881.5735SE +/- 0.005117, N = 31.3986631.3870991.3927461. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 192000 - Buffer Size: 1024abc246810Min: 1.39 / Avg: 1.4 / Max: 1.41. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

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 1.3.1Benchmark: vklBenchmark ISPCabc306090120150SE +/- 0.67, N = 3120120121MIN: 15 / MAX: 1750MIN: 15 / MAX: 1749MIN: 15 / MAX: 1739
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPCabc20406080100Min: 119 / Avg: 120.33 / Max: 121

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4Kabc1326395265SE +/- 0.04, N = 357.9958.4558.381. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4Kabc1224364860Min: 57.94 / Avg: 57.99 / Max: 58.071. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Mallocabc1.1M2.2M3.3M4.4M5.5MSE +/- 19639.79, N = 35226456.995241054.325200128.821. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Mallocabc900K1800K2700K3600K4500KMin: 5187289.02 / Avg: 5226456.99 / Max: 5248603.471. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUabc5001000150020002500SE +/- 12.95, N = 32257.172239.722256.40MIN: 2132.88 / MAX: 3216.45MIN: 2149.54 / MAX: 2500.68MIN: 2134.57 / MAX: 2467.91. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUabc400800120016002000Min: 2233.62 / Avg: 2257.17 / Max: 2278.31. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Scikit-Learn

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

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.1.3Benchmark: TSNE MNIST Datasetabc612182430SE +/- 0.22, N = 326.9126.7126.86
OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.1.3Benchmark: TSNE MNIST Datasetabc612182430Min: 26.48 / Avg: 26.91 / Max: 27.19

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.1.3Benchmark: MNIST Datasetabc20406080100SE +/- 0.23, N = 390.1590.8290.84
OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.1.3Benchmark: MNIST Datasetabc20406080100Min: 89.88 / Avg: 90.15 / Max: 90.61

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.0Binary: Pathtracer ISPC - Model: Asian Dragonabc3691215SE +/- 0.03, N = 311.0611.1411.11MIN: 10.93 / MAX: 11.31MIN: 11.05 / MAX: 11.33MIN: 11.03 / MAX: 11.35
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0Binary: Pathtracer ISPC - Model: Asian Dragonabc3691215Min: 11.02 / Avg: 11.06 / Max: 11.11

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Ultra Fastabc20406080100SE +/- 0.06, N = 380.0179.8179.41
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Ultra Fastabc1530456075Min: 79.91 / Avg: 80.01 / Max: 80.13

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 0.11Encoder Speed: 0abc4080120160200SE +/- 0.94, N = 3195.07193.63195.071. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 0abc4080120160200Min: 193.37 / Avg: 195.07 / Max: 196.611. (CXX) g++ options: -O3 -fPIC -lm

miniBUDE

MiniBUDE is a mini application for the the core computation of the Bristol University Docking Engine (BUDE). This test profile currently makes use of the OpenMP implementation of miniBUDE for CPU benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2abc50100150200250SE +/- 0.41, N = 3243.66245.47245.231. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2abc4080120160200Min: 243.09 / Avg: 243.66 / Max: 244.461. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2abc3691215SE +/- 0.016, N = 39.7479.8199.8091. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm
OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM2abc3691215Min: 9.72 / Avg: 9.75 / Max: 9.781. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 1024abc4080120160200SE +/- 0.06, N = 3177.4177.4176.1
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 1024abc306090120150Min: 177.3 / Avg: 177.4 / Max: 177.5

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUabc5001000150020002500SE +/- 22.77, N = 32406.132394.332388.51MIN: 2255.86MIN: 2261.91MIN: 2252.371. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUabc400800120016002000Min: 2360.77 / Avg: 2406.13 / Max: 2432.21. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4Kabc1224364860SE +/- 0.37, N = 1455.3354.9355.311. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4Kabc1122334455Min: 54.04 / Avg: 55.33 / Max: 58.841. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4Kabc1020304050SE +/- 0.05, N = 345.8746.2046.081. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4Kabc918273645Min: 45.8 / Avg: 45.87 / Max: 45.961. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptabc20406080100SE +/- 0.41, N = 3110.87111.65111.661. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptabc20406080100Min: 110.1 / Avg: 110.87 / Max: 111.491. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: AlexNetabc20406080100SE +/- 0.49, N = 388.6088.4589.08
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: AlexNetabc20406080100Min: 88.04 / Avg: 88.6 / Max: 89.58

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Memory Copyingabc5001000150020002500SE +/- 14.39, N = 32301.692295.122285.431. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Memory Copyingabc400800120016002000Min: 2278.29 / Avg: 2301.69 / Max: 2327.911. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9 video format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 0 - Input: Bosphorus 4Kabc1.2962.5923.8885.1846.48SE +/- 0.02, N = 35.765.725.731. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.13Speed: Speed 0 - Input: Bosphorus 4Kabc246810Min: 5.74 / Avg: 5.76 / Max: 5.81. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.34VGR Performance Metricabc30K60K90K120K150K1329311320581323681. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: AlexNetabc20406080100SE +/- 0.36, N = 377.9577.8077.44
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: AlexNetabc1530456075Min: 77.52 / Avg: 77.95 / Max: 78.67

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 44100 - Buffer Size: 1024abc0.60871.21741.82612.43483.0435SE +/- 0.000534, N = 32.7051222.6974382.6874491. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 44100 - Buffer Size: 1024abc246810Min: 2.7 / Avg: 2.71 / Max: 2.711. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 96000 - Buffer Size: 512abc0.43450.8691.30351.7382.1725SE +/- 0.000948, N = 31.9309601.9184431.9233521. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 96000 - Buffer Size: 512abc246810Min: 1.93 / Avg: 1.93 / Max: 1.931. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on Windshieldabc80160240320400SE +/- 1.35, N = 3378.02375.59376.26
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on Windshieldabc70140210280350Min: 376.54 / Avg: 378.02 / Max: 380.72

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: SENDFILEabc40K80K120K160K200KSE +/- 184.17, N = 3177365.84176973.24178115.641. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: SENDFILEabc30K60K90K120K150KMin: 177032.65 / Avg: 177365.84 / Max: 177668.411. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop Testabc306090120150SE +/- 0.15, N = 3141.00141.91141.10
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop Testabc306090120150Min: 140.79 / Avg: 141 / Max: 141.3

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUabc80160240320400SE +/- 1.63, N = 3348.49347.83346.341. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUabc60120180240300Min: 346.34 / Avg: 348.49 / Max: 351.681. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.0Binary: Pathtracer ISPC - Model: Asian Dragon Objabc3691215SE +/- 0.0123, N = 39.57069.62969.5837MIN: 9.47 / MAX: 9.79MIN: 9.56 / MAX: 9.84MIN: 9.5 / MAX: 9.8
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0Binary: Pathtracer ISPC - Model: Asian Dragon Objabc3691215Min: 9.55 / Avg: 9.57 / Max: 9.59

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Uploadabc510152025SE +/- 0.03, N = 321.2021.1121.241. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Uploadabc510152025Min: 21.15 / Avg: 21.2 / Max: 21.251. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: GoogLeNetabc918273645SE +/- 0.09, N = 341.0540.8040.83
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 64 - Model: GoogLeNetabc918273645Min: 40.87 / Avg: 41.05 / Max: 41.15

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Uploadabc306090120150SE +/- 0.17, N = 3119.13119.63118.901. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Uploadabc20406080100Min: 118.81 / Avg: 119.13 / Max: 119.361. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUabc3691215SE +/- 0.05, N = 311.4711.4911.54MIN: 10.23 / MAX: 55.04MIN: 10.24 / MAX: 17.75MIN: 10.23 / MAX: 58.551. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUabc3691215Min: 11.37 / Avg: 11.47 / Max: 11.541. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Streamabc510152025SE +/- 0.08, N = 320.3320.3920.26
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Streamabc510152025Min: 20.17 / Avg: 20.33 / Max: 20.41

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Streamabc1122334455SE +/- 0.19, N = 349.1849.0349.33
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Streamabc1020304050Min: 48.98 / Avg: 49.18 / Max: 49.55

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: GoogLeNetabc918273645SE +/- 0.11, N = 340.1740.3240.08
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 32 - Model: GoogLeNetabc816243240Min: 39.95 / Avg: 40.17 / Max: 40.33

OpenEMS

OpenEMS is a free and open electromagnetic field solver using the FDTD method. This test profile runs OpenEMS and pyEMS benchmark demos. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMCells/s, More Is BetterOpenEMS 0.0.35-86Test: pyEMS Couplerabc918273645SE +/- 0.06, N = 339.7239.8639.951. (CXX) g++ options: -O3 -rdynamic -ltinyxml -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -lexpat
OpenBenchmarking.orgMCells/s, More Is BetterOpenEMS 0.0.35-86Test: pyEMS Couplerabc816243240Min: 39.62 / Avg: 39.72 / Max: 39.821. (CXX) g++ options: -O3 -rdynamic -ltinyxml -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -lexpat

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 1024abc6K12K18K24K30KSE +/- 369.95, N = 326050.826158.426199.1
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 1024abc5K10K15K20K25KMin: 25610.6 / Avg: 26050.77 / Max: 26785.9

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUabc0.45260.90521.35781.81042.263SE +/- 0.01841, N = 152.000202.011542.01050MIN: 1.92MIN: 2MIN: 21. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUabc246810Min: 1.93 / Avg: 2 / Max: 2.211. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

uvg266

uvg266 is an open-source VVC/H.266 (Versatile Video Coding) encoder based on Kvazaar as part of the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Super Fastabc1530456075SE +/- 0.36, N = 367.0967.4667.36
OpenBenchmarking.orgFrames Per Second, More Is Betteruvg266 0.4.1Video Input: Bosphorus 1080p - Video Preset: Super Fastabc1326395265Min: 66.39 / Avg: 67.09 / Max: 67.6

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 1.4Encoder Mode: Preset 8 - Input: Bosphorus 1080pabc20406080100SE +/- 0.56, N = 381.4581.4081.021. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 8 - Input: Bosphorus 1080pabc1632486480Min: 80.34 / Avg: 81.45 / Max: 82.061. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: GoogLeNetabc1020304050SE +/- 0.04, N = 343.3743.6043.37
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 256 - Model: GoogLeNetabc918273645Min: 43.28 / Avg: 43.37 / Max: 43.42

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1Mabc12002400360048006000SE +/- 13.58, N = 35359.85362.55388.21. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1Mabc9001800270036004500Min: 5333.4 / Avg: 5359.77 / Max: 5378.61. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamabc80160240320400SE +/- 0.80, N = 3371.27371.97373.24
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamabc70140210280350Min: 369.83 / Avg: 371.27 / Max: 372.58

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamabc246810SE +/- 0.0174, N = 38.07988.06478.0373
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamabc3691215Min: 8.05 / Avg: 8.08 / Max: 8.11

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: AlexNetabc306090120150SE +/- 0.47, N = 3117.88118.22118.50
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: AlexNetabc20406080100Min: 117.11 / Avg: 117.88 / Max: 118.73

Neural Magic DeepSparse

This is a benchmark of Neural Magic's DeepSparse using its built-in deepsparse.benchmark utility and various models from their SparseZoo (https://sparsezoo.neuralmagic.com/). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamabc4080120160200SE +/- 0.43, N = 3184.65183.89183.69
OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamabc306090120150Min: 183.8 / Avg: 184.65 / Max: 185.16

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamabc1.22482.44963.67444.89926.124SE +/- 0.0127, N = 35.41545.43795.4437
OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.3.2Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Streamabc246810Min: 5.4 / Avg: 5.42 / Max: 5.44

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Uploadabc3691215SE +/- 0.02, N = 313.5013.4313.451. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Uploadabc48121620Min: 13.47 / Avg: 13.5 / Max: 13.531. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Only - Average Latencyabc0.35570.71141.06711.42281.7785SE +/- 0.012, N = 31.5731.5811.5731. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Only - Average Latencyabc246810Min: 1.56 / Avg: 1.57 / Max: 1.61. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Decompression Speedabc30060090012001500SE +/- 1.17, N = 31444.41451.71447.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19, Long Mode - Decompression Speedabc30060090012001500Min: 1442.2 / Avg: 1444.4 / Max: 1446.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Onlyabc70K140K210K280K350KSE +/- 2326.54, N = 33178723162883178631. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Onlyabc60K120K180K240K300KMin: 313259.63 / Avg: 317872.26 / Max: 320708.71. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Uploadabc4080120160200SE +/- 0.26, N = 3187.09188.02187.761. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Uploadabc306090120150Min: 186.61 / Avg: 187.09 / Max: 187.491. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Compression Speedabc4080120160200SE +/- 0.20, N = 3202.6203.6203.61. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Compression Speedabc4080120160200Min: 202.4 / Avg: 202.6 / Max: 2031. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

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 1.4Encoder Mode: Preset 12 - Input: Bosphorus 1080pabc80160240320400SE +/- 1.08, N = 3348.11348.91347.221. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 1080pabc60120180240300Min: 347 / Avg: 348.11 / Max: 350.271. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Mutexabc900K1800K2700K3600K4500KSE +/- 2569.53, N = 34254720.234273551.984254389.291. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Mutexabc700K1400K2100K2800K3500KMin: 4251494.92 / Avg: 4254720.23 / Max: 4259797.781. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 1.4Encoder Mode: Preset 13 - Input: Bosphorus 1080pabc90180270360450SE +/- 0.84, N = 3398.86399.08397.311. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq
OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 1080pabc70140210280350Min: 397.85 / Avg: 398.86 / Max: 400.531. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Pabellon Barcelona - Compute: CPU-Onlyabc130260390520650SE +/- 0.41, N = 3602.17601.73604.27
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Pabellon Barcelona - Compute: CPU-Onlyabc110220330440550Min: 601.39 / Avg: 602.17 / Max: 602.75

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUabc14002800420056007000SE +/- 19.43, N = 36331.016344.826357.311. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared
OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUabc11002200330044005500Min: 6304.14 / Avg: 6331.01 / Max: 6368.761. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 1080p - Video Preset: Very Fastabc20406080100SE +/- 0.07, N = 374.7974.7375.041. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 1080p - Video Preset: Very Fastabc1428425670Min: 74.65 / Avg: 74.79 / Max: 74.871. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 192000 - Buffer Size: 512abc0.28640.57280.85921.14561.432SE +/- 0.003098, N = 31.2726771.2720571.2674441. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 192000 - Buffer Size: 512abc246810Min: 1.27 / Avg: 1.27 / Max: 1.281. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: System V Message Passingabc2M4M6M8M10MSE +/- 10695.76, N = 38216877.608225295.798192747.721. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: System V Message Passingabc1.4M2.8M4.2M5.6M7MMin: 8200625.78 / Avg: 8216877.6 / Max: 8237049.521. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4Kabc1326395265SE +/- 0.42, N = 1258.7658.6658.531. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4Kabc1224364860Min: 57.68 / Avg: 58.76 / Max: 62.451. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: CPU-Onlyabc400800120016002000SE +/- 2.57, N = 31913.711918.111921.09
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Barbershop - Compute: CPU-Onlyabc30060090012001500Min: 1909.86 / Avg: 1913.71 / Max: 1918.57

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platformabc1020304050SE +/- 0.03, N = 344.4044.3644.531. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platformabc918273645Min: 44.34 / Avg: 44.4 / Max: 44.451. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platformabc4080120160200SE +/- 0.12, N = 3170.61170.76170.111. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platformabc306090120150Min: 170.41 / Avg: 170.61 / Max: 170.841. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUabc0.81161.62322.43483.24644.058SE +/- 0.00059, N = 33.597193.593643.60705MIN: 3.53MIN: 3.53MIN: 3.541. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUabc246810Min: 3.6 / Avg: 3.6 / Max: 3.61. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Compression Speedabc2004006008001000SE +/- 6.47, N = 3869.3866.5866.11. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Compression Speedabc150300450600750Min: 856.7 / Avg: 869.27 / Max: 878.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Decompression Speedabc30060090012001500SE +/- 0.86, N = 31592.61596.21598.41. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3 - Decompression Speedabc30060090012001500Min: 1591 / Avg: 1592.63 / Max: 1593.91. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500abc11K22K33K44K55KSE +/- 61.89, N = 351922.4851890.4551737.261. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500abc9K18K27K36K45KMin: 51833 / Avg: 51922.48 / Max: 52041.291. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

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 2023Implementation: MPI CPU - Input: water_GMX50_bareabc0.19170.38340.57510.76680.9585SE +/- 0.003, N = 30.8520.8510.8491. (CXX) g++ options: -O3
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2023Implementation: MPI CPU - Input: water_GMX50_bareabc246810Min: 0.85 / Avg: 0.85 / Max: 0.861. (CXX) g++ options: -O3

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Matrix Mathabc10K20K30K40K50KSE +/- 69.96, N = 345892.7946046.4845978.841. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Matrix Mathabc8K16K24K32K40KMin: 45800.83 / Avg: 45892.79 / Max: 46030.11. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 44100 - Buffer Size: 512abc0.58021.16041.74062.32082.901SE +/- 0.005177, N = 32.5785992.5705972.5753631. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 44100 - Buffer Size: 512abc246810Min: 2.57 / Avg: 2.58 / Max: 2.591. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

OpenRadioss

OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper Beamabc4080120160200SE +/- 0.32, N = 3187.49187.53188.06
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper Beamabc306090120150Min: 186.94 / Avg: 187.49 / Max: 188.06

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 512abc9K18K27K36K45KSE +/- 83.04, N = 343281.743180.143303.9
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 512abc8K16K24K32K40KMin: 43124.7 / Avg: 43281.7 / Max: 43407.1

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, Onecoinabc20K40K60K80K100KSE +/- 108.37, N = 37815777950779401. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, Onecoinabc14K28K42K56K70KMin: 77940 / Avg: 78156.67 / Max: 782701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: CPU-Onlyabc110220330440550SE +/- 0.90, N = 3493.25491.93492.05
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Classroom - Compute: CPU-Onlyabc90180270360450Min: 491.95 / Avg: 493.25 / Max: 494.99

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Decompression Speedabc30060090012001500SE +/- 0.83, N = 151516.31512.31515.11. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 19 - Decompression Speedabc30060090012001500Min: 1509.7 / Avg: 1516.34 / Max: 1522.71. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Compression Speedabc20406080100SE +/- 0.23, N = 377.177.277.31. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Compression Speedabc1530456075Min: 76.7 / Avg: 77.07 / Max: 77.51. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 1080p - Video Preset: Super Fastabc20406080100SE +/- 0.19, N = 393.4893.5293.301. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 1080p - Video Preset: Super Fastabc20406080100Min: 93.09 / Avg: 93.48 / Max: 93.671. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 1024abc7K14K21K28K35KSE +/- 65.26, N = 330338.730358.130286.7
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 1024abc5K10K15K20K25KMin: 30270.1 / Avg: 30338.73 / Max: 30469.2

Unvanquished

Unvanquished is a modern fork of the Tremulous first person shooter. Unvanquished is powered by the Daemon engine, a combination of the ioquake3 (id Tech 3) engine with the graphically-beautiful XreaL engine. Unvanquished supports a modern OpenGL 3 renderer and other advanced graphics features for this open-source, cross-platform shooter game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.54Resolution: 1920 x 1080 - Effects Quality: Highabc4080120160200SE +/- 0.20, N = 3170.8171.1170.7
OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.54Resolution: 1920 x 1080 - Effects Quality: Highabc306090120150Min: 170.5 / Avg: 170.83 / Max: 171.2

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Zlibabc150300450600750SE +/- 0.29, N = 3717.35717.62716.141. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Zlibabc130260390520650Min: 716.87 / Avg: 717.35 / Max: 717.881. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Unvanquished

Unvanquished is a modern fork of the Tremulous first person shooter. Unvanquished is powered by the Daemon engine, a combination of the ioquake3 (id Tech 3) engine with the graphically-beautiful XreaL engine. Unvanquished supports a modern OpenGL 3 renderer and other advanced graphics features for this open-source, cross-platform shooter game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.54Resolution: 1920 x 1080 - Effects Quality: Ultraabc20406080100SE +/- 0.03, N = 3100.6100.5100.7
OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.54Resolution: 1920 x 1080 - Effects Quality: Ultraabc20406080100Min: 100.6 / Avg: 100.63 / Max: 100.7

CockroachDB

CockroachDB is a cloud-native, distributed SQL database for data intensive applications. This test profile uses a server-less CockroachDB configuration to test various Coackroach workloads on the local host with a single node. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 1024abc9K18K27K36K45KSE +/- 52.10, N = 340249.940176.240255.3
OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 1024abc7K14K21K28K35KMin: 40147.7 / Avg: 40249.9 / Max: 40318.6

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Decompression Speedabc400800120016002000SE +/- 3.74, N = 31758.91761.71762.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 12 - Decompression Speedabc30060090012001500Min: 1751.8 / Avg: 1758.9 / Max: 1764.51. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries too. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: GoogLeNetabc1020304050SE +/- 0.04, N = 344.6544.7044.62
OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.10Device: CPU - Batch Size: 512 - Model: GoogLeNetabc918273645Min: 44.57 / Avg: 44.65 / Max: 44.72

Kvazaar

This is a test of Kvazaar as a CPU-based H.265/HEVC 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 2.2Video Input: Bosphorus 1080p - Video Preset: Mediumabc816243240SE +/- 0.14, N = 334.4634.5234.501. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.orgFrames Per Second, More Is BetterKvazaar 2.2Video Input: Bosphorus 1080p - Video Preset: Mediumabc714212835Min: 34.18 / Avg: 34.46 / Max: 34.631. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Decompression Speedabc400800120016002000SE +/- 1.39, N = 31727.91725.81728.61. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8, Long Mode - Decompression Speedabc30060090012001500Min: 1725.5 / Avg: 1727.93 / Max: 1730.31. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

OpenEMS

OpenEMS is a free and open electromagnetic field solver using the FDTD method. This test profile runs OpenEMS and pyEMS benchmark demos. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMCells/s, More Is BetterOpenEMS 0.0.35-86Test: openEMS MSL_NotchFilterabc20406080100SE +/- 0.06, N = 398.8898.7298.771. (CXX) g++ options: -O3 -rdynamic -ltinyxml -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -lexpat
OpenBenchmarking.orgMCells/s, More Is BetterOpenEMS 0.0.35-86Test: openEMS MSL_NotchFilterabc20406080100Min: 98.79 / Avg: 98.88 / Max: 98.991. (CXX) g++ options: -O3 -rdynamic -ltinyxml -lcrypto -lcurl -lpthread -lsz -lz -ldl -lm -lexpat

Scikit-Learn

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

OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.1.3Benchmark: Sparse Random Projections, 100 Iterationsabc306090120150SE +/- 0.14, N = 3142.73142.60142.50
OpenBenchmarking.orgSeconds, Fewer Is BetterScikit-Learn 1.1.3Benchmark: Sparse Random Projections, 100 Iterationsabc306090120150Min: 142.47 / Avg: 142.73 / Max: 142.92

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Quad SHA-256, Pyriteabc10K20K30K40K50KSE +/- 70.24, N = 34825048310483201. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Quad SHA-256, Pyriteabc8K16K24K32K40KMin: 48110 / Avg: 48250 / Max: 483301. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Fishy Cat - Compute: CPU-Onlyabc50100150200250SE +/- 0.25, N = 3242.51242.40242.75
OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: Fishy Cat - Compute: CPU-Onlyabc4080120160200Min: 242.25 / Avg: 242.51 / Max: 243

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUabc48121620SE +/- 0.00, N = 316.5916.6116.59MIN: 16.45MIN: 16.47MIN: 16.431. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUabc48121620Min: 16.59 / Avg: 16.59 / Max: 16.591. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Semaphoresabc300K600K900K1200K1500KSE +/- 593.53, N = 31189258.321190334.271188758.841. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Semaphoresabc200K400K600K800K1000KMin: 1188074.85 / Avg: 1189258.32 / Max: 1189929.861. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Vector Mathabc9K18K27K36K45KSE +/- 3.69, N = 341733.9741733.4741779.981. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Vector Mathabc7K14K21K28K35KMin: 41728.08 / Avg: 41733.97 / Max: 41740.771. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Decompression Speedabc400800120016002000SE +/- 0.97, N = 31732.41733.21734.31. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 8 - Decompression Speedabc30060090012001500Min: 1730.9 / Avg: 1732.37 / Max: 1734.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Cryptoabc3K6K9K12K15KSE +/- 2.58, N = 314276.3914261.2314271.881. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Cryptoabc2K4K6K8K10KMin: 14273.32 / Avg: 14276.39 / Max: 14281.511. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Hashabc300K600K900K1200K1500KSE +/- 854.18, N = 31548492.991549710.051548182.161. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Hashabc300K600K900K1200K1500KMin: 1547054.3 / Avg: 1548492.99 / Max: 1550010.131. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Pollabc200K400K600K800K1000KSE +/- 379.65, N = 3823762.13823948.19823174.051. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Pollabc140K280K420K560K700KMin: 823003.4 / Avg: 823762.13 / Max: 824167.091. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (silesia.tar) using Zstd (Zstandard) compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Decompression Speedabc30060090012001500SE +/- 0.35, N = 31619.41618.71620.21. (CC) gcc options: -O3 -pthread -lz -llzma -llz4
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.4Compression Level: 3, Long Mode - Decompression Speedabc30060090012001500Min: 1618.7 / Avg: 1619.4 / Max: 1619.81. (CC) gcc options: -O3 -pthread -lz -llzma -llz4

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) 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 4.0Binary: Pathtracer - Model: Asian Dragon Objabc246810SE +/- 0.0103, N = 38.71028.71688.7124MIN: 8.64 / MAX: 8.85MIN: 8.66 / MAX: 8.83MIN: 8.66 / MAX: 8.86
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0Binary: Pathtracer - Model: Asian Dragon Objabc3691215Min: 8.7 / Avg: 8.71 / Max: 8.73

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Function Callabc11K22K33K44K55KSE +/- 1.32, N = 349787.6249807.0549784.191. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Function Callabc9K18K27K36K45KMin: 49786.23 / Avg: 49787.62 / Max: 49790.261. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Unvanquished

Unvanquished is a modern fork of the Tremulous first person shooter. Unvanquished is powered by the Daemon engine, a combination of the ioquake3 (id Tech 3) engine with the graphically-beautiful XreaL engine. Unvanquished supports a modern OpenGL 3 renderer and other advanced graphics features for this open-source, cross-platform shooter game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.54Resolution: 1920 x 1080 - Effects Quality: Mediumabc50100150200250SE +/- 0.09, N = 3248.2248.1248.2
OpenBenchmarking.orgFrames Per Second, More Is BetterUnvanquished 0.54Resolution: 1920 x 1080 - Effects Quality: Mediumabc4080120160200Min: 248.1 / Avg: 248.23 / Max: 248.4

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: CPU Stressabc4K8K12K16K20KSE +/- 32.25, N = 317781.8617780.7617782.211. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: CPU Stressabc3K6K9K12K15KMin: 17725.9 / Avg: 17781.86 / Max: 17837.621. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: x86_64 RdRandabc20K40K60K80K100KSE +/- 0.57, N = 381909.4581907.9781907.391. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: x86_64 RdRandabc14K28K42K56K70KMin: 81908.36 / Avg: 81909.45 / Max: 81910.251. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pabc3691215SE +/- 0.13, N = 313.3113.3113.311. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080pabc48121620Min: 13.18 / Avg: 13.31 / Max: 13.561. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

Connections: 1

a: The test quit with a non-zero exit status.

b: The test quit with a non-zero exit status.

c: The test quit with a non-zero exit status.

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Random Fillabc200K400K600K800K1000KSE +/- 20633.51, N = 1584645786638111024391. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenBenchmarking.orgOp/s, More Is BetterRocksDB 7.9.2Test: Random Fillabc200K400K600K800K1000KMin: 777779 / Avg: 846456.93 / Max: 11158531. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Context Switchingabc400K800K1200K1600K2000KSE +/- 48908.21, N = 151897936.071941869.461906264.571. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Context Switchingabc300K600K900K1200K1500KMin: 1508578.89 / Avg: 1897936.07 / Max: 2317651.451. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Socket Activityabc2K4K6K8K10KSE +/- 146.72, N = 158703.549385.839379.731. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Socket Activityabc16003200480064008000Min: 7928.02 / Avg: 8703.54 / Max: 9454.611. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: CPU Cacheabc20406080100SE +/- 2.33, N = 1283.6089.5072.101. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: CPU Cacheabc20406080100Min: 68.13 / Avg: 83.6 / Max: 97.631. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: IO_uringabc6001200180024003000SE +/- 36.56, N = 151965.751723.132856.891. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: IO_uringabc5001000150020002500Min: 1593.44 / Avg: 1965.75 / Max: 2161.121. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Atomicabc4080120160200SE +/- 3.12, N = 15146.11145.01162.651. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Atomicabc306090120150Min: 130.54 / Avg: 146.11 / Max: 169.81. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: MEMFDabc1326395265SE +/- 3.98, N = 1258.5741.6552.991. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: MEMFDabc1224364860Min: 32.85 / Avg: 58.57 / Max: 76.41. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Futexabc400K800K1200K1600K2000KSE +/- 41062.68, N = 151579624.881723742.381689762.381. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Futexabc300K600K900K1200K1500KMin: 1333105.69 / Avg: 1579624.88 / Max: 1896924.71. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: MMAPabc20406080100SE +/- 2.50, N = 1574.7050.2649.731. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: MMAPabc1428425670Min: 45.42 / Avg: 74.7 / Max: 83.011. (CC) gcc options: -std=gnu99 -O2 -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Write - Average Latencyabc306090120150SE +/- 3.53, N = 12113.08103.7182.831. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Write - Average Latencyabc20406080100Min: 94.39 / Avg: 113.08 / Max: 128.31. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Writeabc13002600390052006500SE +/- 146.05, N = 124472482160361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 500 - Mode: Read Writeabc10002000300040005000Min: 3897.24 / Avg: 4471.78 / Max: 5297.181. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latencyabc1530456075SE +/- 1.63, N = 1260.8068.0741.631. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latencyabc1326395265Min: 50.29 / Avg: 60.8 / Max: 66.481. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Writeabc13002600390052006500SE +/- 122.04, N = 124148367360051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 250 - Mode: Read Writeabc10002000300040005000Min: 3760.31 / Avg: 4147.67 / Max: 4971.361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latencyabc612182430SE +/- 0.75, N = 1224.5523.6419.601. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latencyabc612182430Min: 21.47 / Avg: 24.55 / Max: 28.191. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Writeabc11002200330044005500SE +/- 122.37, N = 124114423051021. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Writeabc9001800270036004500Min: 3546.82 / Avg: 4114.04 / Max: 4657.171. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

KeyDB

A benchmark of KeyDB as an open-source, multi-threaded fork of the Redis server now developed by Snapchat. The developers self-describe the KeyDB in-memory data structure store as a "faster drop in alternative to Redis". Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPOP - Parallel Connections: 100abc300K600K900K1200K1500KSE +/- 37181.44, N = 121301994.001336505.381303645.001. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPOP - Parallel Connections: 100abc200K400K600K800K1000KMin: 915181.06 / Avg: 1301994 / Max: 1400795.621. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SADD - Parallel Connections: 50abc400K800K1200K1600K2000KSE +/- 44518.65, N = 151345962.361920122.881225640.501. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SADD - Parallel Connections: 50abc300K600K900K1200K1500KMin: 1124328.25 / Avg: 1345962.36 / Max: 1631214.881. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPOP - Parallel Connections: 50abc300K600K900K1200K1500KSE +/- 22923.50, N = 151403220.971585238.251479465.001. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: LPOP - Parallel Connections: 50abc300K600K900K1200K1500KMin: 1301100.75 / Avg: 1403220.97 / Max: 1596118.251. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SET - Parallel Connections: 50abc300K600K900K1200K1500KSE +/- 40517.71, N = 121249454.751253729.881422030.121. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: SET - Parallel Connections: 50abc200K400K600K800K1000KMin: 1076403.12 / Avg: 1249454.75 / Max: 1488848.51. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: GET - Parallel Connections: 50abc400K800K1200K1600K2000KSE +/- 63048.39, N = 151481042.421373626.381690102.751. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD
OpenBenchmarking.orgRequests Per Second, More Is BetterKeyDB 6.3.2Test: GET - Parallel Connections: 50abc300K600K900K1200K1500KMin: 1239679.62 / Avg: 1481042.42 / Max: 21778901. (CXX) g++ options: -ggdb -rdynamic -lm -lz -lcrypto -lbz2 -lzstd -llz4 -lsnappy -latomic -ldl -pthread -lrt -luuid -lcurl -lssl -std=c++14 -pedantic -fno-rtti -O2 -flto -MMD

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 toolkit. Learn more via the OpenBenchmarking.org test page.

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU

a: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

a: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

a: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

a: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUabc3691215SE +/- 0.66035, N = 158.938858.133739.33670MIN: 7.53MIN: 8.1MIN: 9.281. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUabc3691215Min: 7.61 / Avg: 8.94 / Max: 13.911. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUabc246810SE +/- 0.17183, N = 158.697488.372398.44040MIN: 5.76MIN: 5.85MIN: 5.81. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUabc3691215Min: 7.89 / Avg: 8.7 / Max: 10.031. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

a: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

a: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUabc0.36920.73841.10761.47681.846SE +/- 0.105801, N = 151.2518711.6407600.959942MIN: 0.91MIN: 1.59MIN: 0.911. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUabc246810Min: 0.96 / Avg: 1.25 / Max: 1.881. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -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 0.11Encoder Speed: 10, Losslessabc246810SE +/- 0.185, N = 156.5716.4386.4391. (CXX) g++ options: -O3 -fPIC -lm
OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, Losslessabc3691215Min: 6.09 / Avg: 6.57 / Max: 8.061. (CXX) g++ options: -O3 -fPIC -lm

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pabc4080120160200SE +/- 3.61, N = 15175.80177.46168.561. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080pabc306090120150Min: 161.27 / Avg: 175.8 / Max: 202.531. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pabc4080120160200SE +/- 3.99, N = 15188.81171.53201.291. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080pabc4080120160200Min: 168.17 / Avg: 188.81 / Max: 205.051. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pabc4080120160200SE +/- 2.71, N = 15147.69145.58169.961. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.6Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080pabc306090120150Min: 139.98 / Avg: 147.69 / Max: 173.61. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

359 Results Shown

RocksDB
PostgreSQL:
  100 - 800 - Read Write - Average Latency
  100 - 800 - Read Write
oneDNN
SVT-AV1
PostgreSQL:
  100 - 1000 - Read Write
  100 - 1000 - Read Write - Average Latency
Neural Magic DeepSparse:
  NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream:
    items/sec
    ms/batch
OpenVINO:
  Vehicle Detection FP16 - CPU:
    ms
    FPS
FLAC Audio Encoding
EnCodec
VP9 libvpx Encoding
CockroachDB
KeyDB:
  SET - 100
  GET - 100
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
oneDNN
AOM AV1
Cpuminer-Opt:
  Blake-2 S
  x25x
EnCodec
Cpuminer-Opt
AOM AV1
EnCodec
FFmpeg:
  libx264 - Video On Demand:
    Seconds
    FPS
RocksDB
VP9 libvpx Encoding
KeyDB
Stress-NG
PostgreSQL
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    items/sec
    ms/batch
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    items/sec
    ms/batch
PostgreSQL
KeyDB:
  HMSET - 100
  SADD - 100
PostgreSQL:
  1 - 500 - Read Only - Average Latency
  1 - 500 - Read Only
Kvazaar
ClickHouse
CockroachDB
Kvazaar
PostgreSQL:
  1 - 800 - Read Write - Average Latency
  1 - 800 - Read Write
KeyDB
PostgreSQL:
  100 - 800 - Read Only - Average Latency
  100 - 800 - Read Only
  1 - 500 - Read Write - Average Latency
  100 - 250 - Read Only
  100 - 250 - Read Only - Average Latency
oneDNN
JPEG XL Decoding libjxl
miniBUDE
EnCodec
miniBUDE
PostgreSQL:
  1 - 500 - Read Write
  1 - 800 - Read Only - Average Latency
  1 - 800 - Read Only
Cpuminer-Opt
ClickHouse
PostgreSQL
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream:
    ms/batch
    items/sec
uvg266
PostgreSQL
uvg266
CockroachDB
RocksDB
Timed FFmpeg Compilation
VVenC
Neural Magic DeepSparse
CockroachDB
VVenC
Neural Magic DeepSparse
RocksDB
PostgreSQL
Stress-NG
PostgreSQL
SVT-AV1
PostgreSQL
Kvazaar
PostgreSQL
Neural Magic DeepSparse
PostgreSQL:
  1 - 250 - Read Write - Average Latency
  1 - 250 - Read Write
CockroachDB
Neural Magic DeepSparse
FFmpeg:
  libx265 - Video On Demand:
    FPS
    Seconds
uvg266:
  Bosphorus 4K - Ultra Fast
  Bosphorus 1080p - Very Fast
nekRS
oneDNN
Kvazaar
CockroachDB
Stress-NG
Embree
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream:
    ms/batch
    items/sec
uvg266
FFmpeg
CockroachDB
OpenVINO
FFmpeg
AOM AV1
OpenVINO
Zstd Compression
uvg266
KeyDB
Neural Magic DeepSparse
uvg266
OpenVINO
Kvazaar
PostgreSQL
nginx
oneDNN
PostgreSQL
JPEG XL libjxl
Embree
Neural Magic DeepSparse
OpenVINO
JPEG XL libjxl
libavif avifenc
CockroachDB
Neural Magic DeepSparse
JPEG XL libjxl
CockroachDB
libavif avifenc
AOM AV1
OpenVINO
JPEG XL libjxl
CockroachDB
PostgreSQL:
  1 - 1000 - Read Only
  1 - 1000 - Read Only - Average Latency
oneDNN
Neural Magic DeepSparse
Xmrig
TensorFlow
ClickHouse
Cpuminer-Opt:
  Ringcoin
  Skeincoin
VVenC
Kvazaar
oneDNN
CockroachDB
oneDNN
PostgreSQL
OpenVINO
SVT-AV1
CockroachDB
JPEG XL Decoding libjxl
PostgreSQL
FFmpeg
Timed Linux Kernel Compilation
OpenVINO
FFmpeg
CockroachDB
oneDNN
Cpuminer-Opt
OpenVKL
JPEG XL libjxl:
  JPEG - 100
  PNG - 100
OpenRadioss
Neural Magic DeepSparse
RocksDB
CockroachDB
VVenC
Zstd Compression
Cpuminer-Opt
uvg266
VP9 libvpx Encoding
Neural Magic DeepSparse
Stress-NG
OpenVINO
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    ms/batch
    items/sec
SVT-AV1
Kvazaar
nginx
oneDNN
Stargate Digital Audio Workstation
Blender
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    items/sec
    ms/batch
OpenVINO:
  Person Detection FP16 - CPU
  Face Detection FP16 - CPU
libavif avifenc
Stargate Digital Audio Workstation
AOM AV1
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU
  Weld Porosity Detection FP16 - CPU
  Weld Porosity Detection FP16 - CPU
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU
Neural Magic DeepSparse:
  CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream:
    ms/batch
    items/sec
AOM AV1
OpenVINO
Zstd Compression
CockroachDB
Stress-NG
OpenRadioss
Stargate Digital Audio Workstation
Timed Linux Kernel Compilation
oneDNN
Cpuminer-Opt
SVT-AV1
Embree
RocksDB
nginx
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    FPS
    ms
TensorFlow
OpenVINO
FFmpeg
Zstd Compression
FFmpeg
TensorFlow
nginx
Stargate Digital Audio Workstation
OpenVKL
AOM AV1
Stress-NG
OpenVINO
Scikit-Learn:
  TSNE MNIST Dataset
  MNIST Dataset
Embree
uvg266
libavif avifenc
miniBUDE:
  OpenMP - BM2:
    GFInst/s
    Billion Interactions/s
CockroachDB
oneDNN
AOM AV1:
  Speed 6 Realtime - Bosphorus 4K
  Speed 8 Realtime - Bosphorus 4K
Cpuminer-Opt
TensorFlow
Stress-NG
VP9 libvpx Encoding
BRL-CAD
TensorFlow
Stargate Digital Audio Workstation:
  44100 - 1024
  96000 - 512
OpenRadioss
Stress-NG
OpenRadioss
OpenVINO
Embree
FFmpeg
TensorFlow
FFmpeg
OpenVINO
Neural Magic DeepSparse:
  NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Stream:
    ms/batch
    items/sec
TensorFlow
OpenEMS
CockroachDB
oneDNN
uvg266
SVT-AV1
TensorFlow
Xmrig
Neural Magic DeepSparse:
  CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream:
    ms/batch
    items/sec
TensorFlow
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
FFmpeg
PostgreSQL
Zstd Compression
PostgreSQL
FFmpeg
Zstd Compression
SVT-AV1
Stress-NG
SVT-AV1
Blender
OpenVINO
Kvazaar
Stargate Digital Audio Workstation
Stress-NG
AOM AV1
Blender
FFmpeg:
  libx265 - Platform:
    FPS
    Seconds
oneDNN
Zstd Compression:
  3, Long Mode - Compression Speed
  3 - Decompression Speed
nginx
GROMACS
Stress-NG
Stargate Digital Audio Workstation
OpenRadioss
CockroachDB
Cpuminer-Opt
Blender
Zstd Compression:
  19 - Decompression Speed
  12 - Compression Speed
Kvazaar
CockroachDB
Unvanquished
Stress-NG
Unvanquished
CockroachDB
Zstd Compression
TensorFlow
Kvazaar
Zstd Compression
OpenEMS
Scikit-Learn
Cpuminer-Opt
Blender
oneDNN
Stress-NG:
  Semaphores
  Vector Math
Zstd Compression
Stress-NG:
  Crypto
  Hash
  Poll
Zstd Compression
Embree
Stress-NG
Unvanquished
Stress-NG:
  CPU Stress
  x86_64 RdRand
AOM AV1
RocksDB
Stress-NG:
  Context Switching
  Socket Activity
  CPU Cache
  IO_uring
  Atomic
  MEMFD
  Futex
  MMAP
PostgreSQL:
  100 - 500 - Read Write - Average Latency
  100 - 500 - Read Write
  100 - 250 - Read Write - Average Latency
  100 - 250 - Read Write
  100 - 100 - Read Write - Average Latency
  100 - 100 - Read Write
KeyDB:
  LPOP - 100
  SADD - 50
  LPOP - 50
  SET - 50
  GET - 50
oneDNN:
  Deconvolution Batch shapes_3d - f32 - CPU
  Deconvolution Batch shapes_1d - f32 - CPU
  IP Shapes 1D - u8s8f32 - CPU
libavif avifenc
AOM AV1:
  Speed 10 Realtime - Bosphorus 1080p
  Speed 9 Realtime - Bosphorus 1080p
  Speed 6 Realtime - Bosphorus 1080p

a

Processor: Intel Core i5-12400 @ 5.60GHz (6 Cores / 12 Threads), Motherboard: MSI PRO Z690-A WIFI DDR4(MS-7D25) v1.0 (Dasharo coreboot+UEFI v1.0.0 BIOS), Chipset: Intel Device 7aa7, Memory: 16GB, Disk: Western Digital WD_BLACK SN750 SE 500GB, Graphics: MSI Intel ADL-S GT1 14GB (1450MHz), Audio: Realtek ALC897, Monitor: DELL S2409W, Network: Intel I225-V + Intel Device 7af0

OS: Ubuntu 22.04, Kernel: 5.15.0-47-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, Vulkan: 1.2.204, Compiler: GCC 11.2.0, 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-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 3 March 2023 16:03 by user phoronix.

b

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 5 March 2023 07:45 by user phoronix.

c

Processor: Intel Core i5-12400 @ 5.60GHz (6 Cores / 12 Threads), Motherboard: MSI PRO Z690-A WIFI DDR4(MS-7D25) v1.0 (Dasharo coreboot+UEFI v1.0.0 BIOS), Chipset: Intel Device 7aa7, Memory: 16GB, Disk: Western Digital WD_BLACK SN750 SE 500GB, Graphics: MSI Intel ADL-S GT1 14GB (1450MHz), Audio: Realtek ALC897, Monitor: DELL S2409W, Network: Intel I225-V + Intel Device 7af0

OS: Ubuntu 22.04, Kernel: 5.15.0-47-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, Vulkan: 1.2.204, Compiler: GCC 11.3.0, 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-cet --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-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x1f - Thermald 2.4.9
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 5 March 2023 19:04 by user phoronix.