Retbleed Call Depth Tracking Skylake

Call Depth Tracking mitigation benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2212262-NE-RETBLEEDS60
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:

Audio Encoding 2 Tests
AV1 3 Tests
C++ Boost Tests 2 Tests
Web Browsers 1 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 9 Tests
CPU Massive 21 Tests
Creator Workloads 16 Tests
Cryptocurrency Benchmarks, CPU Mining Tests 2 Tests
Cryptography 3 Tests
Database Test Suite 11 Tests
Encoding 5 Tests
Game Development 2 Tests
HPC - High Performance Computing 10 Tests
Imaging 6 Tests
Java 2 Tests
Common Kernel Benchmarks 7 Tests
Machine Learning 6 Tests
Multi-Core 18 Tests
NVIDIA GPU Compute 2 Tests
Intel oneAPI 3 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 5 Tests
Python Tests 8 Tests
Renderers 2 Tests
Software Defined Radio 2 Tests
Server 12 Tests
Server CPU Tests 15 Tests
Single-Threaded 3 Tests
Video Encoding 3 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Default - IBRS
December 20 2022
  1 Day, 20 Hours, 11 Minutes
retbleed=stuff
December 22 2022
  1 Day, 21 Hours, 58 Minutes
mitigations=off
December 23 2022
  1 Day, 19 Hours, 48 Minutes
retbleed=off
December 25 2022
  1 Day, 19 Hours, 29 Minutes
Invert Hiding All Results Option
  1 Day, 20 Hours, 21 Minutes

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

Default - IBRS

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

retbleed=stuff

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Stuffing + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

mitigations=off

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + mmio_stale_data: Vulnerable + retbleed: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Not affected + srbds: Vulnerable + tsx_async_abort: Mitigation of TSX disabled

retbleed=off

Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (4 Cores / 8 Threads), Motherboard: MSI Z170A SLI PLUS (MS-7998) v1.0 (2.A0 BIOS), Chipset: Intel Xeon E3-1200 v5/E3-1500, Memory: 32GB, Disk: 256GB TOSHIBA RD400, Graphics: ASUS AMD Radeon HD 7850 / R7 265 R9 270 1024SP, Audio: Realtek ALC1150, Monitor: VA2431, Network: Intel I219-V

OS: Ubuntu 20.04, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 3.36.9, Display Server: X Server 1.20.13, OpenGL: 4.5 Mesa 21.2.6 (LLVM 12.0.0), Compiler: GCC 9.4.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Vulnerable + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

Retbleed Call Depth Tracking SkylakeOpenBenchmarking.orgPhoronix Test SuiteIntel Xeon E3-1280 v5 @ 4.00GHz (4 Cores / 8 Threads)MSI Z170A SLI PLUS (MS-7998) v1.0 (2.A0 BIOS)Intel Xeon E3-1200 v5/E3-150032GB256GB TOSHIBA RD400ASUS AMD Radeon HD 7850 / R7 265 R9 270 1024SPRealtek ALC1150VA2431Intel I219-VUbuntu 20.046.1.0-phx (x86_64)GNOME Shell 3.36.9X Server 1.20.134.5 Mesa 21.2.6 (LLVM 12.0.0)GCC 9.4.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionRetbleed Call Depth Tracking Skylake BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - NONE / errors=remount-ro,relatime,rw / Block Size: 4096- Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1 - OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)- Python 3.8.10- Default - IBRS: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled - retbleed=stuff: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Stuffing + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled - mitigations=off: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + mmio_stale_data: Vulnerable + retbleed: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Not affected + srbds: Vulnerable + tsx_async_abort: Mitigation of TSX disabled - retbleed=off: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Vulnerable + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

Default - IBRSretbleed=stuffmitigations=offretbleed=offLogarithmic Result OverviewPhoronix Test SuiteHackbenchctx_clockSockperfPostMarkKeyDBNatronStress-NGOSBenchMemcachedEnCodecnginxJPEG XL Decoding libjxlFacebook RocksDBJPEG XL libjxlDragonflydbMobile Neural Networkmemtier_benchmarkTimed Erlang/OTP CompilationTimed Linux Kernel CompilationGraphicsMagickRenaissancePostgreSQLNCNNClickHouseTimed PHP CompilationMariaDBRedisDaCapo BenchmarkCockroachDBSeleniumTimed Node.js CompilationUnpacking The Linux KernelApache SparkTimed CPython Compilationlibavif avifencLuaRadioSVT-AV1OpenRadioss7-Zip CompressiononeDNNFLAC Audio EncodingNumenta Anomaly Benchmarkrav1eOpenFOAMWebP2 Image EncodeNeural Magic DeepSparseBRL-CADWebP Image EncodeStargate Digital Audio WorkstationCpuminer-OptOpenVINOminiBUDEsrsRANBlenderSMHashernekRSXmrigOpenVKL

Retbleed Call Depth Tracking Skylakestress-ng: x86_64 RdRandstress-ng: System V Message Passingstress-ng: Mallocstress-ng: Mutexhackbench: 8 - Processhackbench: 8 - Threadhackbench: 4 - Processhackbench: 16 - Processhackbench: 16 - Threadctx-clock: Context Switch Timestress-ng: Socket Activitypostmark: Disk Transaction Performancekeydb: stress-ng: Futexstress-ng: Context Switchingosbench: Create Filessockperf: Latency Ping Pongmemcached: 1:10stress-ng: SENDFILEstress-ng: MEMFDmemcached: 1:5osbench: Create Threadsmemcached: 1:1osbench: Launch Programsgraphics-magick: Rotatememcached: 5:1renaissance: Genetic Algorithm Using Jenetics + Futuresosbench: Create Processesrocksdb: Seq Fillencodec: 24 kbpsencodec: 1.5 kbpsencodec: 6 kbpsstress-ng: MMAPencodec: 3 kbpsrocksdb: Rand Fillstress-ng: Forkinggraphics-magick: HWB Color Spacencnn: CPU - blazefaceosbench: Memory Allocationsncnn: CPU - regnety_400mrocksdb: Update Randnginx: 20renaissance: Apache Spark ALSnginx: 100renaissance: Akka Unbalanced Cobwebbed Treemnn: mobilenetV3pgbench: 1 - 50 - Read Onlypgbench: 1 - 50 - Read Only - Average Latencynginx: 200dragonflydb: 50 - 1:5nginx: 500pgbench: 100 - 100 - Read Only - Average Latencynginx: 4000pgbench: 100 - 100 - Read Onlyrenaissance: Savina Reactors.IOnginx: 1000pgbench: 100 - 50 - Read Only - Average Latencypgbench: 100 - 50 - Read Onlyjpegxl-decode: Alljpegxl: JPEG - 80pgbench: 1 - 100 - Read Onlyrenaissance: ALS Movie Lenspgbench: 1 - 100 - Read Only - Average Latencyjpegxl: JPEG - 90jpegxl: PNG - 80jpegxl-decode: 1jpegxl: PNG - 90rocksdb: Read Rand Write Randmnn: nasnetdragonflydb: 50 - 5:1mnn: squeezenetv1.1mnn: resnet-v2-50stress-ng: NUMArenaissance: In-Memory Database Shootoutmemtier-benchmark: Redis - 50 - 1:10pgbench: 100 - 100 - Read Write - Average Latencypgbench: 100 - 100 - Read Writerenaissance: Finagle HTTP Requestsncnn: CPU - shufflenet-v2dragonflydb: 200 - 1:1ncnn: CPU - FastestDetpgbench: 100 - 50 - Read Writepgbench: 100 - 50 - Read Write - Average Latencybuild-erlang: Time To Compiledragonflydb: 200 - 1:5build-linux-kernel: defconfigspark: 1000000 - 500 - SHA-512 Benchmark Timememtier-benchmark: Redis - 100 - 1:10pgbench: 1 - 1 - Read Onlymnn: MobileNetV2_224redis: GET - 50ncnn: CPU - mnasnetncnn: CPU-v3-v3 - mobilenet-v3mnn: SqueezeNetV1.0pgbench: 1 - 1 - Read Only - Average Latencydragonflydb: 200 - 5:1clickhouse: 100M Rows Web Analytics Dataset, Second Runspark: 1000000 - 500 - Repartition Test Timepgbench: 1 - 100 - Read Writemnn: inception-v3stress-ng: Semaphorespgbench: 1 - 100 - Read Write - Average Latencyspark: 1000000 - 500 - Group By Test Timespark: 1000000 - 100 - Group By Test Timencnn: CPU-v2-v2 - mobilenet-v2clickhouse: 100M Rows Web Analytics Dataset, First Run / Cold Cacheavifenc: 10, Losslesscockroach: KV, 95% Reads - 128cockroach: KV, 50% Reads - 128cockroach: KV, 60% Reads - 128cockroach: KV, 95% Reads - 256renaissance: Scala Dottymemtier-benchmark: Redis - 100 - 1:1build-php: Time To Compileselenium: PSPDFKit WASM - Firefoxcockroach: KV, 50% Reads - 256cockroach: KV, 10% Reads - 128pgbench: 1 - 50 - Read Write - Average Latencycockroach: KV, 50% Reads - 512cockroach: KV, 60% Reads - 256ncnn: CPU - efficientnet-b0pgbench: 1 - 50 - Read Writecockroach: KV, 10% Reads - 256dacapobench: Tradesoaprocksdb: Rand Fill Syncjpegxl: JPEG - 100renaissance: Apache Spark PageRankselenium: Jetstream 2 - Firefoxcockroach: MoVR - 128selenium: Kraken - Firefoxmysqlslap: 16cockroach: KV, 95% Reads - 512openvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUspark: 1000000 - 100 - SHA-512 Benchmark Timerocksdb: Read While Writingdacapobench: Jythoncockroach: KV, 10% Reads - 512spark: 1000000 - 100 - Repartition Test Timeclickhouse: 100M Rows Web Analytics Dataset, Third Runnumenta-nab: KNN CADgraphics-magick: Noise-Gaussiancockroach: KV, 60% Reads - 512mysqlslap: 8build-python: Defaultluaradio: Complex Phasedeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Detection,YOLOv5s COCO - Synchronous Single-Streampgbench: 100 - 1 - Read Only - Average Latencysvt-av1: Preset 13 - Bosphorus 1080psmhasher: FarmHash128build-nodejs: Time To Compilememtier-benchmark: Redis - 100 - 10:1spark: 1000000 - 500 - Inner Join Test Timerenaissance: Rand Forestjpegxl: PNG - 100cockroach: MoVR - 512stress-ng: CPU Cacheopenradioss: Bumper Beamncnn: CPU - googlenetavifenc: 6, Losslessunpack-linux: linux-5.19.tar.xzsmhasher: t1ha0_aes_avx2 x86_64compress-7zip: Compression Ratingmnn: mobilenet-v1-1.0svt-av1: Preset 12 - Bosphorus 1080ppgbench: 100 - 1 - Read Onlygraphics-magick: Resizingpgbench: 1 - 1 - Read Write - Average Latencypgbench: 1 - 1 - Read Writeluaradio: Five Back to Back FIR Filtersavifenc: 6spark: 1000000 - 500 - Broadcast Inner Join Test Timespark: 1000000 - 500 - Calculate Pi Benchmark Using Dataframeopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUsvt-av1: Preset 12 - Bosphorus 4Kncnn: CPU - squeezenet_ssdstress-ng: CPU Stressncnn: CPU - mobilenetopenvino: Vehicle Detection FP16 - CPUopenvino: Vehicle Detection FP16 - CPUrenaissance: Apache Spark Bayesopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUsmhasher: FarmHash32 x86_64 AVXopenvino: Age Gender Recognition Retail 0013 FP16 - CPUspark: 1000000 - 100 - Calculate Pi Benchmark Using Dataframedeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamopenradioss: Rubber O-Ring Seal Installationncnn: CPU - vision_transformerdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamsmhasher: wyhashopenvino: Age Gender Recognition Retail 0013 FP16 - CPUsvt-av1: Preset 13 - Bosphorus 4Ksmhasher: t1ha2_atoncestargate: 44100 - 1024stress-ng: Matrix Mathstress-ng: Memory Copyingcpuminer-opt: Triple SHA-256, Onecoinopenradioss: INIVOL and Fluid Structure Interaction Drop Containernumenta-nab: Earthgecko Skylinepgbench: 100 - 1 - Read Write - Average Latencyrav1e: 10ncnn: CPU - resnet18ncnn: CPU - resnet50spark: 1000000 - 100 - Inner Join Test Timeluaradio: FM Deemphasis Filterbuild-python: Released Build, PGO + LTO Optimizedpgbench: 100 - 1 - Read Writecpuminer-opt: Magiwebp2: Defaultredis: SET - 50graphics-magick: Swirlsvt-av1: Preset 8 - Bosphorus 4Knumenta-nab: Contextual Anomaly Detector OSEcpuminer-opt: Blake-2 Sopenvino: Person Detection FP16 - CPUrocksdb: Rand Readnumenta-nab: Windowed Gaussianncnn: CPU - alexnetencode-flac: WAV To FLACsmhasher: SHA3-256selenium: WASM collisionDetection - Firefoxgraphics-magick: Enhancedopenfoam: drivaerFastback, Small Mesh Size - Mesh Timecpuminer-opt: Ringcoindeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMstress-ng: Atomiccpuminer-opt: Garlicoinstress-ng: Glibc C String Functionsdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamcpuminer-opt: Quad SHA-256, Pyriterav1e: 5deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamavifenc: 2deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamwebp: Quality 100ncnn: CPU - yolov4-tinydeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamcpuminer-opt: x25xsvt-av1: Preset 8 - Bosphorus 1080ponednn: IP Shapes 3D - f32 - CPUsmhasher: Spooky32rav1e: 6selenium: WASM imageConvolute - Firefoxonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamopenvino: Face Detection FP16 - CPUavifenc: 0openvino: Face Detection FP16 - CPUstargate: 44100 - 512srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMstargate: 480000 - 1024webp: Quality 100, Losslessdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamopenvino: Machine Translation EN To DE FP16 - CPUsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMopenvino: Person Detection FP16 - CPUopenvino: Machine Translation EN To DE FP16 - CPUcpuminer-opt: Myriad-Groestlopenfoam: drivaerFastback, Small Mesh Size - Execution Timeluaradio: Hilbert Transformopenradioss: Bird Strike on Windshieldopenvino: Vehicle Detection FP16-INT8 - CPUopenradioss: Cell Phone Drop Testopenvino: Vehicle Detection FP16-INT8 - CPUdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamsrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamwebp: Quality 100, Highest Compressioncompress-7zip: Decompression Ratingsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMsvt-av1: Preset 4 - Bosphorus 1080pwebp: Defaultsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMbrl-cad: VGR Performance Metricsvt-av1: Preset 4 - Bosphorus 4Kblender: BMW27 - CPU-Onlystress-ng: Glibc Qsort Data Sortingopenvino: Person Detection FP32 - CPUcpuminer-opt: Deepcoindeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamsmhasher: MeowHash x86_64 AES-NIminibude: OpenMP - BM1minibude: OpenMP - BM1onednn: Deconvolution Batch shapes_3d - f32 - CPUcpuminer-opt: LBC, LBRY Creditssrsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMstargate: 96000 - 1024srsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamstargate: 96000 - 512deepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamspark: 1000000 - 500 - Calculate Pi Benchmarkspark: 1000000 - 100 - Calculate Pi Benchmarksrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamsrsran: OFDM_Testsrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMxmrig: Wownero - 1Mcpuminer-opt: scryptnekrs: TurboPipe Periodicstress-ng: Vector Mathstargate: 480000 - 512onednn: Recurrent Neural Network Inference - f32 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUncnn: CPU - vgg16xmrig: Monero - 1Monednn: IP Shapes 1D - f32 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16 - CPUstress-ng: Cryptoopenvino: Weld Porosity Detection FP16 - CPUcpuminer-opt: Skeincoinopenvino: Face Detection FP16-INT8 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUsmhasher: fasthash32openvino: Face Detection FP16-INT8 - CPUopenvino: Person Detection FP32 - CPUopenvkl: vklBenchmark Scalaropenvkl: vklBenchmark ISPCgraphics-magick: Sharpenwebp2: Quality 95, Compression Effort 7webp2: Quality 75, Compression Effort 7webp: Quality 100, Lossless, Highest Compressionnatron: Spaceshipmemtier-benchmark: Redis - 50 - 10:1memtier-benchmark: Redis - 50 - 1:1cockroach: MoVR - 256spark: 1000000 - 100 - Broadcast Inner Join Test Timehackbench: 2 - Processhackbench: 1 - Processhackbench: 4 - Threadhackbench: 2 - Threadhackbench: 1 - Threadonednn: Deconvolution Batch shapes_1d - f32 - CPUdacapobench: Tradebeansdacapobench: H2smhasher: MeowHash x86_64 AES-NIsmhasher: t1ha0_aes_avx2 x86_64smhasher: FarmHash32 x86_64 AVXsmhasher: t1ha2_atoncesmhasher: FarmHash128smhasher: fasthash32smhasher: Spooky32smhasher: SHA3-256smhasher: wyhashsockperf: ThroughputDefault - IBRSretbleed=stuffmitigations=offretbleed=off4462.862107608.22388441.98745688.97107.092108.74652.230217.355221.69628232586.503275111189.18830087.70924510.2829.6730164.718823970.8154816.64112.42827558.9325.280317833888.7593.320211520939991.362797.247.05031750905178.02364.15666.94920.3965.36737610713111.864471.29107.71838813.5519808923974.2527451.423838.5313903.12.2401023940.48823460.51902480.4122618.361.00221430.719983710635.921713.790.50199895103.435.1710209922195.30.9795.055.4531.475.3264056116.571844039.714.49241.05463.584432.31231773.0021.59046334310.64.21893182.925.38391612.768220.868937232.81287.9026.491372487.30200374.6223367879.175.105.276.6620.05865694.3176.823.9824846.874699438.78402.5846.145.406.5474.8712.43121848.012263.714206.921820.81142.41360130.93142.630343815334.76837.8188.24215417.916214.110.1826612182.171146740.534244.485.239106.91250.025820258.0157.8525.325.90721604445213556.13.6979.54423.2199915948.628337.535456.757.375317.42390.050207.79113059.241783.4921291653.164.05849.90.55108.1158.46412.0116.8628.5289.65439760.79280964.570185.795199393985.085197577.319.9383.3824.332968.9248.27022.957987.2922.1286.0246.503210.11.3220898.252662.4824.5411.5044565.84422.28173.821820542.661.4851.12312505.381.28526518152.96850.74266971649.05262.9765.3226.07314.3731.843.58356.7442.21618897.922.682315119.8313812.79692.5211550600.731755909120.20611.7121.655122.26600.18189.670778830.673.0824324.4131130.9180741.361140.68243451.44322.63353.0994192671.80227.5446171.4723.19399.0332.5172.5893626.1738156.4436.01612.370012205.362.42633.25.5133296.512910.36003306.20379.2181.21.215741110.01.2419261.3041.351924.176713.20106.25354.18302.545371.74916.4420568.2735.24136.56308.9429.2717.4858323.0114.35223.0722169328.63.76414.16115.9500861.064369.6155.555351.083840.30143.089213.975332896.515.017125.41914.44411055053.00.928978137.234.808328.71520.85975981.759312.2294625.92833.1951436.79441.23354.454.559536.6421102366667359.62030.772.551926380000014401.851.1810603967.4118.2878.301429.87.68644218.59113.014815.3035.4163601797.1320.92007409.456164.112.220.733667390.020.050.421.21109992.571175178.96108.53.1425.05212.80659.16727.97613.04514.14304195376364.51333.71344.58034.00751.41036.38850.0253226.91923.8813382904492.763337126.74603448.111111966.0180.42681.92739.089166.813167.70411943422.604076128725.84971383.121035968.5326.4983324.270885942.9252701.66106.29883794.5623.783843891561.7885.8203576151053867.932086.643.42317652016666.18654.66256.78122.0555.98438975313778.465181.1193.20322711.7521810625852.1723571.825553.0512739.72.0581119820.44624909.111068334.7524113.330.93022968.361075659115.422945.290.461108308114.195.7411067119597.70.9035.66.0334.885.8869391015.896927313.864.35839.58460.344200.61286001.5522.37844704122.23.91942396.755.03375313.323207.353998481.69271.3296.111440930.09201794.4893089573.834.854.996.4830.050898246.1581.373.7426145.846721201.26383.8415.885.186.2978.1811.85222736.113101.015153.922382.31141.51447187.46138.065331916052.67257.6179.10216043.617025.69.8327912888.470286640.554158.888.651111.71209.827020766.5153.3726.065.93759004444114251.33.5481.55410.61110216550.429536.71458.759.622816.76780.05212.11813103.281748.8681319663.844.03834.50.56112.0152.96404.1316.5428.0879.64138438.77285074.501189.070201404034.999200576.219.7643.3424.183031.9048.75622.558039.9121.7585.9546.533177.81.320869.882703.2224.1611.5245559.22421.62173.517120125.381.4651.92812566.661.28493917799.14834.84272131633.50264.6295.3606.14414.2531.503.56357.0438.59918799.402.712289592.5013912.92492.4271559000.731733183620.23411.6621.437122.13596.88289.594193822.673.0464328.2445130.9182114.031153.43243636.88323.13183.0946190971.79927.3999171.5683.16219.0932.3172.9674632.3771157.7936.26012.252812211.482.44833.15.4833895.844010.43233287.93380.5811.211.217403110.41.2511871.3141.668723.993013.14105.45337.23303.875400.62913.5020568.7732.52137.20307.3429.1417.4051321.9114.87853.0922160329.23.77614.21115.8500641.068368.0955.865339.563840.87143.742413.911932807.205.017125.43214.47521059053.10.926504137.134.924228.62000.85676082.020612.1905624.35193.2032437.77439.91353.754.419836.7359102400000359.02031.172.711922676666714404.731.1811433968.4718.2778.381428.67.69614218.69112.974817.2535.38163471797.9320.91717409.076164.442.220.733667390.020.050.420.71180753.401232491.42111.92.9919.1649.87641.52121.55710.57714.05983878353864.48233.70042.43634.30051.40636.39250.2683226.72724.103402526194722.996558503.531165429.652131852.0343.54644.48421.65395.568101.2102304834.075859195756.151440190.041563150.0118.1538802.9201198373.9076342.18114.071159234.3918.1667011157720.2067.2491397171290274.482040.734.50711668793258.60448.26350.37226.8849.64549290017166.225850.9983.18932810.6125225030466.9221755.329875.4611106.71.7971265120.39528879.671108896.9127769.280.82326090.311215448754.126356.080.414120818124.816.2012243318521.40.8176.056.5037.366.3175081714.194979021.513.87435.44469.783839.91403515.5422.59244293809.83.741004168.044.80364513.717198.2221037459.90260.6225.891485496.09218884.2333171723.174.684.846.1240.046939884.9583.143.7126343.406755001.97380.4465.705.036.1180.1311.62123352.312637.014750.123274.71150.81422407.47134.186323716279.67068.9179.91816312.117086.49.6027812714.467286630.564071.490.050111.41185.126921346.1150.4626.575.63735461436514119.53.5183.60402.72510416635.828835.937476.359.663516.75600.048216.17413107.911716.4891341587.323.90819.70.57111.6157.33398.0116.2927.5699.33139559.81290554.421191.993205984114.961202593.719.3583.3624.893056.0349.63422.358179.8521.5685.4746.783215.21.2920829.822689.5324.0011.2716553.46413.10177.262420425.041.4752.05512500.301.26131118017.30834.50267231617.86262.7075.3566.18214.1231.293.53362.6435.07818798.242.722321265.7514012.98093.6941550630.741747195719.97311.5621.387120.76597.08289.417796827.693.0587326.9267130.8180081.271140.86240956.56325.92263.0682192991.80227.4149170.0923.16669.1032.1972.9335631.5886156.2636.36212.338212101.052.44833.25.4957996.378810.37433278.21379.2831.211.207418110.41.2517961.3141.626824.017513.18105.75313.90303.095411.58909.7385968.4729.94136.24306.7729.3417.3925322.7114.95803.0922027328.53.78814.20115.9501041.070368.755.825323.793858.75143.464213.938832954.044.996124.91014.50331057753.00.927538136.734.933628.61230.85774481.999512.1937626.48613.1923436.93441.21353.654.412936.7410102366667359.72033.872.641926260000014411.601.1832433963.5918.2678.261430.47.69350218.83113.004816.6735.38163531798.3020.92977405.606165.802.220.733667390.020.050.420.71288891.751481936.11111.23.1911.3586.49723.65415.1017.95514.90973927363064.48833.69442.43034.03451.41136.79950.2653270.41823.8936909764482.553767497.65682715.711285481.5368.06469.70133.163143.008142.98111813720.584807154975.581156275.481249349.8120.9218013.6111057571.9958646.53151.511044494.5319.7800001035956.6274.8682026541147643.612069.538.24472461315561.99951.01153.18024.5152.55745687915910.085511.0889.00467611.7022819427933.1823525.627412.8312396.22.1021155880.43326593.581085041.5925604.080.89624122.051116029170.924256.310.450111247120.775.9411430319637.60.8755.796.2435.846.0671215116.418966283.044.41740.20668.264037.21317556.8219.84850393966.83.89987881.725.05408312.249204.1071033004.97266.8356.291499557.64203904.5153096924.84.844.986.5890.049928724.1583.373.6826846.330728080.34372.9635.845.096.2679.4611.74423189.112798.215102.322888.91215.91410074.30136.236328916169.57123.9177.36116356.317196.39.7928212830.469627010.554017.688.460105.81199.527221086.8158.5225.225.69737668423414248.23.5182.35413.51710316712.829636.564474.659.807616.71580.05212.54412611.701734.9711329633.774.02825.60.56111.6153.84403.7716.5027.8549.47239619.03289534.528190.252201154104.927203593.019.4873.4424.243051.9149.04622.787966.4221.8983.9747.633140.91.2920433.252722.5824.2011.3638557.08422.01175.971820525.161.4552.17112753.071.28678918087.72843.09266971626.25259.6355.2616.18214.3031.503.52362.3436.20319098.312.712323685.5014012.96892.9451537930.741754238520.15511.6221.462122.25604.28290.518025820.723.0623326.5431129.4180772.461144.59243127.33326.19393.0656191001.81827.2657169.8393.16609.1232.4273.3109631.7023157.1836.34212.325312124.072.43533.45.4655095.681010.45003285.23377.4011.211.217322109.51.2517741.3141.661623.997313.24105.45325.14301.635374.93910.8742268.4731.86137.21307.3029.1417.3667320.8115.12913.0822114327.13.78114.25115.2503581.069367.5555.735332.523845.26143.408813.944232887.585.018125.45014.50171058052.90.929955137.234.930828.61480.85941282.042312.1873626.06343.1945436.316232958439.84354.754.391536.7551102133333359.92029.072.611925000000014384.511.1818813961.2618.2978.351428.37.68653218.56112.884812.4735.41163531798.5120.91457405.046164.712.220.733667390.020.050.420.81220692.331329151.41105.92.9816.1288.30333.50218.2918.97013.53533923368664.48734.51843.30334.00651.40936.39050.0283232.77324.148533313OpenBenchmarking.org

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: x86_64 RdRandDefault - IBRSretbleed=stuffmitigations=offretbleed=off40K80K120K160K200KSE +/- 25.86, N = 3SE +/- 39.20, N = 3SE +/- 2.25, N = 3SE +/- 24.86, N = 34462.864492.76194722.994482.551. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: System V Message PassingDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.4M2.8M4.2M5.6M7MSE +/- 521.34, N = 3SE +/- 4395.83, N = 3SE +/- 8723.09, N = 3SE +/- 3066.42, N = 32107608.223337126.746558503.533767497.651. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MallocDefault - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 2251.44, N = 3SE +/- 3175.67, N = 3SE +/- 5092.34, N = 3SE +/- 8981.65, N = 3388441.98603448.111165429.65682715.711. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MutexDefault - IBRSretbleed=stuffmitigations=offretbleed=off500K1000K1500K2000K2500KSE +/- 1727.54, N = 3SE +/- 2417.49, N = 3SE +/- 3583.54, N = 3SE +/- 511.77, N = 3745688.971111966.012131852.031285481.531. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ProcessDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.33, N = 3SE +/- 0.11, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3107.0980.4343.5568.061. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ThreadDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.60, N = 3SE +/- 0.89, N = 3SE +/- 0.06, N = 3SE +/- 0.83, N = 3108.7581.9344.4869.701. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ProcessDefault - IBRSretbleed=stuffmitigations=offretbleed=off1224364860SE +/- 0.49, N = 3SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 352.2339.0921.6533.161. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ProcessDefault - IBRSretbleed=stuffmitigations=offretbleed=off50100150200250SE +/- 0.48, N = 3SE +/- 0.25, N = 3SE +/- 0.33, N = 3SE +/- 0.47, N = 3217.36166.8195.57143.011. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ThreadDefault - IBRSretbleed=stuffmitigations=offretbleed=off50100150200250SE +/- 2.54, N = 6SE +/- 0.52, N = 3SE +/- 1.10, N = 15SE +/- 0.19, N = 3221.70167.70101.21142.981. (CC) gcc options: -lpthread

ctx_clock

Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off6001200180024003000SE +/- 3.33, N = 3SE +/- 0.67, N = 3282311942301181

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Socket ActivityDefault - IBRSretbleed=stuffmitigations=offretbleed=off10002000300040005000SE +/- 4.43, N = 3SE +/- 0.53, N = 3SE +/- 2.42, N = 3SE +/- 3.05, N = 32586.503422.604834.073720.581. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

PostMark

This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceDefault - IBRSretbleed=stuffmitigations=offretbleed=off13002600390052006500SE +/- 14.33, N = 3SE +/- 22.00, N = 3SE +/- 46.33, N = 332754076585948071. (CC) gcc options: -O3

KeyDB

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

OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.2.0Default - IBRSretbleed=stuffmitigations=offretbleed=off40K80K120K160K200KSE +/- 699.76, N = 3SE +/- 752.57, N = 3SE +/- 833.55, N = 3SE +/- 920.86, N = 3111189.18128725.84195756.15154975.581. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: FutexDefault - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 7928.09, N = 3SE +/- 1267.42, N = 3SE +/- 2035.60, N = 3SE +/- 3071.26, N = 3830087.70971383.121440190.041156275.481. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Context SwitchingDefault - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 2033.85, N = 3SE +/- 679.86, N = 3SE +/- 1094.09, N = 3SE +/- 1335.54, N = 3924510.281035968.531563150.011249349.811. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesDefault - IBRSretbleed=stuffmitigations=offretbleed=off714212835SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 3SE +/- 0.04, N = 329.6726.5018.1520.921. (CC) gcc options: -lm

Sockperf

This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.7Test: Latency Ping PongDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.06162.12323.18484.24645.308SE +/- 0.045, N = 5SE +/- 0.051, N = 5SE +/- 0.004, N = 5SE +/- 0.009, N = 54.7184.2702.9203.6111. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread

Memcached

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.16Set To Get Ratio: 1:10Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 1191.60, N = 3SE +/- 1221.65, N = 3SE +/- 2970.13, N = 3SE +/- 1613.00, N = 3823970.81885942.921198373.901057571.991. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SENDFILEDefault - IBRSretbleed=stuffmitigations=offretbleed=off16K32K48K64K80KSE +/- 28.85, N = 3SE +/- 7.12, N = 3SE +/- 11.51, N = 3SE +/- 20.22, N = 354816.6452701.6676342.1858646.531. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MEMFDDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 1.50, N = 3SE +/- 0.99, N = 3SE +/- 1.18, N = 3SE +/- 2.38, N = 3112.42106.29114.07151.511. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Memcached

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.16Set To Get Ratio: 1:5Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 1256.03, N = 3SE +/- 1962.85, N = 3SE +/- 3325.22, N = 3SE +/- 3239.10, N = 3827558.93883794.561159234.391044494.531. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.23, N = 3SE +/- 0.28, N = 3SE +/- 0.11, N = 3SE +/- 0.26, N = 325.2823.7818.1719.781. (CC) gcc options: -lm

Memcached

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.16Set To Get Ratio: 1:1Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 543.55, N = 3SE +/- 3880.05, N = 3SE +/- 4312.41, N = 3SE +/- 7025.44, N = 3833888.75891561.781157720.201035956.621. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.17, N = 3SE +/- 0.21, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 393.3285.8267.2574.871. (CC) gcc options: -lm

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: RotateDefault - IBRSretbleed=stuffmitigations=offretbleed=off150300450600750SE +/- 0.58, N = 3SE +/- 0.88, N = 3SE +/- 1.86, N = 35206157176541. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

Memcached

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.16Set To Get Ratio: 5:1Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 1428.58, N = 3SE +/- 1529.09, N = 3SE +/- 882.77, N = 3SE +/- 1635.50, N = 3939991.361053867.931290274.481147643.611. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Genetic Algorithm Using Jenetics + FuturesDefault - IBRSretbleed=stuffmitigations=offretbleed=off6001200180024003000SE +/- 47.68, N = 3SE +/- 6.73, N = 3SE +/- 21.98, N = 3SE +/- 4.33, N = 32797.22086.62040.72069.5MIN: 2667.81 / MAX: 2893.34MIN: 2049.58 / MAX: 2124.33MIN: 1981.59 / MAX: 2119.04MIN: 2035.71 / MAX: 2118.84

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesDefault - IBRSretbleed=stuffmitigations=offretbleed=off1122334455SE +/- 0.15, N = 3SE +/- 0.11, N = 3SE +/- 0.38, N = 3SE +/- 0.51, N = 347.0543.4234.5138.241. (CC) gcc options: -lm

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Sequential FillDefault - IBRSretbleed=stuffmitigations=offretbleed=off150K300K450K600K750KSE +/- 5351.57, N = 3SE +/- 6126.72, N = 6SE +/- 8478.51, N = 3SE +/- 1867.35, N = 35090515201666879326131551. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

EnCodec

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

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 24 kbpsDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.13, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 378.0266.1958.6062.00

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 1.5 kbpsDefault - IBRSretbleed=stuffmitigations=offretbleed=off1428425670SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 364.1654.6648.2651.01

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 6 kbpsDefault - IBRSretbleed=stuffmitigations=offretbleed=off1530456075SE +/- 0.04, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 366.9556.7850.3753.18

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MMAPDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.31, N = 3SE +/- 0.31, N = 4SE +/- 0.27, N = 8SE +/- 0.29, N = 620.3922.0526.8824.511. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

EnCodec

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

OpenBenchmarking.orgSeconds, Fewer Is BetterEnCodec 0.1.1Target Bandwidth: 3 kbpsDefault - IBRSretbleed=stuffmitigations=offretbleed=off1530456075SE +/- 0.25, N = 3SE +/- 0.61, N = 3SE +/- 0.56, N = 3SE +/- 0.43, N = 365.3755.9849.6552.56

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Random FillDefault - IBRSretbleed=stuffmitigations=offretbleed=off110K220K330K440K550KSE +/- 4120.62, N = 3SE +/- 4573.03, N = 3SE +/- 5841.02, N = 3SE +/- 5715.80, N = 33761073897534929004568791. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: ForkingDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 37.41, N = 3SE +/- 44.23, N = 3SE +/- 42.84, N = 3SE +/- 44.04, N = 313111.8613778.4617166.2215910.081. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: HWB Color SpaceDefault - IBRSretbleed=stuffmitigations=offretbleed=off130260390520650SE +/- 0.33, N = 3SE +/- 0.88, N = 3SE +/- 0.67, N = 3SE +/- 0.58, N = 34475185855511. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

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 20220729Target: CPU - Model: blazefaceDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.29030.58060.87091.16121.4515SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.291.110.991.08MIN: 1.27 / MAX: 1.5MIN: 1.08 / MAX: 1.38MIN: 0.97 / MAX: 1.26MIN: 1.06 / MAX: 1.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.20, N = 3SE +/- 0.01, N = 3SE +/- 0.16, N = 3SE +/- 0.02, N = 3107.7293.2083.1989.001. (CC) gcc options: -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 20220729Target: CPU - Model: regnety_400mDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 313.5511.7510.6111.70MIN: 13.43 / MAX: 16.58MIN: 11.68 / MAX: 14.61MIN: 10.55 / MAX: 10.88MIN: 11.6 / MAX: 23.721. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Update RandomDefault - IBRSretbleed=stuffmitigations=offretbleed=off50K100K150K200K250KSE +/- 1397.10, N = 3SE +/- 160.82, N = 3SE +/- 2189.34, N = 3SE +/- 440.67, N = 31980892181062522502281941. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 20Default - IBRSretbleed=stuffmitigations=offretbleed=off7K14K21K28K35KSE +/- 6.23, N = 3SE +/- 14.90, N = 3SE +/- 28.27, N = 3SE +/- 19.21, N = 323974.2525852.1730466.9227933.181. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark ALSDefault - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 171.52, N = 3SE +/- 337.23, N = 3SE +/- 84.68, N = 3SE +/- 116.97, N = 327451.423571.821755.323525.6MIN: 26970.66 / MAX: 27928.53MIN: 22927.95 / MAX: 24405.07MIN: 21475.58 / MAX: 22054.88MIN: 23177.22 / MAX: 23997.61

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100Default - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 9.32, N = 3SE +/- 16.02, N = 3SE +/- 10.38, N = 3SE +/- 29.95, N = 323838.5325553.0529875.4627412.831. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Akka Unbalanced Cobwebbed TreeDefault - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 54.55, N = 3SE +/- 30.85, N = 3SE +/- 41.07, N = 3SE +/- 92.58, N = 313903.112739.711106.712396.2MIN: 10489.9 / MAX: 14005.14MIN: 9693.86 / MAX: 12792.39MIN: 8323.26 / MAX: 11185.64MIN: 9295.02 / MAX: 12494.23

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: mobilenetV3Default - IBRSretbleed=stuffmitigations=offretbleed=off0.5041.0081.5122.0162.52SE +/- 0.006, N = 3SE +/- 0.007, N = 3SE +/- 0.002, N = 3SE +/- 0.002, N = 32.2402.0581.7972.102MIN: 2.19 / MAX: 18.29MIN: 2.02 / MAX: 3.45MIN: 1.77 / MAX: 5.04MIN: 2.07 / MAX: 6.281. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off30K60K90K120K150KSE +/- 363.26, N = 3SE +/- 934.91, N = 3SE +/- 240.72, N = 3SE +/- 462.80, N = 31023941119821265121155881. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.10980.21960.32940.43920.549SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 30.4880.4460.3950.4331. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200Default - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 23.99, N = 3SE +/- 26.02, N = 3SE +/- 7.45, N = 3SE +/- 37.50, N = 323460.5124909.1128879.6726593.581. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as 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 BetterDragonflydb 0.6Clients: 50 - Set To Get Ratio: 1:5Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 3202.61, N = 3SE +/- 9874.35, N = 3SE +/- 14928.63, N = 3SE +/- 2098.57, N = 3902480.411068334.751108896.911085041.591. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500Default - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 13.33, N = 3SE +/- 23.71, N = 3SE +/- 46.53, N = 3SE +/- 22.49, N = 322618.3624113.3327769.2825604.081. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.22550.4510.67650.9021.1275SE +/- 0.004, N = 3SE +/- 0.011, N = 3SE +/- 0.003, N = 3SE +/- 0.003, N = 31.0020.9300.8230.8961. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 4000Default - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 53.13, N = 3SE +/- 174.05, N = 3SE +/- 81.17, N = 3SE +/- 35.92, N = 321430.7122968.3626090.3124122.051. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off30K60K90K120K150KSE +/- 387.80, N = 3SE +/- 1297.33, N = 3SE +/- 464.85, N = 3SE +/- 363.63, N = 3998371075651215441116021. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Savina Reactors.IODefault - IBRSretbleed=stuffmitigations=offretbleed=off2K4K6K8K10KSE +/- 93.82, N = 3SE +/- 146.02, N = 3SE +/- 28.35, N = 3SE +/- 126.82, N = 310635.99115.48754.19170.9MIN: 10109.53 / MAX: 16338.05MIN: 8944.48 / MAX: 13088.78MIN: 8697.4 / MAX: 12208.1MIN: 8933.81 / MAX: 12964.22

nginx

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

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000Default - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 98.51, N = 3SE +/- 89.46, N = 3SE +/- 181.77, N = 3SE +/- 190.93, N = 321713.7922945.2926356.0824256.311. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.11270.22540.33810.45080.5635SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.003, N = 3SE +/- 0.002, N = 30.5010.4610.4140.4501. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off30K60K90K120K150KSE +/- 252.50, N = 3SE +/- 180.13, N = 3SE +/- 778.58, N = 3SE +/- 623.45, N = 3998951083081208181112471. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

JPEG XL Decoding libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: AllDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3103.43114.19124.81120.77

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 80Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 35.175.746.205.941. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off30K60K90K120K150KSE +/- 22.05, N = 3SE +/- 433.11, N = 3SE +/- 203.82, N = 3SE +/- 543.52, N = 31020991106711224331143031. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: ALS Movie LensDefault - IBRSretbleed=stuffmitigations=offretbleed=off5K10K15K20K25KSE +/- 10.16, N = 3SE +/- 78.16, N = 3SE +/- 104.72, N = 3SE +/- 39.43, N = 322195.319597.718521.419637.6MIN: 22175.14 / MAX: 24464.84MIN: 19441.37 / MAX: 21708.22MIN: 18335.77 / MAX: 20604.44MIN: 19560.27 / MAX: 21682.59

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.22030.44060.66090.88121.1015SE +/- 0.000, N = 3SE +/- 0.004, N = 3SE +/- 0.001, N = 3SE +/- 0.004, N = 30.9790.9030.8170.8751. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 90Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 35.055.606.055.791. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 80Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 35.456.036.506.241. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic

JPEG XL Decoding libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL Decoding libjxl 0.7CPU Threads: 1Default - IBRSretbleed=stuffmitigations=offretbleed=off918273645SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 331.4734.8837.3635.84

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 90Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 35.325.886.316.061. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Read Random Write RandomDefault - IBRSretbleed=stuffmitigations=offretbleed=off160K320K480K640K800KSE +/- 2106.07, N = 3SE +/- 4335.60, N = 3SE +/- 8337.88, N = 3SE +/- 4847.99, N = 36405616939107508177121511. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: nasnetDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.03, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.11, N = 316.5715.9014.1916.42MIN: 16.41 / MAX: 57.47MIN: 15.64 / MAX: 34.5MIN: 14.02 / MAX: 28.42MIN: 16.2 / MAX: 35.71. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as 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 BetterDragonflydb 0.6Clients: 50 - Set To Get Ratio: 5:1Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 14572.82, N = 3SE +/- 10204.79, N = 3SE +/- 1484.63, N = 3SE +/- 4701.88, N = 3844039.71927313.86979021.51966283.041. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: squeezenetv1.1Default - IBRSretbleed=stuffmitigations=offretbleed=off1.01072.02143.03214.04285.0535SE +/- 0.001, N = 3SE +/- 0.016, N = 3SE +/- 0.013, N = 3SE +/- 0.016, N = 34.4924.3583.8744.417MIN: 4.44 / MAX: 9.15MIN: 4.28 / MAX: 29.76MIN: 3.82 / MAX: 4.21MIN: 4.33 / MAX: 23.761. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: resnet-v2-50Default - IBRSretbleed=stuffmitigations=offretbleed=off918273645SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.11, N = 341.0539.5835.4440.21MIN: 40.85 / MAX: 88.28MIN: 38.97 / MAX: 57.82MIN: 35.31 / MAX: 50.16MIN: 39.91 / MAX: 59.341. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: NUMADefault - IBRSretbleed=stuffmitigations=offretbleed=off1632486480SE +/- 0.26, N = 3SE +/- 0.44, N = 3SE +/- 0.60, N = 3SE +/- 0.62, N = 363.5860.3469.7868.261. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: In-Memory Database ShootoutDefault - IBRSretbleed=stuffmitigations=offretbleed=off10002000300040005000SE +/- 32.38, N = 3SE +/- 26.79, N = 3SE +/- 12.21, N = 3SE +/- 50.43, N = 44432.34200.63839.94037.2MIN: 3880.98 / MAX: 5099.49MIN: 3663.15 / MAX: 4744.43MIN: 3364.72 / MAX: 4319.78MIN: 3568.99 / MAX: 4384.41

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 Bettermemtier_benchmark 1.4Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 15330.54, N = 3SE +/- 10395.85, N = 3SE +/- 12805.02, N = 15SE +/- 7047.94, N = 31231773.001286001.551403515.541317556.821. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.21, N = 3SE +/- 0.27, N = 3SE +/- 0.36, N = 3SE +/- 0.18, N = 321.5922.3822.5919.851. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read WriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off11002200330044005500SE +/- 45.81, N = 3SE +/- 53.82, N = 3SE +/- 72.12, N = 3SE +/- 46.69, N = 346334470442950391. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Finagle HTTP RequestsDefault - IBRSretbleed=stuffmitigations=offretbleed=off9001800270036004500SE +/- 7.61, N = 3SE +/- 19.11, N = 3SE +/- 36.43, N = 3SE +/- 11.33, N = 34310.64122.23809.83966.8MIN: 3653.96 / MAX: 6515.95MIN: 3496.19 / MAX: 6318.23MIN: 3207.35 / MAX: 5734.05MIN: 3285.61 / MAX: 5723.05

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 20220729Target: CPU - Model: shufflenet-v2Default - IBRSretbleed=stuffmitigations=offretbleed=off0.94731.89462.84193.78924.7365SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 34.213.913.743.89MIN: 4.16 / MAX: 4.34MIN: 3.87 / MAX: 4.05MIN: 3.69 / MAX: 6.95MIN: 3.85 / MAX: 5.061. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as 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 BetterDragonflydb 0.6Clients: 200 - Set To Get Ratio: 1:1Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 3016.85, N = 3SE +/- 3190.84, N = 3SE +/- 2298.87, N = 3SE +/- 6499.14, N = 3893182.92942396.751004168.04987881.721. (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 20220729Target: CPU - Model: FastestDetDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.21052.4213.63154.8426.0525SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.05, N = 25.385.034.805.05MIN: 5.27 / MAX: 5.74MIN: 4.96 / MAX: 5.18MIN: 4.76 / MAX: 5.78MIN: 4.95 / MAX: 6.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 50 - Mode: Read WriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off9001800270036004500SE +/- 14.43, N = 3SE +/- 15.39, N = 3SE +/- 18.41, N = 3SE +/- 42.75, N = 339163753364540831. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 312.7713.3213.7212.251. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Timed Erlang/OTP Compilation

This test times how long it takes to compile Erlang/OTP. Erlang is a programming language and run-time for massively scalable soft real-time systems with high availability requirements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 25.0Time To CompileDefault - IBRSretbleed=stuffmitigations=offretbleed=off50100150200250SE +/- 0.25, N = 3SE +/- 0.20, N = 3SE +/- 0.18, N = 3SE +/- 0.31, N = 3220.87207.35198.22204.11

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as 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 BetterDragonflydb 0.6Clients: 200 - Set To Get Ratio: 1:5Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 9053.64, N = 3SE +/- 10739.05, N = 3SE +/- 1230.98, N = 3SE +/- 6369.58, N = 3937232.81998481.691037459.901033004.971. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigDefault - IBRSretbleed=stuffmitigations=offretbleed=off60120180240300SE +/- 1.63, N = 3SE +/- 1.01, N = 3SE +/- 0.76, N = 3SE +/- 1.01, N = 3287.90271.33260.62266.84

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - SHA-512 Benchmark TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.07, N = 3SE +/- 0.08, N = 4SE +/- 0.03, N = 3SE +/- 0.02, N = 36.496.115.896.29

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 Bettermemtier_benchmark 1.4Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 19117.51, N = 3SE +/- 7241.76, N = 3SE +/- 11096.23, N = 3SE +/- 18956.48, N = 31372487.301440930.091485496.091499557.641. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1 - Mode: Read OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off5K10K15K20K25KSE +/- 37.34, N = 3SE +/- 18.45, N = 3SE +/- 85.96, N = 3SE +/- 93.24, N = 3200372017921888203901. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: MobileNetV2_224Default - IBRSretbleed=stuffmitigations=offretbleed=off1.042.083.124.165.2SE +/- 0.003, N = 3SE +/- 0.006, N = 3SE +/- 0.015, N = 3SE +/- 0.027, N = 34.6224.4894.2334.515MIN: 4.58 / MAX: 6.62MIN: 4.43 / MAX: 5.73MIN: 4.17 / MAX: 7.41MIN: 4.44 / MAX: 5.581. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 7.0.4Test: GET - Parallel Connections: 50Default - IBRSretbleed=stuffmitigations=offretbleed=off700K1400K2100K2800K3500KSE +/- 1752.83, N = 3SE +/- 623.36, N = 3SE +/- 5141.59, N = 3SE +/- 1509.63, N = 33367879.173089573.833171723.173096924.801. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

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 20220729Target: CPU - Model: mnasnetDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.14752.2953.44254.595.7375SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 35.104.854.684.84MIN: 5.04 / MAX: 5.46MIN: 4.8 / MAX: 5.14MIN: 4.64 / MAX: 7.37MIN: 4.79 / MAX: 7.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU-v3-v3 - Model: mobilenet-v3Default - IBRSretbleed=stuffmitigations=offretbleed=off1.18582.37163.55744.74325.929SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 35.274.994.844.98MIN: 5.2 / MAX: 5.71MIN: 4.95 / MAX: 5.22MIN: 4.78 / MAX: 16.71MIN: 4.93 / MAX: 5.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: SqueezeNetV1.0Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.005, N = 3SE +/- 0.013, N = 3SE +/- 0.029, N = 3SE +/- 0.017, N = 36.6626.4836.1246.589MIN: 6.58 / MAX: 8.88MIN: 6.42 / MAX: 8.8MIN: 6.04 / MAX: 16.59MIN: 6.52 / MAX: 13.191. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.01130.02260.03390.04520.0565SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.0500.0500.0460.0491. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as 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 BetterDragonflydb 0.6Clients: 200 - Set To Get Ratio: 5:1Default - IBRSretbleed=stuffmitigations=offretbleed=off200K400K600K800K1000KSE +/- 4933.91, N = 3SE +/- 13869.80, N = 3SE +/- 9961.71, N = 3SE +/- 6448.91, N = 3865694.31898246.15939884.95928724.151. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

ClickHouse

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

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, Second RunDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.80, N = 3SE +/- 0.47, N = 11SE +/- 0.52, N = 3SE +/- 0.38, N = 476.8281.3783.1483.37MIN: 5.2 / MAX: 20000MIN: 5.83 / MAX: 20000MIN: 5.96 / MAX: 20000MIN: 5.89 / MAX: 200001. ClickHouse server version 22.5.4.19 (official build).

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Repartition Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.89551.7912.68653.5824.4775SE +/- 0.09, N = 3SE +/- 0.02, N = 4SE +/- 0.07, N = 3SE +/- 0.08, N = 33.983.743.713.68

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read WriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off60120180240300SE +/- 2.48, N = 3SE +/- 1.32, N = 3SE +/- 0.56, N = 3SE +/- 1.44, N = 32482612632681. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: inception-v3Default - IBRSretbleed=stuffmitigations=offretbleed=off1122334455SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.18, N = 346.8745.8543.4146.33MIN: 46.67 / MAX: 67.33MIN: 45.65 / MAX: 64.31MIN: 42.98 / MAX: 77.27MIN: 46.01 / MAX: 65.91. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SemaphoresDefault - IBRSretbleed=stuffmitigations=offretbleed=off160K320K480K640K800KSE +/- 6946.74, N = 15SE +/- 7121.86, N = 15SE +/- 5827.42, N = 15SE +/- 7291.15, N = 12699438.78721201.26755001.97728080.341. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 100 - Mode: Read Write - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off90180270360450SE +/- 4.05, N = 3SE +/- 1.95, N = 3SE +/- 0.80, N = 3SE +/- 2.01, N = 3402.58383.84380.45372.961. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Group By Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.04, N = 3SE +/- 0.07, N = 4SE +/- 0.06, N = 3SE +/- 0.11, N = 36.145.885.705.84

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Group By Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.2152.433.6454.866.075SE +/- 0.07, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 35.405.185.035.09

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 20220729Target: CPU-v2-v2 - Model: mobilenet-v2Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 36.546.296.116.26MIN: 6.46 / MAX: 6.94MIN: 6.21 / MAX: 6.49MIN: 6.04 / MAX: 6.27MIN: 6.19 / MAX: 6.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

ClickHouse

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

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, First Run / Cold CacheDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 1.00, N = 3SE +/- 0.67, N = 11SE +/- 1.04, N = 3SE +/- 1.18, N = 474.8778.1880.1379.46MIN: 4.96 / MAX: 15000MIN: 5.58 / MAX: 20000MIN: 5.87 / MAX: 20000MIN: 5.8 / MAX: 200001. ClickHouse server version 22.5.4.19 (official build).

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, LosslessDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 312.4311.8511.6211.741. (CXX) g++ options: -O3 -fPIC -lm

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 128Default - IBRSretbleed=stuffmitigations=offretbleed=off5K10K15K20K25KSE +/- 60.10, N = 3SE +/- 96.17, N = 3SE +/- 72.05, N = 3SE +/- 67.44, N = 321848.022736.123352.323189.1

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 128Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 43.99, N = 3SE +/- 29.50, N = 3SE +/- 71.67, N = 3SE +/- 23.00, N = 312263.713101.012637.012798.2

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 128Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 34.01, N = 3SE +/- 31.65, N = 3SE +/- 43.52, N = 3SE +/- 65.34, N = 314206.915153.914750.115102.3

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 256Default - IBRSretbleed=stuffmitigations=offretbleed=off5K10K15K20K25KSE +/- 38.97, N = 3SE +/- 67.07, N = 3SE +/- 48.49, N = 3SE +/- 30.70, N = 321820.822382.323274.722888.9

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Scala DottyDefault - IBRSretbleed=stuffmitigations=offretbleed=off30060090012001500SE +/- 4.99, N = 3SE +/- 14.29, N = 15SE +/- 16.88, N = 15SE +/- 13.79, N = 31142.41141.51150.81215.9MIN: 867.97 / MAX: 2211.31MIN: 836.06 / MAX: 2372.4MIN: 820.44 / MAX: 2321.46MIN: 824.75 / MAX: 2207.54

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 Bettermemtier_benchmark 1.4Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:1Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 18141.35, N = 3SE +/- 9807.81, N = 3SE +/- 20387.26, N = 4SE +/- 18624.65, N = 31360130.931447187.461422407.471410074.301. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Timed PHP Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 8.1.9Time To CompileDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.35, N = 3SE +/- 0.44, N = 3SE +/- 0.28, N = 3SE +/- 0.44, N = 3142.63138.07134.19136.24

Selenium

OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: FirefoxDefault - IBRSretbleed=stuffmitigations=offretbleed=off7001400210028003500SE +/- 10.17, N = 3SE +/- 3.46, N = 334383319323732891. firefox 108.0

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 256Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 35.11, N = 3SE +/- 38.67, N = 3SE +/- 67.56, N = 3SE +/- 35.46, N = 315334.716052.616279.616169.5

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 128Default - IBRSretbleed=stuffmitigations=offretbleed=off16003200480064008000SE +/- 55.63, N = 3SE +/- 26.36, N = 3SE +/- 31.51, N = 3SE +/- 47.48, N = 36837.87257.67068.97123.9

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.16, N = 3SE +/- 0.34, N = 3SE +/- 0.42, N = 3SE +/- 0.31, N = 3188.24179.10179.92177.361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 65.05, N = 3SE +/- 52.98, N = 3SE +/- 73.08, N = 3SE +/- 24.66, N = 315417.916043.616312.116356.3

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 256Default - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 57.53, N = 3SE +/- 125.50, N = 3SE +/- 40.38, N = 3SE +/- 80.54, N = 316214.117025.617086.417196.3

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 20220729Target: CPU - Model: efficientnet-b0Default - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 310.189.839.609.79MIN: 10.11 / MAX: 10.58MIN: 9.75 / MAX: 21.91MIN: 9.52 / MAX: 21.53MIN: 9.74 / MAX: 9.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 50 - Mode: Read WriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off60120180240300SE +/- 0.23, N = 3SE +/- 0.53, N = 3SE +/- 0.65, N = 3SE +/- 0.48, N = 32662792782821. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 256Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 20.31, N = 3SE +/- 67.67, N = 3SE +/- 20.72, N = 3SE +/- 37.60, N = 312182.112888.412714.412830.4

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapDefault - IBRSretbleed=stuffmitigations=offretbleed=off15003000450060007500SE +/- 50.45, N = 3SE +/- 38.70, N = 4SE +/- 49.58, N = 4SE +/- 90.71, N = 47114702867286962

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Random Fill SyncDefault - IBRSretbleed=stuffmitigations=offretbleed=off150300450600750SE +/- 4.73, N = 3SE +/- 8.33, N = 3SE +/- 9.10, N = 4SE +/- 4.06, N = 36746646637011. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: JPEG - Quality: 100Default - IBRSretbleed=stuffmitigations=offretbleed=off0.1260.2520.3780.5040.63SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.530.550.560.551. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark PageRankDefault - IBRSretbleed=stuffmitigations=offretbleed=off9001800270036004500SE +/- 23.20, N = 3SE +/- 32.03, N = 3SE +/- 2.22, N = 3SE +/- 48.87, N = 34244.44158.84071.44017.6MIN: 3945.79 / MAX: 4429.16MIN: 3865.05 / MAX: 4442.02MIN: 3791.67 / MAX: 4350.58MIN: 3606.13 / MAX: 4304.31

Selenium

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream 2 - Browser: FirefoxDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.74, N = 3SE +/- 0.35, N = 3SE +/- 0.48, N = 3SE +/- 1.06, N = 285.2488.6590.0588.461. firefox 108.0

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 128Default - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.48, N = 3SE +/- 0.19, N = 3SE +/- 0.75, N = 3SE +/- 1.02, N = 3106.9111.7111.4105.8

Selenium

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: Kraken - Browser: FirefoxDefault - IBRSretbleed=stuffmitigations=offretbleed=off30060090012001500SE +/- 3.24, N = 3SE +/- 5.53, N = 3SE +/- 4.04, N = 3SE +/- 2.56, N = 31250.01209.81185.11199.51. firefox 108.0

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.8.2Clients: 16Default - IBRSretbleed=stuffmitigations=offretbleed=off60120180240300SE +/- 0.33, N = 3SE +/- 0.76, N = 3SE +/- 0.38, N = 3SE +/- 0.80, N = 32582702692721. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -pthread -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off5K10K15K20K25KSE +/- 39.90, N = 3SE +/- 95.61, N = 3SE +/- 115.39, N = 3SE +/- 19.89, N = 320258.020766.521346.121086.8

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.26, N = 3SE +/- 1.41, N = 3SE +/- 1.24, N = 3SE +/- 0.93, N = 3157.85153.37150.46158.521. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Vehicle Bike Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.04, N = 3SE +/- 0.24, N = 3SE +/- 0.22, N = 3SE +/- 0.15, N = 325.3226.0626.5725.22MIN: 12.97 / MAX: 44.7MIN: 12.62 / MAX: 46.95MIN: 12.63 / MAX: 42.23MIN: 13.24 / MAX: 45.161. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.33432.66864.00295.33726.6715SE +/- 0.06, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 35.905.935.635.69

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Read While WritingDefault - IBRSretbleed=stuffmitigations=offretbleed=off160K320K480K640K800KSE +/- 10015.04, N = 3SE +/- 10552.54, N = 3SE +/- 4523.86, N = 3SE +/- 11857.59, N = 37216047590047354617376681. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonDefault - IBRSretbleed=stuffmitigations=offretbleed=off10002000300040005000SE +/- 50.46, N = 7SE +/- 22.52, N = 4SE +/- 51.90, N = 5SE +/- 23.53, N = 44452444143654234

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 10% Reads - Concurrency: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 17.17, N = 3SE +/- 24.56, N = 3SE +/- 31.63, N = 3SE +/- 61.83, N = 313556.114251.314119.514248.2

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Repartition Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.83031.66062.49093.32124.1515SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.06, N = 33.693.543.513.51

ClickHouse

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

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, Third RunDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.72, N = 3SE +/- 0.41, N = 11SE +/- 0.21, N = 3SE +/- 1.06, N = 479.5481.5583.6082.35MIN: 5.59 / MAX: 20000MIN: 5.79 / MAX: 20000MIN: 6.05 / MAX: 20000MIN: 5.89 / MAX: 200001. ClickHouse server version 22.5.4.19 (official build).

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: KNN CADDefault - IBRSretbleed=stuffmitigations=offretbleed=off90180270360450SE +/- 5.51, N = 3SE +/- 0.43, N = 3SE +/- 1.85, N = 3SE +/- 2.44, N = 3423.22410.61402.73413.52

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: Noise-GaussianDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100991021041031. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 60% Reads - Concurrency: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 58.53, N = 3SE +/- 125.77, N = 3SE +/- 35.50, N = 3SE +/- 106.96, N = 315948.616550.416635.816712.8

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.8.2Clients: 8Default - IBRSretbleed=stuffmitigations=offretbleed=off60120180240300SE +/- 0.51, N = 3SE +/- 0.28, N = 3SE +/- 0.36, N = 3SE +/- 0.43, N = 32832952882961. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -pthread -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl

Timed CPython Compilation

This test times how long it takes to build the reference Python implementation, CPython, with optimizations and LTO enabled for a release build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed CPython Compilation 3.10.6Build Configuration: DefaultDefault - IBRSretbleed=stuffmitigations=offretbleed=off91827364537.5436.7135.9436.56

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Complex PhaseDefault - IBRSretbleed=stuffmitigations=offretbleed=off100200300400500SE +/- 1.81, N = 3SE +/- 6.95, N = 3SE +/- 4.57, N = 3SE +/- 4.15, N = 3456.7458.7476.3474.6

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off1326395265SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 357.3859.6259.6659.81

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 317.4216.7716.7616.72

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.01130.02260.03390.04520.0565SE +/- 0.001, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.000, N = 30.0500.0500.0480.0501. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 1080pDefault - IBRSretbleed=stuffmitigations=offretbleed=off50100150200250SE +/- 0.29, N = 3SE +/- 0.24, N = 3SE +/- 0.43, N = 3SE +/- 0.92, N = 3207.79212.12216.17212.541. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash128Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 37.86, N = 3SE +/- 6.91, N = 3SE +/- 29.02, N = 3SE +/- 141.54, N = 1513059.2413103.2813107.9112611.701. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: FarmHash128Default - IBRSretbleed=stuffmitigations=offretbleed=off1224364860SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 1551.4151.4151.4151.411. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 18.8Time To CompileDefault - IBRSretbleed=stuffmitigations=offretbleed=off400800120016002000SE +/- 0.05, N = 3SE +/- 0.48, N = 3SE +/- 0.72, N = 3SE +/- 0.16, N = 31783.491748.871716.491734.97

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 Bettermemtier_benchmark 1.4Protocol: Redis - Clients: 100 - Set To Get Ratio: 10:1Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 16343.00, N = 3SE +/- 14429.89, N = 3SE +/- 13202.94, N = 3SE +/- 17472.63, N = 41291653.161319663.841341587.321329633.771. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Inner Join Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.91131.82262.73393.64524.5565SE +/- 0.06, N = 3SE +/- 0.04, N = 4SE +/- 0.01, N = 3SE +/- 0.04, N = 34.054.033.904.02

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Random ForestDefault - IBRSretbleed=stuffmitigations=offretbleed=off2004006008001000SE +/- 2.77, N = 3SE +/- 3.59, N = 3SE +/- 1.51, N = 3SE +/- 4.60, N = 3849.9834.5819.7825.6MIN: 707.65 / MAX: 1223.4MIN: 712.17 / MAX: 1235.53MIN: 700.08 / MAX: 1221.27MIN: 699.38 / MAX: 1198.62

JPEG XL libjxl

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

OpenBenchmarking.orgMP/s, More Is BetterJPEG XL libjxl 0.7Input: PNG - Quality: 100Default - IBRSretbleed=stuffmitigations=offretbleed=off0.12830.25660.38490.51320.6415SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.550.560.570.561. (CXX) g++ options: -fno-rtti -funwind-tables -O3 -O2 -fPIE -pie -lm -pthread -latomic

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.53, N = 3SE +/- 0.29, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 3108.1112.0111.6111.6

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU CacheDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 2.33, N = 3SE +/- 1.55, N = 3SE +/- 2.61, N = 3SE +/- 2.47, N = 3158.46152.96157.33153.841. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenRadioss

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper BeamDefault - IBRSretbleed=stuffmitigations=offretbleed=off90180270360450SE +/- 0.61, N = 3SE +/- 0.31, N = 3SE +/- 0.18, N = 3SE +/- 0.41, N = 3412.01404.13398.01403.77

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 20220729Target: CPU - Model: googlenetDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 316.8616.5416.2916.50MIN: 16.75 / MAX: 18.33MIN: 16.45 / MAX: 16.88MIN: 16.22 / MAX: 16.6MIN: 16.42 / MAX: 16.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, LosslessDefault - IBRSretbleed=stuffmitigations=offretbleed=off714212835SE +/- 0.03, N = 3SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 328.5328.0927.5727.851. (CXX) g++ options: -O3 -fPIC -lm

Unpacking The Linux Kernel

This test measures how long it takes to extract the .tar.xz Linux kernel source tree package. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterUnpacking The Linux Kernel 5.19linux-5.19.tar.xzDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.017, N = 4SE +/- 0.067, N = 4SE +/- 0.005, N = 4SE +/- 0.004, N = 49.6549.6419.3319.472

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_64Default - IBRSretbleed=stuffmitigations=offretbleed=off9K18K27K36K45KSE +/- 146.26, N = 3SE +/- 329.62, N = 3SE +/- 285.30, N = 3SE +/- 312.62, N = 339760.7938438.7739559.8139619.031. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: t1ha0_aes_avx2 x86_64Default - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.82, N = 333.7133.7033.6934.521. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingDefault - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 221.34, N = 3SE +/- 116.09, N = 3SE +/- 44.85, N = 3SE +/- 244.13, N = 3280962850729055289531. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. This MNN test profile is building the OpenMP / CPU threaded version for processor benchmarking and not any GPU-accelerated test. MNN does allow making use of AVX-512 extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 2.1Model: mobilenet-v1-1.0Default - IBRSretbleed=stuffmitigations=offretbleed=off1.02832.05663.08494.11325.1415SE +/- 0.005, N = 3SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.012, N = 34.5704.5014.4214.528MIN: 4.53 / MAX: 24.87MIN: 4.46 / MAX: 5.13MIN: 4.38 / MAX: 19.05MIN: 4.48 / MAX: 8.731. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 1080pDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.31, N = 3SE +/- 0.43, N = 3SE +/- 0.45, N = 3SE +/- 0.54, N = 3185.80189.07191.99190.251. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 204.73, N = 3SE +/- 19.47, N = 3SE +/- 106.24, N = 3SE +/- 14.57, N = 3199392014020598201151. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: ResizingDefault - IBRSretbleed=stuffmitigations=offretbleed=off90180270360450SE +/- 0.88, N = 3SE +/- 1.00, N = 33984034114101. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.14412.28823.43234.57645.7205SE +/- 0.008, N = 3SE +/- 0.017, N = 3SE +/- 0.040, N = 3SE +/- 0.035, N = 35.0854.9994.9614.9271. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 1 - Clients: 1 - Mode: Read WriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.31, N = 3SE +/- 0.68, N = 3SE +/- 1.63, N = 3SE +/- 1.43, N = 31972002022031. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Five Back to Back FIR FiltersDefault - IBRSretbleed=stuffmitigations=offretbleed=off130260390520650SE +/- 2.86, N = 3SE +/- 4.25, N = 3SE +/- 1.66, N = 3SE +/- 2.34, N = 3577.3576.2593.7593.0

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6Default - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.06, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 319.9419.7619.3619.491. (CXX) g++ options: -O3 -fPIC -lm

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Broadcast Inner Join Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.7741.5482.3223.0963.87SE +/- 0.05, N = 3SE +/- 0.02, N = 4SE +/- 0.08, N = 3SE +/- 0.04, N = 33.383.343.363.44

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Calculate Pi Benchmark Using DataframeDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.11, N = 3SE +/- 0.07, N = 4SE +/- 0.66, N = 3SE +/- 0.02, N = 324.3324.1824.8924.24

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off7001400210028003500SE +/- 4.13, N = 3SE +/- 3.18, N = 3SE +/- 1.99, N = 3SE +/- 19.88, N = 32968.923031.903056.033051.911. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4KDefault - IBRSretbleed=stuffmitigations=offretbleed=off1122334455SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.13, N = 348.2748.7649.6349.051. (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 20220729Target: CPU - Model: squeezenet_ssdDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.22, N = 322.9522.5522.3522.78MIN: 22.77 / MAX: 25.96MIN: 22.43 / MAX: 23.74MIN: 22.23 / MAX: 23.47MIN: 22.43 / MAX: 33.941. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU StressDefault - IBRSretbleed=stuffmitigations=offretbleed=off2K4K6K8K10KSE +/- 93.45, N = 3SE +/- 82.80, N = 3SE +/- 5.67, N = 3SE +/- 92.71, N = 37987.298039.918179.857966.421. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

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 20220729Target: CPU - Model: mobilenetDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.16, N = 322.1221.7521.5621.89MIN: 21.87 / MAX: 33.68MIN: 21.62 / MAX: 23.09MIN: 21.42 / MAX: 22.85MIN: 21.62 / MAX: 22.571. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 1.30, N = 3SE +/- 0.95, N = 3SE +/- 0.97, N = 3SE +/- 1.28, N = 386.0285.9585.4783.971. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off1122334455SE +/- 0.69, N = 3SE +/- 0.51, N = 3SE +/- 0.54, N = 3SE +/- 0.72, N = 346.5046.5346.7847.63MIN: 23.98 / MAX: 69.29MIN: 20.85 / MAX: 70.17MIN: 24.61 / MAX: 68.31MIN: 23.65 / MAX: 68.251. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Renaissance

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

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark BayesDefault - IBRSretbleed=stuffmitigations=offretbleed=off7001400210028003500SE +/- 43.03, N = 15SE +/- 33.97, N = 7SE +/- 43.55, N = 15SE +/- 25.04, N = 153210.13177.83215.23140.9MIN: 2206.57 / MAX: 3464.54MIN: 2429.75 / MAX: 3365.98MIN: 2315.41 / MAX: 3533.89MIN: 2328.29 / MAX: 3358.6

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.2970.5940.8911.1881.485SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 31.321.301.291.29MIN: 0.77 / MAX: 18.67MIN: 0.74 / MAX: 17.31MIN: 0.71 / MAX: 19.3MIN: 0.76 / MAX: 19.61. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVXDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 91.06, N = 3SE +/- 49.18, N = 3SE +/- 68.95, N = 3SE +/- 245.46, N = 1520898.2520869.8820829.8220433.251. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: FarmHash32 x86_64 AVXDefault - IBRSretbleed=stuffmitigations=offretbleed=off1020304050SE +/- 2.16, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.56, N = 1544.5842.4442.4343.301. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off6001200180024003000SE +/- 6.84, N = 3SE +/- 3.22, N = 3SE +/- 8.23, N = 3SE +/- 7.59, N = 32662.482703.222689.532722.581. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using DataframeDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.12, N = 3SE +/- 0.06, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 324.5424.1624.0024.20

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 311.5011.5211.2711.36

OpenRadioss

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal InstallationDefault - IBRSretbleed=stuffmitigations=offretbleed=off120240360480600SE +/- 0.10, N = 3SE +/- 0.39, N = 3SE +/- 0.05, N = 3SE +/- 0.43, N = 3565.84559.22553.46557.08

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 20220729Target: CPU - Model: vision_transformerDefault - IBRSretbleed=stuffmitigations=offretbleed=off90180270360450SE +/- 5.45, N = 3SE +/- 3.73, N = 3SE +/- 0.09, N = 3SE +/- 8.49, N = 3422.28421.62413.10422.01MIN: 413.3 / MAX: 450.85MIN: 413.1 / MAX: 442MIN: 412.4 / MAX: 425.33MIN: 412.79 / MAX: 552.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.41, N = 3SE +/- 0.28, N = 3SE +/- 2.19, N = 3SE +/- 1.11, N = 3173.82173.52177.26175.97

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: wyhashDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 227.26, N = 3SE +/- 210.32, N = 15SE +/- 200.08, N = 3SE +/- 53.25, N = 320542.6620125.3820425.0420525.161. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: wyhashDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.00, N = 3SE +/- 0.22, N = 15SE +/- 0.01, N = 3SE +/- 0.26, N = 323.8824.1023.8924.151. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Age Gender Recognition Retail 0013 FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.3330.6660.9991.3321.665SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 31.481.461.471.45MIN: 0.93 / MAX: 20.4MIN: 0.89 / MAX: 21.29MIN: 0.88 / MAX: 20.32MIN: 0.83 / MAX: 20.851. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4KDefault - IBRSretbleed=stuffmitigations=offretbleed=off1224364860SE +/- 0.06, N = 3SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.08, N = 351.1251.9352.0652.171. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: t1ha2_atonceDefault - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 141.89, N = 3SE +/- 155.43, N = 12SE +/- 138.84, N = 15SE +/- 56.28, N = 312505.3812566.6612500.3012753.071. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: t1ha2_atonceDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.00, N = 3SE +/- 0.29, N = 12SE +/- 0.03, N = 15SE +/- 0.00, N = 334.0134.3034.0334.011. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 44100 - Buffer Size: 1024Default - IBRSretbleed=stuffmitigations=offretbleed=off0.28950.5790.86851.1581.4475SE +/- 0.000148, N = 3SE +/- 0.000082, N = 3SE +/- 0.017971, N = 3SE +/- 0.000153, N = 31.2852651.2849391.2613111.2867891. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Matrix MathDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 150.56, N = 3SE +/- 237.67, N = 4SE +/- 258.64, N = 3SE +/- 44.22, N = 318152.9617799.1418017.3018087.721. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Memory CopyingDefault - IBRSretbleed=stuffmitigations=offretbleed=off2004006008001000SE +/- 8.53, N = 3SE +/- 9.22, N = 7SE +/- 9.06, N = 3SE +/- 12.29, N = 3850.74834.84834.50843.091. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Triple SHA-256, OnecoinDefault - IBRSretbleed=stuffmitigations=offretbleed=off6K12K18K24K30KSE +/- 3.33, N = 3SE +/- 313.39, N = 3SE +/- 23.33, N = 3SE +/- 3.33, N = 3266972721326723266971. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenRadioss

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop ContainerDefault - IBRSretbleed=stuffmitigations=offretbleed=off400800120016002000SE +/- 6.44, N = 3SE +/- 4.35, N = 3SE +/- 0.41, N = 3SE +/- 1.45, N = 31649.051633.501617.861626.25

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko SkylineDefault - IBRSretbleed=stuffmitigations=offretbleed=off60120180240300SE +/- 2.15, N = 3SE +/- 1.77, N = 3SE +/- 0.88, N = 3SE +/- 1.62, N = 3262.98264.63262.71259.64

PostgreSQL

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average LatencyDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.2062.4123.6184.8246.03SE +/- 0.009, N = 3SE +/- 0.035, N = 3SE +/- 0.006, N = 3SE +/- 0.030, N = 35.3225.3605.3565.2611. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

rav1e

Xiph rav1e is a Rust-written AV1 video encoder that claims to be the fastest and safest AV1 encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.6.1Speed: 10Default - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.020, N = 3SE +/- 0.006, N = 3SE +/- 0.017, N = 3SE +/- 0.026, N = 36.0736.1446.1826.182

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 20220729Target: CPU - Model: resnet18Default - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.08, N = 314.3714.2514.1214.30MIN: 14.25 / MAX: 17.15MIN: 14.14 / MAX: 15.65MIN: 14.02 / MAX: 14.36MIN: 14.1 / MAX: 29.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20220729Target: CPU - Model: resnet50Default - IBRSretbleed=stuffmitigations=offretbleed=off714212835SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 331.8431.5031.2931.50MIN: 31.6 / MAX: 45.19MIN: 31.34 / MAX: 34.28MIN: 31.16 / MAX: 32.7MIN: 31.34 / MAX: 32.411. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Inner Join Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.80551.6112.41653.2224.0275SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 33.583.563.533.52

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: FM Deemphasis FilterDefault - IBRSretbleed=stuffmitigations=offretbleed=off80160240320400SE +/- 0.35, N = 3SE +/- 1.01, N = 3SE +/- 1.09, N = 3SE +/- 0.71, N = 3356.7357.0362.6362.3

Timed CPython Compilation

This test times how long it takes to build the reference Python implementation, CPython, with optimizations and LTO enabled for a release build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed CPython Compilation 3.10.6Build Configuration: Released Build, PGO + LTO OptimizedDefault - IBRSretbleed=stuffmitigations=offretbleed=off100200300400500442.22438.60435.08436.20

PostgreSQL

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 1 - Mode: Read WriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.32, N = 3SE +/- 1.22, N = 3SE +/- 0.22, N = 3SE +/- 1.10, N = 31881871871901. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: MagiDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.03, N = 3SE +/- 1.28, N = 5SE +/- 0.17, N = 3SE +/- 0.17, N = 397.9299.4098.2498.311. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: DefaultDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.6121.2241.8362.4483.06SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 32.682.712.722.711. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 7.0.4Test: SET - Parallel Connections: 50Default - IBRSretbleed=stuffmitigations=offretbleed=off500K1000K1500K2000K2500KSE +/- 9932.25, N = 3SE +/- 10945.95, N = 3SE +/- 8978.75, N = 3SE +/- 13282.97, N = 32315119.832289592.502321265.752323685.501. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: SwirlDefault - IBRSretbleed=stuffmitigations=offretbleed=off3060901201501381391401401. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -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 1.4Encoder Mode: Preset 8 - Input: Bosphorus 4KDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 312.8012.9212.9812.971. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Contextual Anomaly Detector OSEDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.29, N = 3SE +/- 0.07, N = 3SE +/- 0.68, N = 3SE +/- 0.75, N = 392.5292.4393.6992.95

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Blake-2 SDefault - IBRSretbleed=stuffmitigations=offretbleed=off30K60K90K120K150KSE +/- 1465.07, N = 3SE +/- 1476.97, N = 3SE +/- 1213.35, N = 3SE +/- 18.56, N = 31550601559001550631537931. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.16650.3330.49950.6660.8325SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.730.730.740.741. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Random ReadDefault - IBRSretbleed=stuffmitigations=offretbleed=off4M8M12M16M20MSE +/- 180456.61, N = 3SE +/- 82599.84, N = 3SE +/- 104196.67, N = 3SE +/- 227211.26, N = 4175590911733183617471957175423851. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial time-series data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Windowed GaussianDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 320.2120.2319.9720.16

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 20220729Target: CPU - Model: alexnetDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 311.7111.6611.5611.62MIN: 11.65 / MAX: 11.92MIN: 11.59 / MAX: 11.8MIN: 11.5 / MAX: 12.06MIN: 11.57 / MAX: 12.041. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

FLAC Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.4WAV To FLACDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.05, N = 5SE +/- 0.06, N = 5SE +/- 0.13, N = 5SE +/- 0.14, N = 521.6621.4421.3921.461. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: SHA3-256Default - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.18, N = 3SE +/- 0.16, N = 3SE +/- 1.74, N = 3SE +/- 0.27, N = 3122.26122.13120.76122.251. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: SHA3-256Default - IBRSretbleed=stuffmitigations=offretbleed=off7001400210028003500SE +/- 0.08, N = 3SE +/- 0.16, N = 3SE +/- 44.34, N = 3SE +/- 5.45, N = 33226.923226.733270.423232.771. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

Selenium

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: WASM collisionDetection - Browser: FirefoxDefault - IBRSretbleed=stuffmitigations=offretbleed=off130260390520650SE +/- 0.46, N = 3SE +/- 1.11, N = 3SE +/- 0.38, N = 3SE +/- 5.12, N = 3600.1596.8597.0604.21. firefox 108.0

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: EnhancedDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100818282821. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off2040608010089.6789.5989.4290.521. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lphysicalProperties -lspecie -lfiniteVolume -lfvModels -lgenericPatchFields -lmeshTools -lsampling -lOpenFOAM -ldl -lm

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: RingcoinDefault - IBRSretbleed=stuffmitigations=offretbleed=off2004006008001000SE +/- 6.22, N = 3SE +/- 4.41, N = 3SE +/- 9.06, N = 7SE +/- 4.63, N = 3830.67822.67827.69820.721. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.69351.3872.08052.7743.4675SE +/- 0.0017, N = 3SE +/- 0.0054, N = 3SE +/- 0.0019, N = 3SE +/- 0.0082, N = 33.08243.04643.05873.0623

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off70140210280350SE +/- 0.18, N = 3SE +/- 0.58, N = 3SE +/- 0.20, N = 3SE +/- 0.87, N = 3324.41328.24326.93326.54

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.17, N = 3SE +/- 0.24, N = 3SE +/- 0.25, N = 3SE +/- 0.57, N = 3130.9130.9130.8129.41. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: AtomicDefault - IBRSretbleed=stuffmitigations=offretbleed=off40K80K120K160K200KSE +/- 2552.57, N = 15SE +/- 2632.45, N = 15SE +/- 2604.63, N = 15SE +/- 2454.73, N = 15180741.36182114.03180081.27180772.461. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: GarlicoinDefault - IBRSretbleed=stuffmitigations=offretbleed=off2004006008001000SE +/- 2.76, N = 3SE +/- 6.67, N = 3SE +/- 2.68, N = 3SE +/- 1.78, N = 31140.681153.431140.861144.591. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc C String FunctionsDefault - IBRSretbleed=stuffmitigations=offretbleed=off50K100K150K200K250KSE +/- 443.77, N = 3SE +/- 612.75, N = 3SE +/- 1033.29, N = 3SE +/- 488.89, N = 3243451.44243636.88240956.56243127.331. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off70140210280350SE +/- 0.47, N = 3SE +/- 0.71, N = 3SE +/- 1.16, N = 3SE +/- 0.71, N = 3322.63323.13325.92326.19

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.69741.39482.09222.78963.487SE +/- 0.0045, N = 3SE +/- 0.0068, N = 3SE +/- 0.0109, N = 3SE +/- 0.0067, N = 33.09943.09463.06823.0656

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Quad SHA-256, PyriteDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 176.67, N = 3SE +/- 3.33, N = 3SE +/- 193.10, N = 8192671909719299191001. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

rav1e

Xiph rav1e is a Rust-written AV1 video encoder that claims to be the fastest and safest AV1 encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.6.1Speed: 5Default - IBRSretbleed=stuffmitigations=offretbleed=off0.40910.81821.22731.63642.0455SE +/- 0.003, N = 3SE +/- 0.004, N = 3SE +/- 0.003, N = 3SE +/- 0.006, N = 31.8021.7991.8021.818

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.11, N = 327.5427.4027.4127.27

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 2Default - IBRSretbleed=stuffmitigations=offretbleed=off4080120160200SE +/- 0.26, N = 3SE +/- 0.27, N = 3SE +/- 0.56, N = 3SE +/- 0.22, N = 3171.47171.57170.09169.841. (CXX) g++ options: -O3 -fPIC -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.71861.43722.15582.87443.593SE +/- 0.0024, N = 3SE +/- 0.0058, N = 3SE +/- 0.0080, N = 3SE +/- 0.0115, N = 33.19393.16213.16663.1660

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100Default - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 39.039.099.109.121. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread

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 20220729Target: CPU - Model: yolov4-tinyDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.09, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.10, N = 332.5132.3132.1932.42MIN: 32.3 / MAX: 35.45MIN: 32.17 / MAX: 40.14MIN: 32.06 / MAX: 35.39MIN: 32.2 / MAX: 35.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off1632486480SE +/- 0.01, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.29, N = 372.5972.9772.9373.31

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off140280420560700SE +/- 0.48, N = 3SE +/- 1.07, N = 3SE +/- 1.60, N = 3SE +/- 2.30, N = 3626.17632.38631.59631.70

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: x25xDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 1.00, N = 3SE +/- 1.84, N = 3SE +/- 1.22, N = 3SE +/- 1.01, N = 3156.44157.79156.26157.181. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 8 - Input: Bosphorus 1080pDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.11, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 336.0236.2636.3636.341. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 312.3712.2512.3412.33MIN: 12.21MIN: 12.02MIN: 12.15MIN: 12.171. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: Spooky32Default - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 10.25, N = 3SE +/- 9.32, N = 3SE +/- 38.37, N = 3SE +/- 105.88, N = 312205.3612211.4812101.0512124.071. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: Spooky32Default - IBRSretbleed=stuffmitigations=offretbleed=off1122334455SE +/- 0.00, N = 3SE +/- 0.25, N = 3SE +/- 0.24, N = 3SE +/- 0.01, N = 350.0350.2750.2750.031. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder that claims to be the fastest and safest AV1 encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.6.1Speed: 6Default - IBRSretbleed=stuffmitigations=offretbleed=off0.55081.10161.65242.20322.754SE +/- 0.006, N = 3SE +/- 0.006, N = 3SE +/- 0.006, N = 3SE +/- 0.005, N = 32.4262.4482.4482.435

Selenium

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: WASM imageConvolute - Browser: FirefoxDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.10, N = 3SE +/- 0.17, N = 3SE +/- 0.09, N = 3SE +/- 0.09, N = 333.233.133.233.41. firefox 108.0

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off1.24052.4813.72154.9626.2025SE +/- 0.01808, N = 3SE +/- 0.00519, N = 3SE +/- 0.01647, N = 3SE +/- 0.00478, N = 35.513325.483385.495795.46550MIN: 5.41MIN: 5.41MIN: 5.41MIN: 5.41. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.30, N = 3SE +/- 0.30, N = 3SE +/- 0.12, N = 3SE +/- 0.30, N = 396.5195.8496.3895.68

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 310.3610.4310.3710.45

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off7001400210028003500SE +/- 10.60, N = 3SE +/- 21.15, N = 3SE +/- 7.27, N = 3SE +/- 4.07, N = 33306.203287.933278.213285.23MIN: 3204.79 / MAX: 3381.05MIN: 3168.84 / MAX: 3380.01MIN: 3175.59 / MAX: 3371.08MIN: 3201.26 / MAX: 3383.31. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

libavif avifenc

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

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 0Default - IBRSretbleed=stuffmitigations=offretbleed=off80160240320400SE +/- 0.22, N = 3SE +/- 0.48, N = 3SE +/- 1.22, N = 3SE +/- 0.25, N = 3379.22380.58379.28377.401. (CXX) g++ options: -O3 -fPIC -lm

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.27230.54460.81691.08921.3615SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 31.201.211.211.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 44100 - Buffer Size: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off0.27390.54780.82171.09561.3695SE +/- 0.000342, N = 3SE +/- 0.000113, N = 3SE +/- 0.008493, N = 3SE +/- 0.000124, N = 31.2157411.2174031.2074181.2173221. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.20, N = 3SE +/- 0.65, N = 3110.0110.4110.4109.51. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 480000 - Buffer Size: 1024Default - IBRSretbleed=stuffmitigations=offretbleed=off0.28170.56340.84511.12681.4085SE +/- 0.005641, N = 3SE +/- 0.000662, N = 3SE +/- 0.000583, N = 3SE +/- 0.000223, N = 31.2419261.2511871.2517961.2517741. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, LosslessDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.29480.58960.88441.17921.474SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.301.311.311.311. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off1020304050SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 341.3541.6741.6341.66

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 324.1823.9924.0224.00

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.09, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.05, N = 313.2013.1413.1813.241. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.46, N = 3SE +/- 0.70, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 3106.2105.4105.7105.41. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off11002200330044005500SE +/- 1.25, N = 3SE +/- 1.80, N = 3SE +/- 3.89, N = 3SE +/- 2.51, N = 35354.185337.235313.905325.14MIN: 5048.06 / MAX: 5756.34MIN: 5005.81 / MAX: 5704.35MIN: 4979.6 / MAX: 5629.68MIN: 5015.25 / MAX: 5648.971. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Machine Translation EN To DE FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off70140210280350SE +/- 1.98, N = 3SE +/- 0.65, N = 3SE +/- 1.48, N = 3SE +/- 1.21, N = 3302.54303.87303.09301.63MIN: 242.68 / MAX: 511.74MIN: 164.49 / MAX: 327.96MIN: 253.91 / MAX: 325.11MIN: 253.21 / MAX: 326.21. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: Myriad-GroestlDefault - IBRSretbleed=stuffmitigations=offretbleed=off12002400360048006000SE +/- 2.14, N = 3SE +/- 24.39, N = 3SE +/- 17.24, N = 3SE +/- 2.34, N = 35371.745400.625411.585374.931. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off2004006008001000916.44913.50909.74910.871. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lphysicalProperties -lspecie -lfiniteVolume -lfvModels -lgenericPatchFields -lmeshTools -lsampling -lOpenFOAM -ldl -lm

LuaRadio

LuaRadio is a lightweight software-defined radio (SDR) framework built atop LuaJIT. LuaRadio provides a suite of source, sink, and processing blocks, with a simple API for defining flow graphs, running flow graphs, creating blocks, and creating data types. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterLuaRadio 0.9.1Test: Hilbert TransformDefault - IBRSretbleed=stuffmitigations=offretbleed=off1530456075SE +/- 0.10, N = 3SE +/- 0.19, N = 3SE +/- 0.15, N = 3SE +/- 0.13, N = 368.268.768.468.4

OpenRadioss

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on WindshieldDefault - IBRSretbleed=stuffmitigations=offretbleed=off160320480640800SE +/- 0.01, N = 3SE +/- 0.53, N = 3SE +/- 0.74, N = 3SE +/- 0.27, N = 3735.24732.52729.94731.86

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.38, N = 3SE +/- 0.43, N = 3SE +/- 0.26, N = 3SE +/- 0.06, N = 3136.56137.20136.24137.211. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenRadioss

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop TestDefault - IBRSretbleed=stuffmitigations=offretbleed=off70140210280350SE +/- 0.54, N = 3SE +/- 0.36, N = 3SE +/- 0.38, N = 3SE +/- 0.17, N = 3308.94307.34306.77307.30

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Vehicle Detection FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off714212835SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 329.2729.1429.3429.14MIN: 16.56 / MAX: 45.71MIN: 16.05 / MAX: 46.79MIN: 15.88 / MAX: 47.74MIN: 16.39 / MAX: 45.231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 317.4917.4117.3917.37

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off70140210280350SE +/- 0.55, N = 3SE +/- 0.26, N = 3SE +/- 0.19, N = 3SE +/- 1.24, N = 3323.0321.9322.7320.81. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.13, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3114.35114.88114.96115.13

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Highest CompressionDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.69531.39062.08592.78123.4765SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 33.073.093.093.081. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression RatingDefault - IBRSretbleed=stuffmitigations=offretbleed=off5K10K15K20K25KSE +/- 57.42, N = 3SE +/- 55.21, N = 3SE +/- 132.11, N = 3SE +/- 7.81, N = 3221692216022027221141. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off70140210280350SE +/- 0.85, N = 3SE +/- 0.31, N = 3SE +/- 0.26, N = 3SE +/- 0.39, N = 3328.6329.2328.5327.11. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 4 - Input: Bosphorus 1080pDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.85231.70462.55693.40924.2615SE +/- 0.002, N = 3SE +/- 0.007, N = 3SE +/- 0.007, N = 3SE +/- 0.003, N = 33.7643.7763.7883.7811. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: DefaultDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 314.1614.2114.2014.251. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.61, N = 3SE +/- 0.13, N = 3SE +/- 0.46, N = 3SE +/- 0.66, N = 3115.9115.8115.9115.21. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

BRL-CAD

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

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.32.6VGR Performance MetricDefault - IBRSretbleed=stuffmitigations=offretbleed=off11K22K33K44K55K500865006450104503581. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -pthread -ldl -lm

SVT-AV1

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

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 4 - Input: Bosphorus 4KDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.24080.48160.72240.96321.204SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 31.0641.0681.0701.0691. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Blender

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.4Blend File: BMW27 - Compute: CPU-OnlyDefault - IBRSretbleed=stuffmitigations=offretbleed=off80160240320400SE +/- 0.15, N = 3SE +/- 0.72, N = 3SE +/- 0.52, N = 3SE +/- 0.06, N = 3369.61368.09368.70367.55

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc Qsort Data SortingDefault - IBRSretbleed=stuffmitigations=offretbleed=off1326395265SE +/- 0.13, N = 3SE +/- 0.08, N = 3SE +/- 0.27, N = 3SE +/- 0.22, N = 355.5555.8655.8255.731. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off11002200330044005500SE +/- 1.67, N = 3SE +/- 2.07, N = 3SE +/- 2.90, N = 3SE +/- 4.26, N = 35351.085339.565323.795332.52MIN: 5029.61 / MAX: 5757.54MIN: 5008.95 / MAX: 5703.69MIN: 4982.73 / MAX: 5635.02MIN: 5010.51 / MAX: 5668.141. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: DeepcoinDefault - IBRSretbleed=stuffmitigations=offretbleed=off8001600240032004000SE +/- 0.76, N = 3SE +/- 0.36, N = 3SE +/- 24.38, N = 3SE +/- 3.29, N = 33840.303840.873858.753845.261. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.09, N = 3SE +/- 0.25, N = 3SE +/- 0.07, N = 3SE +/- 0.08, N = 3143.09143.74143.46143.41

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 313.9813.9113.9413.94

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NIDefault - IBRSretbleed=stuffmitigations=offretbleed=off7K14K21K28K35KSE +/- 91.91, N = 3SE +/- 20.33, N = 3SE +/- 137.62, N = 3SE +/- 31.93, N = 332896.5132807.2032954.0432887.581. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: MeowHash x86_64 AES-NIDefault - IBRSretbleed=stuffmitigations=offretbleed=off1428425670SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 364.5164.4864.4964.491. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

miniBUDE

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

OpenBenchmarking.orgBillion Interactions/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Default - IBRSretbleed=stuffmitigations=offretbleed=off1.12912.25823.38734.51645.6455SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 35.0175.0174.9965.0181. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

OpenBenchmarking.orgGFInst/s, More Is BetterminiBUDE 20210901Implementation: OpenMP - Input Deck: BM1Default - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3125.42125.43124.91125.451. (CC) gcc options: -std=c99 -Ofast -ffast-math -fopenmp -march=native -lm

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 314.4414.4814.5014.50MIN: 14.25MIN: 14.33MIN: 14.35MIN: 14.361. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: LBC, LBRY CreditsDefault - IBRSretbleed=stuffmitigations=offretbleed=off2K4K6K8K10KSE +/- 40.00, N = 3SE +/- 16.67, N = 3SE +/- 15.28, N = 3105501059010577105801. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off1224364860SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.06, N = 353.053.153.052.91. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 96000 - Buffer Size: 1024Default - IBRSretbleed=stuffmitigations=offretbleed=off0.20920.41840.62760.83681.046SE +/- 0.000224, N = 3SE +/- 0.000807, N = 3SE +/- 0.000455, N = 3SE +/- 0.000035, N = 30.9289780.9265040.9275380.9299551. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.47, N = 3SE +/- 0.18, N = 3SE +/- 0.20, N = 3SE +/- 0.77, N = 3137.2137.1136.7137.21. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 334.8134.9234.9334.93

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off714212835SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 328.7228.6228.6128.61

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 96000 - Buffer Size: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off0.19340.38680.58020.77360.967SE +/- 0.000732, N = 3SE +/- 0.000670, N = 3SE +/- 0.000243, N = 3SE +/- 0.000460, N = 30.8597590.8567600.8577440.8594121. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.16, N = 3SE +/- 0.07, N = 381.7682.0282.0082.04

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 312.2312.1912.1912.19

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off140280420560700SE +/- 0.17, N = 3SE +/- 0.64, N = 3SE +/- 0.59, N = 3SE +/- 0.72, N = 3625.93624.35626.49626.06

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.72071.44142.16212.88283.6035SE +/- 0.0008, N = 3SE +/- 0.0033, N = 3SE +/- 0.0031, N = 3SE +/- 0.0037, N = 33.19513.20323.19233.1945

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Calculate Pi BenchmarkDefault - IBRSretbleed=stuffmitigations=offretbleed=off90180270360450SE +/- 0.18, N = 3SE +/- 1.00, N = 4SE +/- 0.45, N = 3SE +/- 0.22, N = 3436.79437.77436.93436.32

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Calculate Pi BenchmarkDefault - IBRSretbleed=stuffmitigations=offretbleed=off100200300400500SE +/- 0.93, N = 3SE +/- 0.48, N = 3SE +/- 1.14, N = 3SE +/- 0.28, N = 3441.23439.91441.21439.84

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off80160240320400SE +/- 0.99, N = 3SE +/- 0.30, N = 3SE +/- 0.12, N = 3SE +/- 0.19, N = 3354.4353.7353.6354.71. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off1224364860SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 354.5654.4254.4154.39

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 336.6436.7436.7436.76

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples / Second, More Is BettersrsRAN 22.04.1Test: OFDM_TestDefault - IBRSretbleed=stuffmitigations=offretbleed=off20M40M60M80M100MSE +/- 317979.73, N = 3SE +/- 450924.98, N = 3SE +/- 348010.22, N = 3SE +/- 409606.86, N = 31023666671024000001023666671021333331. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMDefault - IBRSretbleed=stuffmitigations=offretbleed=off80160240320400SE +/- 0.38, N = 3SE +/- 0.64, N = 3SE +/- 0.43, N = 3SE +/- 0.64, N = 3359.6359.0359.7359.91. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -lpthread -ldl -lm

Xmrig

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

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Wownero - Hash Count: 1MDefault - IBRSretbleed=stuffmitigations=offretbleed=off400800120016002000SE +/- 1.18, N = 3SE +/- 1.62, N = 3SE +/- 0.87, N = 3SE +/- 2.44, N = 32030.72031.12033.82029.01. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: scryptDefault - IBRSretbleed=stuffmitigations=offretbleed=off1632486480SE +/- 0.01, N = 3SE +/- 0.16, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 372.5572.7172.6472.611. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

nekRS

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

OpenBenchmarking.orgFLOP/s, More Is BetternekRS 22.0Input: TurboPipe PeriodicDefault - IBRSretbleed=stuffmitigations=offretbleed=off4000M8000M12000M16000M20000MSE +/- 6547518.61, N = 3SE +/- 26938345.24, N = 3SE +/- 7918543.64, N = 3SE +/- 11775822.69, N = 3192638000001922676666719262600000192500000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -pthread -lmpi_cxx -lmpi

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Vector MathDefault - IBRSretbleed=stuffmitigations=offretbleed=off3K6K9K12K15KSE +/- 3.47, N = 3SE +/- 8.67, N = 3SE +/- 1.97, N = 3SE +/- 25.94, N = 314401.8514404.7314411.6014384.511. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

Stargate Digital Audio Workstation

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

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 480000 - Buffer Size: 512Default - IBRSretbleed=stuffmitigations=offretbleed=off0.26620.53240.79861.06481.331SE +/- 0.000675, N = 3SE +/- 0.000431, N = 3SE +/- 0.000280, N = 3SE +/- 0.000339, N = 31.1810601.1811431.1832431.1818811. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off9001800270036004500SE +/- 1.74, N = 3SE +/- 0.74, N = 3SE +/- 4.77, N = 3SE +/- 1.52, N = 33967.413968.473963.593961.26MIN: 3960.54MIN: 3960.07MIN: 3954MIN: 3956.981. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 318.2818.2718.2618.29MIN: 10 / MAX: 37.03MIN: 9.77 / MAX: 32.89MIN: 9.82 / MAX: 32.91MIN: 9.75 / MAX: 33.941. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

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 20220729Target: CPU - Model: vgg16Default - IBRSretbleed=stuffmitigations=offretbleed=off20406080100SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 378.3078.3878.2678.35MIN: 78.11 / MAX: 81.22MIN: 78.15 / MAX: 90.52MIN: 78.05 / MAX: 89.98MIN: 78.11 / MAX: 89.131. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread -pthread

Xmrig

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

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.18.1Variant: Monero - Hash Count: 1MDefault - IBRSretbleed=stuffmitigations=offretbleed=off30060090012001500SE +/- 0.41, N = 3SE +/- 0.94, N = 3SE +/- 1.27, N = 3SE +/- 1.47, N = 31429.81428.61430.41428.31. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off246810SE +/- 0.00789, N = 3SE +/- 0.00849, N = 3SE +/- 0.01398, N = 3SE +/- 0.01002, N = 37.686447.696147.693507.68653MIN: 7.53MIN: 7.55MIN: 7.53MIN: 7.531. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off50100150200250SE +/- 0.06, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.14, N = 3218.59218.69218.83218.561. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.11, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3113.01112.97113.00112.881. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Stress-NG

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

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CryptoDefault - IBRSretbleed=stuffmitigations=offretbleed=off10002000300040005000SE +/- 3.82, N = 3SE +/- 0.75, N = 3SE +/- 6.94, N = 3SE +/- 3.32, N = 34815.304817.254816.674812.471. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -laio -lapparmor -latomic -lbsd -lc -lcrypt -ldl -lEGL -lgbm -lGLESv2 -ljpeg -lrt -lsctp -lz -pthread

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 335.4035.3835.3835.41MIN: 29.04 / MAX: 52.04MIN: 25.81 / MAX: 52.06MIN: 27.55 / MAX: 52.33MIN: 25.78 / MAX: 52.571. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

Cpuminer-Opt

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

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.20.3Algorithm: SkeincoinDefault - IBRSretbleed=stuffmitigations=offretbleed=off4K8K12K16K20KSE +/- 3.33, N = 3SE +/- 3.33, N = 3SE +/- 3.33, N = 3163601634716353163531. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenVINO

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

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off400800120016002000SE +/- 0.52, N = 3SE +/- 0.78, N = 3SE +/- 0.01, N = 3SE +/- 1.34, N = 31797.131797.931798.301798.51MIN: 1747.57 / MAX: 1809.8MIN: 1733.26 / MAX: 1816.74MIN: 1724.05 / MAX: 1818.82MIN: 1690.1 / MAX: 1919.151. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 320.9220.9220.9320.91MIN: 20.86MIN: 20.86MIN: 20.84MIN: 20.851. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off16003200480064008000SE +/- 2.05, N = 3SE +/- 2.32, N = 3SE +/- 3.51, N = 3SE +/- 3.60, N = 37409.457409.077405.607405.04MIN: 7399.64MIN: 7401.09MIN: 7394.43MIN: 7395.871. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

SMHasher

SMHasher is a hash function tester supporting various algorithms and able to make use of AVX and other modern CPU instruction set extensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/sec, More Is BetterSMHasher 2022-08-22Hash: fasthash32Default - IBRSretbleed=stuffmitigations=offretbleed=off13002600390052006500SE +/- 1.21, N = 3SE +/- 2.06, N = 3SE +/- 1.39, N = 3SE +/- 1.00, N = 36164.116164.446165.806164.711. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread
OpenBenchmarking.orgcycles/hash, Fewer Is BetterSMHasher 2022-08-22Hash: fasthash32Default - IBRSretbleed=stuffmitigations=offretbleed=off816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.39, N = 3SE +/- 0.00, N = 336.3936.3936.8036.391. (CXX) g++ options: -march=native -O3 -flto -fno-fat-lto-objects -lpthread

OpenVINO

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

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.49950.9991.49851.9982.4975SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.222.222.222.221. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP32 - Device: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.16430.32860.49290.65720.8215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.730.730.730.731. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -shared

OpenVKL

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

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ScalarDefault - IBRSretbleed=stuffmitigations=offretbleed=off81624324036363636MIN: 4 / MAX: 701MIN: 4 / MAX: 700MIN: 4 / MAX: 701MIN: 4 / MAX: 701

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPCDefault - IBRSretbleed=stuffmitigations=offretbleed=off1530456075SE +/- 0.33, N = 367676767MIN: 8 / MAX: 1035MIN: 8 / MAX: 1039MIN: 8 / MAX: 1038MIN: 8 / MAX: 1036

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.38Operation: SharpenDefault - IBRSretbleed=stuffmitigations=offretbleed=off918273645393939391. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Quality 95, Compression Effort 7Default - IBRSretbleed=stuffmitigations=offretbleed=off0.00450.0090.01350.0180.0225SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.020.020.020.021. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Quality 75, Compression Effort 7Default - IBRSretbleed=stuffmitigations=offretbleed=off0.01130.02260.03390.04520.0565SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.050.050.050.051. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl -lpthread

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Lossless, Highest CompressionDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.09450.1890.28350.3780.4725SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.420.420.420.421. (CC) gcc options: -fvisibility=hidden -O2 -lm -pthread

Natron

Natron is an open-source, cross-platform compositing software for visual effects (VFX) and motion graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterNatron 2.4.3Input: SpaceshipDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.270.540.811.081.35SE +/- 0.02, N = 15SE +/- 0.01, N = 12SE +/- 0.01, N = 12SE +/- 0.00, N = 31.20.70.70.8

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 Bettermemtier_benchmark 1.4Protocol: Redis - Clients: 50 - Set To Get Ratio: 10:1Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 14120.93, N = 3SE +/- 11938.38, N = 3SE +/- 27491.98, N = 15SE +/- 20904.07, N = 121109992.571180753.401288891.751220692.331. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenBenchmarking.orgOps/sec, More Is Bettermemtier_benchmark 1.4Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:1Default - IBRSretbleed=stuffmitigations=offretbleed=off300K600K900K1200K1500KSE +/- 10509.17, N = 11SE +/- 17147.18, N = 4SE +/- 48717.93, N = 15SE +/- 18395.69, N = 151175178.961232491.421481936.111329151.411. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

CockroachDB

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

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: MoVR - Concurrency: 256Default - IBRSretbleed=stuffmitigations=offretbleed=off306090120150SE +/- 0.31, N = 3SE +/- 0.35, N = 3SE +/- 0.15, N = 3SE +/- 4.63, N = 12108.5111.9111.2105.9

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test TimeDefault - IBRSretbleed=stuffmitigations=offretbleed=off0.71781.43562.15342.87123.589SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.22, N = 3SE +/- 0.06, N = 33.142.993.192.98

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ProcessDefault - IBRSretbleed=stuffmitigations=offretbleed=off612182430SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.52, N = 15SE +/- 0.03, N = 325.0519.1611.3616.131. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ProcessDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.011, N = 3SE +/- 0.012, N = 3SE +/- 0.260, N = 15SE +/- 0.004, N = 312.8069.8766.4978.3031. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ThreadDefault - IBRSretbleed=stuffmitigations=offretbleed=off1326395265SE +/- 0.30, N = 3SE +/- 1.00, N = 15SE +/- 0.95, N = 15SE +/- 0.04, N = 359.1741.5223.6533.501. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ThreadDefault - IBRSretbleed=stuffmitigations=offretbleed=off714212835SE +/- 0.62, N = 15SE +/- 0.78, N = 15SE +/- 0.90, N = 15SE +/- 0.69, N = 1527.9821.5615.1018.291. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ThreadDefault - IBRSretbleed=stuffmitigations=offretbleed=off3691215SE +/- 0.004, N = 3SE +/- 0.158, N = 15SE +/- 0.077, N = 3SE +/- 0.161, N = 1513.04510.5777.9558.9701. (CC) gcc options: -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUDefault - IBRSretbleed=stuffmitigations=offretbleed=off48121620SE +/- 0.26, N = 15SE +/- 0.28, N = 15SE +/- 0.25, N = 3SE +/- 0.25, N = 1514.1414.0614.9113.54MIN: 11.76MIN: 11.74MIN: 11.77MIN: 11.731. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

DaCapo Benchmark

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

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansDefault - IBRSretbleed=stuffmitigations=offretbleed=off9001800270036004500SE +/- 56.91, N = 20SE +/- 44.97, N = 20SE +/- 50.26, N = 20SE +/- 44.62, N = 204195387839273923

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2Default - IBRSretbleed=stuffmitigations=offretbleed=off8001600240032004000SE +/- 55.09, N = 20SE +/- 47.29, N = 4SE +/- 42.46, N = 20SE +/- 65.06, N = 163763353836303686

Sockperf

This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterSockperf 3.7Test: ThroughputDefault - IBRSretbleed=stuffmitigations=offretbleed=off150K300K450K600K750KSE +/- 3153.77, N = 5SE +/- 3430.46, N = 25SE +/- 8501.97, N = 25SE +/- 6503.42, N = 53382904025266909765333131. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread

351 Results Shown

Stress-NG:
  x86_64 RdRand
  System V Message Passing
  Malloc
  Mutex
Hackbench:
  8 - Process
  8 - Thread
  4 - Process
  16 - Process
  16 - Thread
ctx_clock
Stress-NG
PostMark
KeyDB
Stress-NG:
  Futex
  Context Switching
OSBench
Sockperf
Memcached
Stress-NG:
  SENDFILE
  MEMFD
Memcached
OSBench
Memcached
OSBench
GraphicsMagick
Memcached
Renaissance
OSBench
Facebook RocksDB
EnCodec:
  24 kbps
  1.5 kbps
  6 kbps
Stress-NG
EnCodec
Facebook RocksDB
Stress-NG
GraphicsMagick
NCNN
OSBench
NCNN
Facebook RocksDB
nginx
Renaissance
nginx
Renaissance
Mobile Neural Network
PostgreSQL:
  1 - 50 - Read Only
  1 - 50 - Read Only - Average Latency
nginx
Dragonflydb
nginx
PostgreSQL
nginx
PostgreSQL
Renaissance
nginx
PostgreSQL:
  100 - 50 - Read Only - Average Latency
  100 - 50 - Read Only
JPEG XL Decoding libjxl
JPEG XL libjxl
PostgreSQL
Renaissance
PostgreSQL
JPEG XL libjxl:
  JPEG - 90
  PNG - 80
JPEG XL Decoding libjxl
JPEG XL libjxl
Facebook RocksDB
Mobile Neural Network
Dragonflydb
Mobile Neural Network:
  squeezenetv1.1
  resnet-v2-50
Stress-NG
Renaissance
memtier_benchmark
PostgreSQL:
  100 - 100 - Read Write - Average Latency
  100 - 100 - Read Write
Renaissance
NCNN
Dragonflydb
NCNN
PostgreSQL:
  100 - 50 - Read Write
  100 - 50 - Read Write - Average Latency
Timed Erlang/OTP Compilation
Dragonflydb
Timed Linux Kernel Compilation
Apache Spark
memtier_benchmark
PostgreSQL
Mobile Neural Network
Redis
NCNN:
  CPU - mnasnet
  CPU-v3-v3 - mobilenet-v3
Mobile Neural Network
PostgreSQL
Dragonflydb
ClickHouse
Apache Spark
PostgreSQL
Mobile Neural Network
Stress-NG
PostgreSQL
Apache Spark:
  1000000 - 500 - Group By Test Time
  1000000 - 100 - Group By Test Time
NCNN
ClickHouse
libavif avifenc
CockroachDB:
  KV, 95% Reads - 128
  KV, 50% Reads - 128
  KV, 60% Reads - 128
  KV, 95% Reads - 256
Renaissance
memtier_benchmark
Timed PHP Compilation
Selenium
CockroachDB:
  KV, 50% Reads - 256
  KV, 10% Reads - 128
PostgreSQL
CockroachDB:
  KV, 50% Reads - 512
  KV, 60% Reads - 256
NCNN
PostgreSQL
CockroachDB
DaCapo Benchmark
Facebook RocksDB
JPEG XL libjxl
Renaissance
Selenium
CockroachDB
Selenium
MariaDB
CockroachDB
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    FPS
    ms
Apache Spark
Facebook RocksDB
DaCapo Benchmark
CockroachDB
Apache Spark
ClickHouse
Numenta Anomaly Benchmark
GraphicsMagick
CockroachDB
MariaDB
Timed CPython Compilation
LuaRadio
Neural Magic DeepSparse:
  CV Detection,YOLOv5s COCO - Synchronous Single-Stream:
    ms/batch
    items/sec
PostgreSQL
SVT-AV1
SMHasher
Timed Node.js Compilation
memtier_benchmark
Apache Spark
Renaissance
JPEG XL libjxl
CockroachDB
Stress-NG
OpenRadioss
NCNN
libavif avifenc
Unpacking The Linux Kernel
SMHasher
7-Zip Compression
Mobile Neural Network
SVT-AV1
PostgreSQL
GraphicsMagick
PostgreSQL:
  1 - 1 - Read Write - Average Latency
  1 - 1 - Read Write
LuaRadio
libavif avifenc
Apache Spark:
  1000000 - 500 - Broadcast Inner Join Test Time
  1000000 - 500 - Calculate Pi Benchmark Using Dataframe
OpenVINO
SVT-AV1
NCNN
Stress-NG
NCNN
OpenVINO:
  Vehicle Detection FP16 - CPU:
    FPS
    ms
Renaissance
OpenVINO
SMHasher
OpenVINO
Apache Spark
Neural Magic DeepSparse
OpenRadioss
NCNN
Neural Magic DeepSparse
SMHasher
OpenVINO
SVT-AV1
SMHasher
Stargate Digital Audio Workstation
Stress-NG:
  Matrix Math
  Memory Copying
Cpuminer-Opt
OpenRadioss
Numenta Anomaly Benchmark
PostgreSQL
rav1e
NCNN:
  CPU - resnet18
  CPU - resnet50
Apache Spark
LuaRadio
Timed CPython Compilation
PostgreSQL
Cpuminer-Opt
WebP2 Image Encode
Redis
GraphicsMagick
SVT-AV1
Numenta Anomaly Benchmark
Cpuminer-Opt
OpenVINO
Facebook RocksDB
Numenta Anomaly Benchmark
NCNN
FLAC Audio Encoding
SMHasher
Selenium
GraphicsMagick
OpenFOAM
Cpuminer-Opt
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    items/sec
    ms/batch
srsRAN
Stress-NG
Cpuminer-Opt
Stress-NG
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
Cpuminer-Opt
rav1e
Neural Magic DeepSparse
libavif avifenc
Neural Magic DeepSparse
WebP Image Encode
NCNN
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream
Cpuminer-Opt
SVT-AV1
oneDNN
SMHasher
rav1e
Selenium
oneDNN
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenVINO
libavif avifenc
OpenVINO
Stargate Digital Audio Workstation
srsRAN
Stargate Digital Audio Workstation
WebP Image Encode
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    ms/batch
    items/sec
OpenVINO
srsRAN
OpenVINO:
  Person Detection FP16 - CPU
  Machine Translation EN To DE FP16 - CPU
Cpuminer-Opt
OpenFOAM
LuaRadio
OpenRadioss
OpenVINO
OpenRadioss
OpenVINO
Neural Magic DeepSparse
srsRAN
Neural Magic DeepSparse
WebP Image Encode
7-Zip Compression
srsRAN
SVT-AV1
WebP Image Encode
srsRAN
BRL-CAD
SVT-AV1
Blender
Stress-NG
OpenVINO
Cpuminer-Opt
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
SMHasher
miniBUDE:
  OpenMP - BM1:
    Billion Interactions/s
    GFInst/s
oneDNN
Cpuminer-Opt
srsRAN
Stargate Digital Audio Workstation
srsRAN
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    items/sec
    ms/batch
Stargate Digital Audio Workstation
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream:
    ms/batch
    items/sec
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Apache Spark:
  1000000 - 500 - Calculate Pi Benchmark
  1000000 - 100 - Calculate Pi Benchmark
srsRAN
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream:
    ms/batch
    items/sec
srsRAN:
  OFDM_Test
  4G PHY_DL_Test 100 PRB SISO 256-QAM
Xmrig
Cpuminer-Opt
nekRS
Stress-NG
Stargate Digital Audio Workstation
oneDNN
OpenVINO
NCNN
Xmrig
oneDNN
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU
  Weld Porosity Detection FP16 - CPU
Stress-NG
OpenVINO
Cpuminer-Opt
OpenVINO
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Recurrent Neural Network Training - f32 - CPU
SMHasher
OpenVINO:
  Face Detection FP16-INT8 - CPU
  Person Detection FP32 - CPU
OpenVKL:
  vklBenchmark Scalar
  vklBenchmark ISPC
GraphicsMagick
WebP2 Image Encode:
  Quality 95, Compression Effort 7
  Quality 75, Compression Effort 7
WebP Image Encode
Natron
memtier_benchmark:
  Redis - 50 - 10:1
  Redis - 50 - 1:1
CockroachDB
Apache Spark
Hackbench:
  2 - Process
  1 - Process
  4 - Thread
  2 - Thread
  1 - Thread
oneDNN
DaCapo Benchmark:
  Tradebeans
  H2
Sockperf

Default - IBRS

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + 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 PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 20 December 2022 15:14 by user phoronix.

retbleed=stuff

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Stuffing + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 22 December 2022 06:00 by user phoronix.

mitigations=off

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: vulnerable + mds: Vulnerable; SMT vulnerable + meltdown: Vulnerable + mmio_stale_data: Vulnerable + retbleed: Vulnerable + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Not affected + srbds: Vulnerable + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 23 December 2022 15:05 by user phoronix.

retbleed=off

Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (4 Cores / 8 Threads), Motherboard: MSI Z170A SLI PLUS (MS-7998) v1.0 (2.A0 BIOS), Chipset: Intel Xeon E3-1200 v5/E3-1500, Memory: 32GB, Disk: 256GB TOSHIBA RD400, Graphics: ASUS AMD Radeon HD 7850 / R7 265 R9 270 1024SP, Audio: Realtek ALC1150, Monitor: VA2431, Network: Intel I219-V

OS: Ubuntu 20.04, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 3.36.9, Display Server: X Server 1.20.13, OpenGL: 4.5 Mesa 21.2.6 (LLVM 12.0.0), Compiler: GCC 9.4.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0xf0 - Thermald 1.9.1
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu220.04)
Python Notes: Python 3.8.10
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Vulnerable + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Mitigation of Microcode + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 25 December 2022 04:17 by user phoronix.