dddxxx

Tests for a future article. Intel Core i7-8565U testing with a Dell 0KTW76 (1.17.0 BIOS) and Intel UHD 620 WHL GT2 15GB 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 2308061-NE-DDDXXX46317
Jump To Table - Results

View

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

Limit displaying results to tests within:

AV1 2 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 3 Tests
CPU Massive 4 Tests
Creator Workloads 8 Tests
Database Test Suite 5 Tests
Encoding 4 Tests
Game Development 2 Tests
HPC - High Performance Computing 2 Tests
Machine Learning 2 Tests
Multi-Core 10 Tests
NVIDIA GPU Compute 2 Tests
Intel oneAPI 3 Tests
Programmer / Developer System Benchmarks 3 Tests
Python Tests 4 Tests
Server 5 Tests
Server CPU Tests 4 Tests
Video Encoding 3 Tests
Vulkan Compute 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
Performance Per
Dollar
Date
Run
  Test
  Duration
a
August 05 2023
  15 Hours, 4 Minutes
b
August 06 2023
  15 Hours, 10 Minutes
Invert Hiding All Results Option
  15 Hours, 7 Minutes
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):


dddxxxOpenBenchmarking.orgPhoronix Test SuiteIntel Core i7-8565U @ 4.60GHz (4 Cores / 8 Threads)Dell 0KTW76 (1.17.0 BIOS)Intel Cannon Point-LP16GBSK hynix PC401 NVMe 256GBIntel UHD 620 WHL GT2 15GB (1150MHz)Realtek ALC3271Qualcomm Atheros QCA6174 802.11acUbuntu 22.045.19.0-rc6-phx-retbleed (x86_64)GNOME Shell 42.2X Server + Wayland4.6 Mesa 22.0.1OpenCL 3.01.3.204GCC 11.3.0GCC 11.4.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerOpenGLOpenCLVulkanCompilersFile-SystemScreen ResolutionDddxxx 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-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 - 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-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/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 - NONE / errors=remount-ro,relatime,rw / Block Size: 4096- Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 2.4.9 - a: OpenJDK Runtime Environment (build 11.0.18+10-post-Ubuntu-0ubuntu122.04) - b: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04) - a: Python 3.10.6- b: Python 3.10.12- itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of IBRS + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of IBRS IBPB: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Not affected

a vs. b ComparisonPhoronix Test SuiteBaseline+48%+48%+96%+96%+144%+144%+192%+192%61.2%61.1%42.1%40.2%37.1%36.4%34.9%34.6%34.5%33.9%31.1%28.4%27.1%24.3%24.1%23.8%23%21.9%21.7%20.7%19.3%19%18.4%18.4%18.2%17.8%17.6%17.6%17.1%15.5%15%14.1%13.8%13.1%12.1%12%12%11.3%9.1%8.8%8.4%8.1%7.8%7.6%7.3%7.2%6.9%6.5%5.1%5.1%4.9%4.8%4.7%4.3%4.3%4.2%4%3.9%3.7%3.5%3.4%3.3%3.2%3.2%2.8%2.6%2.3%2.3%2%1192.1%B.L.N.Q.A.S.I - A.M.SB.L.N.Q.A.S.I - A.M.SCPU - shufflenet-v2Redis - 500 - 1:5237.7%437.5%Redis - 500 - 1:10MMAPRedis - 100 - 1:5Redis - 50 - 1:5CPU - mnasnetRedis - 100 - 1:10MEMFD31.3%Redis - 50 - 1:10FutexForkingB.L.N.Q.A - A.M.SCPU CacheCryptoZlibPreset 12 - Bosphorus 4K22.5%HashB.L.N.Q.A - A.M.SPthreadCPU-v3-v3 - mobilenet-v3SENDFILEN.T.C.B.b.u.S - A.M.SN.T.C.D.m - A.M.SN.T.C.B.b.u.S - A.M.SN.T.C.D.m - A.M.SAVL TreeN.T.C.B.b.u.c - A.M.SN.T.C.B.b.u.c - A.M.SMallocPreset 8 - Bosphorus 1080p15.2%Preset 4 - Bosphorus 4K15.2%Mutexgravity_spheres_volume/dim_512/scivis/real_timeS.N.1200 - 100 - 50013.3%100 - 100 - 500100 - 100 - 50012.2%NUMAPreset 4 - Bosphorus 1080pPollAtomicPreset 13 - Bosphorus 4K9.9%200 - 100 - 500particle_volume/ao/real_time9.1%Fused Multiply-Add8.8%CPU-v2-v2 - mobilenet-v2C.S.9.P.Y.P - A.M.SV.F.PPipeCPU StressIO_uringG.Q.D.S500 - 100 - 2007.1%500 - 1 - 5007%500 - 100 - 5006.9%CloningCPU - mobilenet6.7%4 - 256 - 576.6%C.S.9.P.Y.P - A.M.SVulkan GPU - regnety_400m6%8 - 256 - 575.7%G.C.S.F5.3%S.V.M.P5.3%C.D.Y.C.S.I - A.M.SFunction Callparticle_volume/pathtracer/real_time5%CPU - resnet50gravity_spheres_volume/dim_512/pathtracer/real_time4.8%128CPU - squeezenet_ssdPathtracer ISPC - Asian Dragon Obj4.4%500 - 100 - 5008 - 256 - 32x86_64 RdRand4.3%500 - 1 - 2004.2%500 - 1 - 500Context Switching200 - 1 - 2004%500 - 100 - 200CPU - efficientnet-b0Vulkan GPU - FastestDet100 - 100 - 200Matrix 3D Math100 - 100 - 2003.3%C.D.Y.C.S.I - A.M.SVulkan GPU - squeezenet_ssd1:103.1%R.5.B - A.M.S3.1%Bosphorus 1080p - Fast3%CPU - FastestDet3%Vulkan GPU - mobilenetR.5.B - A.M.S2.7%200 - 1 - 5002.7%Floating PointTime To Compile2.5%CPU - yolov4-tiny2.4%C.C.R.5.I - A.M.S2.4%C.C.R.5.I - A.M.S2.3%Chimera 1080p8 - 256 - 512CPU - blazeface2.2%particle_volume/scivis/real_time2.2%1:1002.1%Vulkan GPU - blazeface2.1%Preset 12 - Bosphorus 1080pPathtracer - Crown2%SQLiteNeural Magic DeepSparseNeural Magic DeepSparseNCNNRedis 7.0.12 + memtier_benchmarkSQLiteSQLiteRedis 7.0.12 + memtier_benchmarkStress-NGRedis 7.0.12 + memtier_benchmarkRedis 7.0.12 + memtier_benchmarkNCNNRedis 7.0.12 + memtier_benchmarkStress-NGRedis 7.0.12 + memtier_benchmarkStress-NGStress-NGNeural Magic DeepSparseStress-NGStress-NGStress-NGSVT-AV1Stress-NGNeural Magic DeepSparseStress-NGNCNNStress-NGNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseNeural Magic DeepSparseStress-NGNeural Magic DeepSparseNeural Magic DeepSparseStress-NGSVT-AV1SVT-AV1Stress-NGOSPRaydav1dApache IoTDBApache IoTDBApache IoTDBStress-NGSVT-AV1Stress-NGStress-NGSVT-AV1Apache IoTDBOSPRayStress-NGNCNNNeural Magic DeepSparseStress-NGStress-NGStress-NGStress-NGStress-NGApache IoTDBApache IoTDBApache IoTDBStress-NGNCNNLiquid-DSPNeural Magic DeepSparseNCNNLiquid-DSPStress-NGStress-NGNeural Magic DeepSparseStress-NGOSPRayNCNNOSPRaylibxsmmNCNNEmbreeApache IoTDBLiquid-DSPStress-NGApache IoTDBApache IoTDBStress-NGApache IoTDBApache IoTDBNCNNNCNNApache IoTDBStress-NGApache IoTDBNeural Magic DeepSparseNCNNMemcachedNeural Magic DeepSparseVVenCNCNNNCNNNeural Magic DeepSparseApache IoTDBStress-NGBuild2NCNNNeural Magic DeepSparseNeural Magic DeepSparsedav1dLiquid-DSPNCNNOSPRayMemcachedNCNNSVT-AV1Embreeab

dddxxxsqlite: 1deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamncnn: CPU - shufflenet-v2memtier-benchmark: Redis - 500 - 1:5sqlite: 2sqlite: 4memtier-benchmark: Redis - 500 - 1:10stress-ng: MMAPmemtier-benchmark: Redis - 100 - 1:5memtier-benchmark: Redis - 50 - 1:5ncnn: CPU - mnasnetmemtier-benchmark: Redis - 100 - 1:10stress-ng: MEMFDmemtier-benchmark: Redis - 50 - 1:10stress-ng: Futexstress-ng: Forkingdeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamstress-ng: CPU Cachestress-ng: Cryptostress-ng: Zlibsvt-av1: Preset 12 - Bosphorus 4Kstress-ng: Hashdeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamstress-ng: Pthreadncnn: CPU-v3-v3 - mobilenet-v3stress-ng: SENDFILEdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamstress-ng: AVL Treedeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamstress-ng: Mallocsvt-av1: Preset 8 - Bosphorus 1080psvt-av1: Preset 4 - Bosphorus 4Kstress-ng: Mutexospray: gravity_spheres_volume/dim_512/scivis/real_timedav1d: Summer Nature 1080papache-iotdb: 200 - 100 - 500apache-iotdb: 100 - 100 - 500apache-iotdb: 100 - 100 - 500stress-ng: NUMAsvt-av1: Preset 4 - Bosphorus 1080pstress-ng: Pollstress-ng: Atomicsvt-av1: Preset 13 - Bosphorus 4Kapache-iotdb: 200 - 100 - 500ospray: particle_volume/ao/real_timestress-ng: Fused Multiply-Addncnn: CPU-v2-v2 - mobilenet-v2deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamstress-ng: Vector Floating Pointstress-ng: Pipestress-ng: CPU Stressstress-ng: IO_uringstress-ng: Glibc Qsort Data Sortingapache-iotdb: 500 - 100 - 200apache-iotdb: 500 - 1 - 500apache-iotdb: 500 - 100 - 500stress-ng: Cloningncnn: CPU - mobilenetliquid-dsp: 4 - 256 - 57deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamncnn: Vulkan GPU - regnety_400mliquid-dsp: 8 - 256 - 57stress-ng: Glibc C String Functionsstress-ng: System V Message Passingdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamstress-ng: Function Callospray: particle_volume/pathtracer/real_timencnn: CPU - resnet50ospray: gravity_spheres_volume/dim_512/pathtracer/real_timelibxsmm: 128ncnn: CPU - squeezenet_ssdembree: Pathtracer ISPC - Asian Dragon Objapache-iotdb: 500 - 100 - 500liquid-dsp: 8 - 256 - 32stress-ng: x86_64 RdRandapache-iotdb: 500 - 1 - 200apache-iotdb: 500 - 1 - 500stress-ng: Context Switchingapache-iotdb: 200 - 1 - 200apache-iotdb: 500 - 100 - 200ncnn: CPU - efficientnet-b0ncnn: Vulkan GPU - FastestDetapache-iotdb: 100 - 100 - 200stress-ng: Matrix 3D Mathapache-iotdb: 100 - 100 - 200deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamncnn: Vulkan GPU - squeezenet_ssdmemcached: 1:10deepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamvvenc: Bosphorus 1080p - Fastncnn: CPU - FastestDetncnn: Vulkan GPU - mobilenetdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamapache-iotdb: 200 - 1 - 500stress-ng: Floating Pointbuild2: Time To Compilencnn: CPU - yolov4-tinydeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdav1d: Chimera 1080pliquid-dsp: 8 - 256 - 512ncnn: CPU - blazefaceospray: particle_volume/scivis/real_timememcached: 1:100ncnn: Vulkan GPU - blazefacesvt-av1: Preset 12 - Bosphorus 1080pembree: Pathtracer - Crownsvt-av1: Preset 13 - Bosphorus 1080pncnn: Vulkan GPU - vision_transformerdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamstress-ng: Vector Mathdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamncnn: Vulkan GPU - vgg16ospray: gravity_spheres_volume/dim_512/ao/real_timeapache-iotdb: 100 - 1 - 200ncnn: CPU - vgg16apache-iotdb: 100 - 1 - 500memcached: 1:5ncnn: Vulkan GPU - resnet18ncnn: Vulkan GPU - efficientnet-b0build-llvm: Unix Makefilesz3: 1.smt2ncnn: Vulkan GPU - resnet50apache-iotdb: 200 - 100 - 200stress-ng: Matrix Mathembree: Pathtracer ISPC - Crowndeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamquantlib: liquid-dsp: 2 - 256 - 512apache-iotdb: 100 - 1 - 500dav1d: Summer Nature 4Kapache-iotdb: 100 - 1 - 200stress-ng: Memory Copyingapache-iotdb: 200 - 100 - 200ncnn: CPU - regnety_400mbuild-godot: Time To Compiledeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamvvenc: Bosphorus 4K - Fastdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamvvenc: Bosphorus 4K - Fasterncnn: CPU - vision_transformerdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamstress-ng: Wide Vector Mathembree: Pathtracer - Asian Dragon Objstress-ng: Semaphoresdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamliquid-dsp: 1 - 256 - 32libxsmm: 32ncnn: Vulkan GPU - googlenetbuild-llvm: Ninjancnn: Vulkan GPU - mnasnetsvt-av1: Preset 8 - Bosphorus 4Kncnn: Vulkan GPU - yolov4-tinyxonotic: 1920 x 1080 - Ultrastress-ng: Socket Activityembree: Pathtracer ISPC - Asian Dragonapache-iotdb: 200 - 1 - 200deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamembree: Pathtracer - Asian Dragonapache-iotdb: 500 - 1 - 200xonotic: 1920 x 1080 - Ultimatez3: 2.smt2ncnn: CPU - alexnetapache-iotdb: 200 - 1 - 500ncnn: Vulkan GPU - shufflenet-v2ncnn: CPU - googlenetliquid-dsp: 2 - 256 - 57liquid-dsp: 4 - 256 - 32xonotic: 1920 x 1080 - Highdav1d: Chimera 1080p 10-bitdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamncnn: Vulkan GPU - alexnetliquid-dsp: 1 - 256 - 512vvenc: Bosphorus 1080p - Fasterliquid-dsp: 2 - 256 - 32libxsmm: 64deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamencode-opus: WAV To Opus Encodencnn: CPU - resnet18xonotic: 1920 x 1080 - Lowliquid-dsp: 4 - 256 - 512cassandra: Writesstress-ng: Vector Shuffleliquid-dsp: 1 - 256 - 57ncnn: Vulkan GPU-v3-v3 - mobilenet-v3ncnn: Vulkan GPU-v2-v2 - mobilenet-v2oidn: RTLightmap.hdr.4096x4096 - CPU-Onlyoidn: RT.ldr_alb_nrm.3840x2160 - CPU-Onlyoidn: RT.hdr_alb_nrm.3840x2160 - CPU-Onlyvkpeak: fp32-scalarab30.32721.298593.84824.29890374.23123.465125.830914462.0220.861016703.271039311.126.041028641.1956.591109812.44486078.689551.79764.2676929178.434590.53273.0535.497523699.702.667734974.265.1436779.86211.983017.51249.4553116.268316.412.0309979.1797375933.9935.7391.092604156.720.622182273.35550.47342.2113467917.3153.954.260284300.80224.5134.0498609112.971.493342978232.736.692.98157289.321422109.847310.05170492.1669.13212.4375.44948660.19669.4125.93119590000669.18099.821403450002485856.072652462.62162.16472058.7052.590436.021.0616479.916.554.1670984.71779450003267.3115.75611157.73705297.9617.348900311.039.785.6615017343.49383.49107.8412.544115.97502050.4365.95173.9055.3928.1530.56731333775.59743.83555.93638.3729.658568.0491244.12357190000.911.37399496309.760.94154.5973.5218202.852240.0369.559312666.1928.918597.700.74514622.6297.42996242.81525368.3912.739.222967.52843.97333.1812613348.9317318.73.7369855.67692379.81517950034.8167.015295991078.73144.5310.031372.60410.24651.180137.94892.693233.962.3527147314.853.82293218725.51196.78434651650047.716.292892.0885.239.49039.3977.93387362495.574.7883816344.214.6720244.56174.21831036429.9359.1708882130.94111.6829.653.6616.237379550014028000091.1145989191.308.300911.6777218009.8828555050090.965.425030.539835.39612.61205.606232625830500264222431.22434290005.116.670.060.120.12268.3388.58534.323158.23683.021248610.70170.014173.0451253523.6928.461372033.181398883.014.491377833.4843.091455171.86624014.1712140.60614.67931153192.755684.53335.9128.978638650.513.246142229.614.3143752.37179.077220.726411.171998.714319.32.3883835.8517434156.1031.0140.948694562.470.709986310.94485.93387.1512006039.1460.494.773318466.50249.9930.9759394851.371.36922737483.306.153.23207877.701533355.117866.4182936.6874.13198.4170.464629287.79715.5827.68112205000628.363910.411328300002360589.102519916.10154.25862164.2050.082934.331.01295383.715.803.99231026.931855400003134.0415.11637041.62733269.7916.689246224.429.435.4715526243.84396.03104.4312.950215.48486807.4968.00033.7905.5527.3829.76031298988.31762.86569.60939.3128.962269.6181249.62365230000.931.34467485922.630.96157.6923.4542206.769235.5170.821312444.3328.464096.260.73478322.9398.751009840.28518308.2012.569.342929.91143.41933.601246407317116.793.6962864.99322404.81502350034.4866.38524660.181088.75143.2410.121384.76610.3361.170139.04752.672232.172.3352148388.793.85033241408.21195.48584682500048.016.192908.8785.269.43639.1777.51958872508.634.7636812170.8614.6069245.64304.19981040683.1358.9499391131.40511.6429.753.6716.197362050013995000090.9080095190.878.283211.6977108509.8688566400090.865.356430.566435.42612.62205.448626025812500264382431.01434275005.116.670.060.120.12268.33OpenBenchmarking.org

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database with a variable number of concurrent repetitions -- up to the maximum number of CPU threads available. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 1ab20406080100SE +/- 0.86, N = 2SE +/- 0.19, N = 230.3388.591. (CC) gcc options: -O2 -lreadline -ltermcap -lz -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamba816243240SE +/- 0.13, N = 2SE +/- 0.07, N = 234.3221.30

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Streamba20406080100SE +/- 0.22, N = 2SE +/- 0.32, N = 258.2493.85

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: shufflenet-v2ba0.96531.93062.89593.86124.8265SE +/- 0.01, N = 2SE +/- 0.02, N = 23.024.29MIN: 2.73 / MAX: 18.89MIN: 3.99 / MAX: 24.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Redis 7.0.12 + memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 500 - Set To Get Ratio: 1:5ba300K600K900K1200K1500KSE +/- 25673.44, N = 2SE +/- 8374.42, N = 21248610.70890374.231. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database with a variable number of concurrent repetitions -- up to the maximum number of CPU threads available. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 2ab4080120160200SE +/- 24.93, N = 2SE +/- 1.42, N = 2123.47170.011. (CC) gcc options: -O2 -lreadline -ltermcap -lz -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 4ab4080120160200SE +/- 17.49, N = 2SE +/- 0.50, N = 2125.83173.051. (CC) gcc options: -O2 -lreadline -ltermcap -lz -lm

Redis 7.0.12 + memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 500 - Set To Get Ratio: 1:10ba300K600K900K1200K1500KSE +/- 63299.47, N = 2SE +/- 45279.14, N = 21253523.69914462.021. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: MMAPba714212835SE +/- 3.70, N = 2SE +/- 3.19, N = 228.4620.861. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Redis 7.0.12 + memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:5ba300K600K900K1200K1500KSE +/- 39391.55, N = 2SE +/- 44860.53, N = 21372033.181016703.271. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:5ba300K600K900K1200K1500KSE +/- 4729.88, N = 2SE +/- 43465.56, N = 21398883.011039311.121. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mnasnetba246810SE +/- 0.02, N = 2SE +/- 0.02, N = 24.496.04MIN: 4.03 / MAX: 20.38MIN: 5.41 / MAX: 26.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Redis 7.0.12 + memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10ba300K600K900K1200K1500KSE +/- 8107.22, N = 2SE +/- 568.99, N = 21377833.481028641.191. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: MEMFDab1326395265SE +/- 0.79, N = 2SE +/- 0.72, N = 256.5943.091. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Redis 7.0.12 + memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10ba300K600K900K1200K1500KSE +/- 18012.65, N = 2SE +/- 19181.49, N = 21455171.861109812.441. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Futexba130K260K390K520K650KSE +/- 57507.32, N = 2SE +/- 36419.89, N = 2624014.17486078.681. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Forkingba3K6K9K12K15KSE +/- 1330.94, N = 2SE +/- 301.28, N = 212140.609551.791. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Streamba160320480640800SE +/- 2.10, N = 2SE +/- 113.73, N = 2614.68764.27

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU Cacheba200K400K600K800K1000KSE +/- 138960.51, N = 2SE +/- 244998.38, N = 21153192.75929178.431. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Cryptoba12002400360048006000SE +/- 96.10, N = 2SE +/- 47.72, N = 25684.534590.531. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Zlibba70140210280350SE +/- 21.49, N = 2SE +/- 31.94, N = 2335.91273.051. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 1.6Encoder Mode: Preset 12 - Input: Bosphorus 4Kab816243240SE +/- 0.91, N = 2SE +/- 0.81, N = 235.5028.981. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Hashba140K280K420K560K700KSE +/- 38690.57, N = 2SE +/- 38337.11, N = 2638650.51523699.701. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Streamba0.73041.46082.19122.92163.652SE +/- 0.0188, N = 2SE +/- 0.3971, N = 23.24612.6677

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Pthreadba9K18K27K36K45KSE +/- 4375.87, N = 2SE +/- 4732.28, N = 242229.6134974.261. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v3-v3 - Model: mobilenet-v3ba1.15652.3133.46954.6265.7825SE +/- 0.04, N = 2SE +/- 0.80, N = 24.315.14MIN: 4.05 / MAX: 20.17MIN: 4.05 / MAX: 26.811. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: SENDFILEba9K18K27K36K45KSE +/- 1151.61, N = 2SE +/- 1669.87, N = 243752.3736779.861. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streamba50100150200250SE +/- 3.79, N = 2SE +/- 13.47, N = 2179.08211.98

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamba510152025SE +/- 3.12, N = 2SE +/- 2.36, N = 220.7317.51

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streamba3691215SE +/- 0.2364, N = 2SE +/- 0.6045, N = 211.17199.4553

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streamba306090120150SE +/- 14.87, N = 2SE +/- 15.66, N = 298.71116.27

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: AVL Treeba510152025SE +/- 0.15, N = 2SE +/- 0.05, N = 219.3016.411. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamba0.53741.07481.61222.14962.687SE +/- 0.0327, N = 2SE +/- 0.0085, N = 22.38832.0309

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streamba2004006008001000SE +/- 13.19, N = 2SE +/- 0.29, N = 2835.85979.18

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Mallocba90K180K270K360K450KSE +/- 46761.51, N = 2SE +/- 39637.02, N = 2434156.10375933.991. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 1.6Encoder Mode: Preset 8 - Input: Bosphorus 1080pab816243240SE +/- 0.87, N = 2SE +/- 2.58, N = 235.7431.011. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 4 - Input: Bosphorus 4Kab0.24570.49140.73710.98281.2285SE +/- 0.023, N = 2SE +/- 0.003, N = 21.0920.9481. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Mutexba150K300K450K600K750KSE +/- 20730.70, N = 2SE +/- 76155.40, N = 2694562.47604156.721. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OSPRay

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

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeba0.15970.31940.47910.63880.7985SE +/- 0.025213, N = 2SE +/- 0.014296, N = 20.7099860.622182

dav1d

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

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.2.1Video Input: Summer Nature 1080pba70140210280350SE +/- 2.55, N = 2SE +/- 4.15, N = 2310.94273.351. (CC) gcc options: -pthread -lm

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500ab120240360480600550.47485.93MAX: 3020.53MAX: 2859.52

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 100 - Sensor Count: 500ba80160240320400387.15342.21MAX: 2248.18MAX: 2349.14

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 100 - Sensor Count: 500ab3M6M9M12M15M13467917.3112006039.14

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: NUMAba1428425670SE +/- 5.27, N = 2SE +/- 5.63, N = 260.4953.951. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 1.6Encoder Mode: Preset 4 - Input: Bosphorus 1080pba1.07392.14783.22174.29565.3695SE +/- 0.109, N = 2SE +/- 0.495, N = 24.7734.2601. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Pollba70K140K210K280K350KSE +/- 23847.48, N = 2SE +/- 20478.34, N = 2318466.50284300.801. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Atomicba50100150200250SE +/- 11.52, N = 2SE +/- 10.78, N = 2249.99224.511. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 1.6Encoder Mode: Preset 13 - Input: Bosphorus 4Kab816243240SE +/- 2.10, N = 2SE +/- 2.71, N = 234.0530.981. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500ba2M4M6M8M10M9394851.378609112.97

OSPRay

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

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/ao/real_timeab0.3360.6721.0081.3441.68SE +/- 0.09686, N = 2SE +/- 0.02805, N = 21.493341.36920

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Fused Multiply-Addab600K1200K1800K2400K3000KSE +/- 63230.65, N = 2SE +/- 262560.86, N = 22978232.732737483.301. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v2-v2 - Model: mobilenet-v2ba246810SE +/- 0.07, N = 2SE +/- 0.70, N = 26.156.69MIN: 5.65 / MAX: 26.38MIN: 5.6 / MAX: 28.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamba0.72721.45442.18162.90883.636SE +/- 0.4116, N = 2SE +/- 0.0495, N = 23.23202.9815

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Floating Pointba2K4K6K8K10KSE +/- 310.83, N = 2SE +/- 364.11, N = 27877.707289.321. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Pipeba300K600K900K1200K1500KSE +/- 131648.17, N = 2SE +/- 30380.48, N = 21533355.111422109.841. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU Stressba2K4K6K8K10KSE +/- 158.65, N = 2SE +/- 567.82, N = 27866.407310.051. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: IO_uringba40K80K120K160K200KSE +/- 499.46, N = 2SE +/- 2497.26, N = 2182936.68170492.161. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Glibc Qsort Data Sortingba1632486480SE +/- 3.74, N = 2SE +/- 5.90, N = 274.1369.131. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200ab50100150200250212.43198.41MAX: 2080.14MAX: 2427.81

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500ab2040608010075.4070.46MAX: 1528.52MAX: 1429.13

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500ab1.1M2.2M3.3M4.4M5.5M4948660.194629287.79

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Cloningba150300450600750SE +/- 2.47, N = 2SE +/- 28.62, N = 2715.58669.411. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mobilenetab714212835SE +/- 0.19, N = 2SE +/- 0.03, N = 225.9327.68MIN: 23.82 / MAX: 45.07MIN: 24.92 / MAX: 47.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 4 - Buffer Length: 256 - Filter Length: 57ab30M60M90M120M150MSE +/- 6360000.00, N = 2SE +/- 3395000.00, N = 21195900001122050001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Streamba140280420560700SE +/- 79.47, N = 2SE +/- 12.84, N = 2628.36669.18

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: regnety_400mab3691215SE +/- 0.01, N = 2SE +/- 0.07, N = 29.8210.41MIN: 9.26 / MAX: 25.33MIN: 9.74 / MAX: 26.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 8 - Buffer Length: 256 - Filter Length: 57ab30M60M90M120M150MSE +/- 5615000.00, N = 2SE +/- 7270000.00, N = 21403450001328300001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Glibc C String Functionsab500K1000K1500K2000K2500KSE +/- 92139.37, N = 2SE +/- 158119.57, N = 22485856.072360589.101. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: System V Message Passingab600K1200K1800K2400K3000KSE +/- 119752.37, N = 2SE +/- 200089.21, N = 22652462.622519916.101. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamba4080120160200SE +/- 1.05, N = 2SE +/- 21.74, N = 2154.26162.16

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Function Callba5001000150020002500SE +/- 40.41, N = 2SE +/- 96.37, N = 22164.202058.701. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

OSPRay

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

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/pathtracer/real_timeab1224364860SE +/- 0.59, N = 2SE +/- 0.02, N = 252.5950.08

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet50ba816243240SE +/- 1.96, N = 2SE +/- 0.05, N = 234.3336.02MIN: 30.77 / MAX: 59.68MIN: 31.18 / MAX: 61.61. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OSPRay

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

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeab0.23890.47780.71670.95561.1945SE +/- 0.011835, N = 2SE +/- 0.016867, N = 21.0616401.012953

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 128ba20406080100SE +/- 4.10, N = 2SE +/- 4.60, N = 283.779.91. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: squeezenet_ssdba48121620SE +/- 0.37, N = 2SE +/- 0.30, N = 215.8016.55MIN: 13.81 / MAX: 36.58MIN: 15.48 / MAX: 37.441. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 4.1Binary: Pathtracer ISPC - Model: Asian Dragon Objab0.93761.87522.81283.75044.688SE +/- 0.0272, N = 2SE +/- 0.1882, N = 24.16703.9923MIN: 3.5 / MAX: 5.07MIN: 3.51 / MAX: 5.03

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500ba20040060080010001026.93984.70MAX: 6211.26MAX: 4944.15

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 8 - Buffer Length: 256 - Filter Length: 32ba40M80M120M160M200MSE +/- 5480000.00, N = 2SE +/- 11475000.00, N = 21855400001779450001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: x86_64 RdRandab7001400210028003500SE +/- 13.37, N = 2SE +/- 68.45, N = 23267.313134.041. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200ab4812162015.7515.11MAX: 1425.54MAX: 1458.64

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500ba140K280K420K560K700K637041.62611157.73

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Context Switchingba160K320K480K640K800KSE +/- 16234.88, N = 2SE +/- 53162.51, N = 2733269.79705297.961. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200ab4812162017.3416.68MAX: 866.08MAX: 1251.48

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200ba2M4M6M8M10M9246224.428900311.03

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: efficientnet-b0ba3691215SE +/- 0.02, N = 2SE +/- 0.65, N = 29.439.78MIN: 8.62 / MAX: 26.5MIN: 8.38 / MAX: 32.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: FastestDetba1.27352.5473.82055.0946.3675SE +/- 0.04, N = 2SE +/- 0.20, N = 25.475.66MIN: 5.21 / MAX: 24.93MIN: 5.23 / MAX: 24.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 100 - Sensor Count: 200ba3M6M9M12M15M15526243.8415017343.49

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix 3D Mathba90180270360450SE +/- 6.17, N = 2SE +/- 5.36, N = 2396.03383.491. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 100 - Sensor Count: 200ab20406080100107.84104.43MAX: 1809.87MAX: 2007.51

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Streamba3691215SE +/- 0.10, N = 2SE +/- 1.68, N = 212.9512.54

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: squeezenet_ssdba48121620SE +/- 0.84, N = 2SE +/- 0.47, N = 215.4815.97MIN: 13.66 / MAX: 36.39MIN: 13.93 / MAX: 35.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:10ab110K220K330K440K550KSE +/- 9547.84, N = 2SE +/- 4490.17, N = 2502050.43486807.491. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamab1530456075SE +/- 6.09, N = 2SE +/- 7.55, N = 265.9568.00

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 1.9Video Input: Bosphorus 1080p - Video Preset: Fastab0.87861.75722.63583.51444.393SE +/- 0.015, N = 2SE +/- 0.042, N = 23.9053.7901. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: FastestDetab1.24882.49763.74644.99526.244SE +/- 0.04, N = 2SE +/- 0.16, N = 25.395.55MIN: 5.09 / MAX: 24.31MIN: 5.23 / MAX: 25.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mobilenetba714212835SE +/- 0.57, N = 2SE +/- 0.19, N = 227.3828.15MIN: 24.23 / MAX: 47.57MIN: 26.87 / MAX: 48.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Streamab714212835SE +/- 2.83, N = 2SE +/- 3.30, N = 230.5729.76

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500ab300K600K900K1200K1500K1333775.591298988.31

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Floating Pointba160320480640800SE +/- 25.67, N = 2SE +/- 25.00, N = 2762.86743.831. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Build2

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.15Time To Compileab120240360480600SE +/- 1.14, N = 2SE +/- 4.22, N = 2555.94569.61

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: yolov4-tinyab918273645SE +/- 0.08, N = 2SE +/- 0.83, N = 238.3739.31MIN: 36.54 / MAX: 54.65MIN: 36.44 / MAX: 59.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamab714212835SE +/- 2.87, N = 2SE +/- 2.67, N = 229.6628.96

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streamab1530456075SE +/- 6.60, N = 2SE +/- 6.42, N = 268.0569.62

dav1d

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

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.2.1Video Input: Chimera 1080pba50100150200250SE +/- 1.95, N = 2SE +/- 2.76, N = 2249.62244.121. (CC) gcc options: -pthread -lm

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 8 - Buffer Length: 256 - Filter Length: 512ba8M16M24M32M40MSE +/- 2570000.00, N = 2SE +/- 3468000.00, N = 236523000357190001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: blazefaceab0.20930.41860.62790.83721.0465SE +/- 0.01, N = 2SE +/- 0.00, N = 20.910.93MIN: 0.78 / MAX: 3.03MIN: 0.8 / MAX: 3.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OSPRay

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

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/scivis/real_timeab0.30910.61820.92731.23641.5455SE +/- 0.00084, N = 2SE +/- 0.00287, N = 21.373991.34467

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100ab110K220K330K440K550KSE +/- 10780.90, N = 2SE +/- 14781.65, N = 2496309.76485922.631. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: blazefaceab0.2160.4320.6480.8641.08SE +/- 0.00, N = 2SE +/- 0.01, N = 20.940.96MIN: 0.86 / MAX: 3.07MIN: 0.8 / MAX: 3.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 1.6Encoder Mode: Preset 12 - Input: Bosphorus 1080pba306090120150SE +/- 0.70, N = 2SE +/- 0.24, N = 2157.69154.601. (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 4.1Binary: Pathtracer - Model: Crownab0.79241.58482.37723.16963.962SE +/- 0.0068, N = 2SE +/- 0.0668, N = 23.52183.4542MIN: 2.81 / MAX: 4.3MIN: 2.82 / MAX: 4.4

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 1.6Encoder Mode: Preset 13 - Input: Bosphorus 1080pba50100150200250SE +/- 0.84, N = 2SE +/- 0.66, N = 2206.77202.851. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vision_transformerba50100150200250SE +/- 3.18, N = 2SE +/- 3.20, N = 2235.51240.03MIN: 187.15 / MAX: 291.28MIN: 196.33 / MAX: 300.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Streamab1632486480SE +/- 5.54, N = 2SE +/- 6.64, N = 269.5670.82

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Mathab3K6K9K12K15KSE +/- 425.64, N = 2SE +/- 913.67, N = 212666.1912444.331. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Streamab714212835SE +/- 2.30, N = 2SE +/- 2.67, N = 228.9228.46

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: vgg16ba20406080100SE +/- 0.11, N = 2SE +/- 0.23, N = 296.2697.70MIN: 93.43 / MAX: 114.68MIN: 94.39 / MAX: 117.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OSPRay

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

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timeab0.16770.33540.50310.67080.8385SE +/- 0.009807, N = 2SE +/- 0.008961, N = 20.7451460.734783

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 1 - Sensor Count: 200ba51015202522.9322.62MAX: 1216.89MAX: 1174.88

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vgg16ab20406080100SE +/- 0.26, N = 2SE +/- 2.34, N = 297.4298.75MIN: 94.57 / MAX: 115.91MIN: 90.66 / MAX: 1242.521. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 1 - Sensor Count: 500ba200K400K600K800K1000K1009840.28996242.81

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:5ab110K220K330K440K550KSE +/- 12286.93, N = 2SE +/- 13966.62, N = 2525368.39518308.201. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet18ba3691215SE +/- 0.04, N = 2SE +/- 0.04, N = 212.5612.73MIN: 11.83 / MAX: 29.03MIN: 11.94 / MAX: 28.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: efficientnet-b0ab3691215SE +/- 0.07, N = 2SE +/- 0.02, N = 29.229.34MIN: 8.52 / MAX: 24.96MIN: 8.61 / MAX: 25.621. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix Makefilesba6001200180024003000SE +/- 7.61, N = 2SE +/- 45.61, N = 22929.912967.53

Z3 Theorem Prover

The Z3 Theorem Prover / SMT solver is developed by Microsoft Research under the MIT license. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterZ3 Theorem Prover 4.12.1SMT File: 1.smt2ba1020304050SE +/- 0.18, N = 2SE +/- 0.05, N = 243.4243.971. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: resnet50ab816243240SE +/- 0.55, N = 2SE +/- 1.36, N = 233.1833.60MIN: 31.12 / MAX: 58.5MIN: 30.7 / MAX: 59.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200ab3M6M9M12M15M12613348.9312464073.00

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix Mathab4K8K12K16K20KSE +/- 1025.15, N = 2SE +/- 854.63, N = 217318.7017116.791. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 4.1Binary: Pathtracer ISPC - Model: Crownab0.84081.68162.52243.36324.204SE +/- 0.0660, N = 2SE +/- 0.1040, N = 23.73693.6962MIN: 3.08 / MAX: 4.7MIN: 3.06 / MAX: 4.78

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamab2004006008001000SE +/- 75.66, N = 2SE +/- 86.04, N = 2855.68864.99

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.30ba5001000150020002500SE +/- 234.25, N = 2SE +/- 210.20, N = 22404.82379.81. (CXX) g++ options: -O3 -march=native -fPIE -pie

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 2 - Buffer Length: 256 - Filter Length: 512ab3M6M9M12M15MSE +/- 158500.00, N = 2SE +/- 149500.00, N = 215179500150235001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 1 - Sensor Count: 500ab81624324034.8134.48MAX: 1426.89MAX: 1495.07

dav1d

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

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.2.1Video Input: Summer Nature 4Kab1530456075SE +/- 0.31, N = 2SE +/- 2.35, N = 267.0166.381. (CC) gcc options: -pthread -lm

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 100 - Batch Size Per Write: 1 - Sensor Count: 200ab110K220K330K440K550K529599.00524660.18

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Memory Copyingba2004006008001000SE +/- 40.05, N = 2SE +/- 40.86, N = 21088.751078.731. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200ab306090120150144.53143.24MAX: 1992.27MAX: 2073.51

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: regnety_400mab3691215SE +/- 0.14, N = 2SE +/- 0.04, N = 210.0310.12MIN: 8.93 / MAX: 25.51MIN: 9.48 / MAX: 25.781. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To Compileab30060090012001500SE +/- 0.29, N = 2SE +/- 12.96, N = 21372.601384.77

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamab3691215SE +/- 1.04, N = 2SE +/- 1.15, N = 210.2510.34

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 1.9Video Input: Bosphorus 4K - Video Preset: Fastab0.26550.5310.79651.0621.3275SE +/- 0.013, N = 2SE +/- 0.013, N = 21.1801.1701. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamab306090120150SE +/- 15.16, N = 2SE +/- 17.32, N = 2137.95139.05

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 1.9Video Input: Bosphorus 4K - Video Preset: Fasterab0.60591.21181.81772.42363.0295SE +/- 0.052, N = 2SE +/- 0.045, N = 22.6932.6721. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vision_transformerba50100150200250SE +/- 5.04, N = 2SE +/- 0.17, N = 2232.17233.96MIN: 187.55 / MAX: 291.61MIN: 196.52 / MAX: 292.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streamab0.52941.05881.58822.11762.647SE +/- 0.2053, N = 2SE +/- 0.2323, N = 22.35272.3352

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Wide Vector Mathba30K60K90K120K150KSE +/- 3329.63, N = 2SE +/- 2993.76, N = 2148388.79147314.851. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 4.1Binary: Pathtracer - Model: Asian Dragon Objba0.86631.73262.59893.46524.3315SE +/- 0.0122, N = 2SE +/- 0.0521, N = 23.85033.8229MIN: 3.22 / MAX: 4.83MIN: 3.19 / MAX: 4.76

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Semaphoresba700K1400K2100K2800K3500KSE +/- 70160.30, N = 2SE +/- 287315.12, N = 23241408.213218725.511. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Streamab4080120160200SE +/- 20.06, N = 2SE +/- 21.75, N = 2196.78195.49

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 32ba10M20M30M40M50MSE +/- 77000.00, N = 2SE +/- 451500.00, N = 246825000465165001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 32ba1122334455SE +/- 0.55, N = 2SE +/- 0.35, N = 248.047.71. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: googlenetba48121620SE +/- 0.01, N = 2SE +/- 0.00, N = 216.1916.29MIN: 14.91 / MAX: 32.9MIN: 14.94 / MAX: 33.611. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Ninjaab6001200180024003000SE +/- 4.75, N = 2SE +/- 4.95, N = 22892.092908.88

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: mnasnetab1.18352.3673.55054.7345.9175SE +/- 0.81, N = 2SE +/- 0.78, N = 25.235.26MIN: 4.05 / MAX: 26.68MIN: 4.06 / MAX: 22.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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 1.6Encoder Mode: Preset 8 - Input: Bosphorus 4Kab3691215SE +/- 0.307, N = 2SE +/- 1.160, N = 29.4909.4361. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: yolov4-tinyba918273645SE +/- 1.11, N = 2SE +/- 1.02, N = 239.1739.39MIN: 36.39 / MAX: 59.83MIN: 36.4 / MAX: 60.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game for this open-source first person shooter title. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.6Resolution: 1920 x 1080 - Effects Quality: Ultraab20406080100SE +/- 0.16, N = 2SE +/- 0.41, N = 277.9377.52MIN: 34 / MAX: 120MIN: 32 / MAX: 120

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Socket Activityba5001000150020002500SE +/- 271.08, N = 2SE +/- 249.85, N = 22508.632495.571. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

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 4.1Binary: Pathtracer ISPC - Model: Asian Dragonab1.07742.15483.23224.30965.387SE +/- 0.0697, N = 2SE +/- 0.0364, N = 24.78834.7636MIN: 4.05 / MAX: 5.89MIN: 4.05 / MAX: 5.98

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200ab200K400K600K800K1000K816344.20812170.86

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Streamab48121620SE +/- 1.61, N = 2SE +/- 1.82, N = 214.6714.61

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streamab50100150200250SE +/- 31.56, N = 2SE +/- 32.45, N = 2244.56245.64

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 4.1Binary: Pathtracer - Model: Asian Dragonab0.94911.89822.84733.79644.7455SE +/- 0.0414, N = 2SE +/- 0.0239, N = 24.21834.1998MIN: 3.54 / MAX: 5.31MIN: 3.54 / MAX: 5.43

Apache IoTDB

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200ba200K400K600K800K1000K1040683.131036429.93

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game for this open-source first person shooter title. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.6Resolution: 1920 x 1080 - Effects Quality: Ultimateab1326395265SE +/- 0.07, N = 2SE +/- 0.06, N = 259.1758.95MIN: 24 / MAX: 94MIN: 24 / MAX: 94

Z3 Theorem Prover

The Z3 Theorem Prover / SMT solver is developed by Microsoft Research under the MIT license. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterZ3 Theorem Prover 4.12.1SMT File: 2.smt2ab306090120150SE +/- 0.11, N = 2SE +/- 0.05, N = 2130.94131.411. (CXX) g++ options: -lpthread -std=c++17 -fvisibility=hidden -mfpmath=sse -msse -msse2 -O3 -fPIC

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: alexnetba3691215SE +/- 0.07, N = 2SE +/- 0.06, N = 211.6411.68MIN: 10.99 / MAX: 27.75MIN: 10.92 / MAX: 27.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Apache IoTDB

OpenBenchmarking.orgAverage Latency, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500ba71421283529.7529.65MAX: 1271.26MAX: 1193.78

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: shufflenet-v2ab0.82581.65162.47743.30324.129SE +/- 0.63, N = 2SE +/- 0.62, N = 23.663.67MIN: 2.74 / MAX: 16.28MIN: 2.75 / MAX: 19.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: googlenetba48121620SE +/- 0.10, N = 2SE +/- 0.08, N = 216.1916.23MIN: 14.97 / MAX: 32.25MIN: 15.01 / MAX: 32.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 2 - Buffer Length: 256 - Filter Length: 57ab16M32M48M64M80MSE +/- 128500.00, N = 2SE +/- 620500.00, N = 273795500736205001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 4 - Buffer Length: 256 - Filter Length: 32ab30M60M90M120M150MSE +/- 2650000.00, N = 2SE +/- 2610000.00, N = 21402800001399500001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game for this open-source first person shooter title. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.6Resolution: 1920 x 1080 - Effects Quality: Highab20406080100SE +/- 0.04, N = 2SE +/- 0.32, N = 291.1190.91MIN: 39 / MAX: 130MIN: 38 / MAX: 130

dav1d

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

OpenBenchmarking.orgFPS, More Is Betterdav1d 1.2.1Video Input: Chimera 1080p 10-bitab4080120160200SE +/- 19.41, N = 2SE +/- 12.81, N = 2191.30190.871. (CC) gcc options: -pthread -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streamab246810SE +/- 1.0816, N = 2SE +/- 1.0939, N = 28.30098.2832

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU - Model: alexnetab3691215SE +/- 0.03, N = 2SE +/- 0.04, N = 211.6711.69MIN: 11.06 / MAX: 27.75MIN: 10.95 / MAX: 28.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 512ab1.7M3.4M5.1M6.8M8.5MSE +/- 8100.00, N = 2SE +/- 20850.00, N = 2772180077108501. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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 1.9Video Input: Bosphorus 1080p - Video Preset: Fasterab3691215SE +/- 0.090, N = 2SE +/- 0.603, N = 29.8829.8681. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 2 - Buffer Length: 256 - Filter Length: 32ba20M40M60M80M100MSE +/- 1161000.00, N = 2SE +/- 873500.00, N = 285664000855505001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

libxsmm

Libxsmm is an open-source library for specialized dense and sparse matrix operations and deep learning primitives. Libxsmm supports making use of Intel AMX, AVX-512, and other modern CPU instruction set capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS/s, More Is Betterlibxsmm 2-1.17-3645M N K: 64ab20406080100SE +/- 4.45, N = 2SE +/- 4.80, N = 290.990.81. (CXX) g++ options: -dynamic -Bstatic -static-libgcc -lgomp -lm -lrt -ldl -lquadmath -lstdc++ -pthread -fPIC -std=c++14 -O2 -fopenmp-simd -funroll-loops -ftree-vectorize -fdata-sections -ffunction-sections -fvisibility=hidden -march=core-avx2

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamab1530456075SE +/- 0.37, N = 2SE +/- 0.29, N = 265.4365.36

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Streamab714212835SE +/- 0.17, N = 2SE +/- 0.13, N = 230.5430.57

Opus Codec Encoding

Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus five times. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.4WAV To Opus Encodeab816243240SE +/- 0.10, N = 2SE +/- 0.06, N = 235.4035.431. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet18ab3691215SE +/- 0.04, N = 2SE +/- 0.11, N = 212.6112.62MIN: 11.89 / MAX: 29.06MIN: 11.83 / MAX: 28.81. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game for this open-source first person shooter title. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.6Resolution: 1920 x 1080 - Effects Quality: Lowab50100150200250SE +/- 0.72, N = 2SE +/- 0.60, N = 2205.61205.45MIN: 99 / MAX: 336MIN: 95 / MAX: 334

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 4 - Buffer Length: 256 - Filter Length: 512ab6M12M18M24M30MSE +/- 466500.00, N = 2SE +/- 478500.00, N = 225830500258125001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Apache Cassandra

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

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 4.1.3Test: Writesba6K12K18K24K30KSE +/- 71.50, N = 2SE +/- 149.00, N = 22643826422

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Shuffleab5001000150020002500SE +/- 33.39, N = 2SE +/- 17.07, N = 22431.222431.011. (CXX) g++ options: -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lpthread -lrt -lsctp -lz

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 57ab9M18M27M36M45MSE +/- 225000.00, N = 2SE +/- 154500.00, N = 243429000434275001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3ab1.14982.29963.44944.59925.749SE +/- 0.77, N = 2SE +/- 0.80, N = 25.115.11MIN: 4.07 / MAX: 20.14MIN: 4.05 / MAX: 26.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2ab246810SE +/- 0.67, N = 2SE +/- 0.65, N = 26.676.67MIN: 5.54 / MAX: 27.83MIN: 5.58 / MAX: 27.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Intel Open Image Denoise

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

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.0Run: RTLightmap.hdr.4096x4096 - Device: CPU-Onlyba0.01350.0270.04050.0540.0675SE +/- 0.00, N = 2SE +/- 0.00, N = 20.060.06

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.0Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Onlyba0.0270.0540.0810.1080.135SE +/- 0.00, N = 2SE +/- 0.00, N = 20.120.12

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.0Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Onlyba0.0270.0540.0810.1080.135SE +/- 0.00, N = 2SE +/- 0.00, N = 20.120.12

vkpeak

Vkpeak is a Vulkan compute benchmark inspired by OpenCL's clpeak. Vkpeak provides Vulkan compute performance measurements for FP16 / FP32 / FP64 / INT16 / INT32 scalar and vec4 performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is Bettervkpeak 20230730fp32-scalarba60120180240300SE +/- 0.01, N = 2SE +/- 0.00, N = 2268.33268.33

197 Results Shown

SQLite
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
NCNN
Redis 7.0.12 + memtier_benchmark
SQLite:
  2
  4
Redis 7.0.12 + memtier_benchmark
Stress-NG
Redis 7.0.12 + memtier_benchmark:
  Redis - 100 - 1:5
  Redis - 50 - 1:5
NCNN
Redis 7.0.12 + memtier_benchmark
Stress-NG
Redis 7.0.12 + memtier_benchmark
Stress-NG:
  Futex
  Forking
Neural Magic DeepSparse
Stress-NG:
  CPU Cache
  Crypto
  Zlib
SVT-AV1
Stress-NG
Neural Magic DeepSparse
Stress-NG
NCNN
Stress-NG
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream
Stress-NG
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
Stress-NG
SVT-AV1:
  Preset 8 - Bosphorus 1080p
  Preset 4 - Bosphorus 4K
Stress-NG
OSPRay
dav1d
Apache IoTDB:
  200 - 100 - 500
  100 - 100 - 500
  100 - 100 - 500
Stress-NG
SVT-AV1
Stress-NG:
  Poll
  Atomic
SVT-AV1
Apache IoTDB
OSPRay
Stress-NG
NCNN
Neural Magic DeepSparse
Stress-NG:
  Vector Floating Point
  Pipe
  CPU Stress
  IO_uring
  Glibc Qsort Data Sorting
Apache IoTDB:
  500 - 100 - 200
  500 - 1 - 500
  500 - 100 - 500
Stress-NG
NCNN
Liquid-DSP
Neural Magic DeepSparse
NCNN
Liquid-DSP
Stress-NG:
  Glibc C String Functions
  System V Message Passing
Neural Magic DeepSparse
Stress-NG
OSPRay
NCNN
OSPRay
libxsmm
NCNN
Embree
Apache IoTDB
Liquid-DSP
Stress-NG
Apache IoTDB:
  500 - 1 - 200
  500 - 1 - 500
Stress-NG
Apache IoTDB:
  200 - 1 - 200
  500 - 100 - 200
NCNN:
  CPU - efficientnet-b0
  Vulkan GPU - FastestDet
Apache IoTDB
Stress-NG
Apache IoTDB
Neural Magic DeepSparse
NCNN
Memcached
Neural Magic DeepSparse
VVenC
NCNN:
  CPU - FastestDet
  Vulkan GPU - mobilenet
Neural Magic DeepSparse
Apache IoTDB
Stress-NG
Build2
NCNN
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    items/sec
    ms/batch
dav1d
Liquid-DSP
NCNN
OSPRay
Memcached
NCNN
SVT-AV1
Embree
SVT-AV1
NCNN
Neural Magic DeepSparse
Stress-NG
Neural Magic DeepSparse
NCNN
OSPRay
Apache IoTDB
NCNN
Apache IoTDB
Memcached
NCNN:
  Vulkan GPU - resnet18
  Vulkan GPU - efficientnet-b0
Timed LLVM Compilation
Z3 Theorem Prover
NCNN
Apache IoTDB
Stress-NG
Embree
Neural Magic DeepSparse
QuantLib
Liquid-DSP
Apache IoTDB
dav1d
Apache IoTDB
Stress-NG
Apache IoTDB
NCNN
Timed Godot Game Engine Compilation
Neural Magic DeepSparse
VVenC
Neural Magic DeepSparse
VVenC
NCNN
Neural Magic DeepSparse
Stress-NG
Embree
Stress-NG
Neural Magic DeepSparse
Liquid-DSP
libxsmm
NCNN
Timed LLVM Compilation
NCNN
SVT-AV1
NCNN
Xonotic
Stress-NG
Embree
Apache IoTDB
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream
Embree
Apache IoTDB
Xonotic
Z3 Theorem Prover
NCNN
Apache IoTDB
NCNN:
  Vulkan GPU - shufflenet-v2
  CPU - googlenet
Liquid-DSP:
  2 - 256 - 57
  4 - 256 - 32
Xonotic
dav1d
Neural Magic DeepSparse
NCNN
Liquid-DSP
VVenC
Liquid-DSP
libxsmm
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream:
    items/sec
    ms/batch
Opus Codec Encoding
NCNN
Xonotic
Liquid-DSP
Apache Cassandra
Stress-NG
Liquid-DSP
NCNN:
  Vulkan GPU-v3-v3 - mobilenet-v3
  Vulkan GPU-v2-v2 - mobilenet-v2
Intel Open Image Denoise:
  RTLightmap.hdr.4096x4096 - CPU-Only
  RT.ldr_alb_nrm.3840x2160 - CPU-Only
  RT.hdr_alb_nrm.3840x2160 - CPU-Only
vkpeak