AMD EPYC 7763 1P spec_rstack_overflow

Benchmarks by Michael Larabel for a future article looking at AMD Inception impact.

HTML result view exported from: https://openbenchmarking.org/result/2308112-NE-EPYC7763124&sro&grr.

AMD EPYC 7763 1P spec_rstack_overflowProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen Resolutionoffsafe RET no microcodesafe RETIBPBAMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads)AMD DAYTONA_X (RYM1009B BIOS)AMD Starship/Matisse256GB800GB INTEL SSDPF21Q800GBASPEEDVE2282 x Mellanox MT27710Ubuntu 22.046.5.0-rc5-phx-tues (x86_64)GNOME Shell 42.5X Server 1.21.1.31.3.224GCC 11.3.0 + LLVM 14.0.0ext41920x1080OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Disk Details- NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Details- off: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173- safe RET no microcode: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173- safe RET: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa0011d1- IBPB: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa0011d1Java Details- OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Details- Python 3.10.6Security Details- off: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Vulnerable no microcode + 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: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - safe RET no microcode: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET no microcode + 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: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - safe RET: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of safe RET + 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: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - IBPB: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of IBPB + 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: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

AMD EPYC 7763 1P spec_rstack_overflowopenvkl: vklBenchmark ISPCtensorflow: CPU - 64 - ResNet-50mysqlslap: 8192build-linux-kernel: allmodconfigcockroach: KV, 50% Reads - 128cockroach: KV, 95% Reads - 128clickhouse: 100M Rows Hits Dataset, Third Runclickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachemysqlslap: 4096openfoam: drivaerFastback, Medium Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timebuild-llvm: Ninjaospray: particle_volume/pathtracer/real_timebuild-nodejs: Time To Compileopenradioss: INIVOL and Fluid Structure Interaction Drop Containerapache-iotdb: 200 - 100 - 500apache-iotdb: 200 - 100 - 500numpy: openradioss: Bird Strike on Windshieldospray: particle_volume/scivis/real_timedeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streampgbench: 100 - 800 - Read Only - Average Latencypgbench: 100 - 800 - Read Onlypgbench: 100 - 800 - Read Write - Average Latencypgbench: 100 - 800 - Read Writemrbayes: Primate Phylogeny Analysiscassandra: Writesbuild-godot: Time To Compilespark: 1000000 - 100 - Broadcast Inner Join Test Timespark: 1000000 - 100 - Inner Join Test Timespark: 1000000 - 100 - Repartition Test Timespark: 1000000 - 100 - Group By Test Timespark: 1000000 - 100 - Calculate Pi Benchmarkspark: 1000000 - 100 - SHA-512 Benchmark Timeapache-iotdb: 500 - 100 - 500apache-iotdb: 500 - 100 - 500memtier-benchmark: Redis - 100 - 1:10openradioss: Bumper Beamospray: particle_volume/ao/real_timeopenradioss: Rubber O-Ring Seal Installationnginx: 1000nginx: 500deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamblender: Pabellon Barcelona - CPU-Onlyapache-iotdb: 200 - 100 - 200apache-iotdb: 200 - 100 - 200build-linux-kernel: defconfigopenvino: Person Detection FP16 - CPUopenvino: Person Detection FP16 - CPUmemtier-benchmark: Redis - 100 - 1:5memtier-benchmark: Redis - 50 - 1:5memtier-benchmark: Redis - 50 - 1:10openvino: Face Detection FP16-INT8 - CPUopenvino: Face Detection FP16-INT8 - CPUrocksdb: Read Rand Write Randospray: gravity_spheres_volume/dim_512/scivis/real_timeospray: gravity_spheres_volume/dim_512/ao/real_timespark: 1000000 - 100 - Calculate Pi Benchmark Using Dataframeopenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamrocksdb: Update Randdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamospray: gravity_spheres_volume/dim_512/pathtracer/real_timeapache-iotdb: 500 - 100 - 200apache-iotdb: 500 - 100 - 200deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamapache-iotdb: 200 - 1 - 200apache-iotdb: 200 - 1 - 200apache-iotdb: 500 - 1 - 500apache-iotdb: 500 - 1 - 500openradioss: Cell Phone Drop Testgromacs: MPI CPU - water_GMX50_baredeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamcompress-7zip: Decompression Ratingcompress-7zip: Compression Ratingamg: deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamremhos: Sample Remap Examplespecfem3d: Layered Halfspacespecfem3d: Water-layered Halfspaceapache-iotdb: 200 - 1 - 500apache-iotdb: 200 - 1 - 500apache-iotdb: 500 - 1 - 200apache-iotdb: 500 - 1 - 200namd: ATPase Simulation - 327,506 Atomsblender: BMW27 - CPU-Onlyspecfem3d: Homogeneous Halfspacespecfem3d: Tomographic Modeldacapobench: Jythonmt-dgemm: Sustained Floating-Point Ratedacapobench: Tradebeansspecfem3d: Mount St. Helensembree: Pathtracer ISPC - Crownembree: Pathtracer ISPC - Asian Dragonsqlite: 16sqlite: 8offsafe RET no microcodesafe RETIBPB45317.78355289.063103635.0135187.2362.64361.81349.43590633.51902140.61562176.374157.829164.268162.13117.9739463981.42457.23144.8317.7511678.933046.70200.256312871912.98861604136.686238741121.9481.301.882.094.9131.843.3978.9258682618.182195705.5187.7218.022677.46166499.89169583.1555.3922576.972284.5037.7043665846.2831.1924092.917.682197287.302204628.922177211.801141.4327.8329516848.331748.9605128.401126.0365.5911487.2450462287596.591553.596813.135536.5849501499.13839.720137.603713.83960525.6631.491415756.3333.105.68068.1483468.830638558538437410117990008.30783840.630717.37531.84563042429.77253538632.361271946.5714.051176385.350.3813027.3417.41712093314.134265606419324.200551399311.80123873257.422964.59646.2733.75545215.56301344.242100851.4131487.0329.19337.01323.42412644.36223145.06069182.169155.165172.749163.02123.6137720117.40418.95152.9117.7305679.576646.72390.296270728014.49955175137.518233069125.6631.392.222.385.1732.023.4279.1958073516.792154339.2693.6818.028885.04140555.98144020.0355.3747577.058984.6935.1046538766.0137.6234124.747.582167181.092218601.792173694.771142.2927.7928727658.327498.9694128.381126.6465.8910485.0069428112596.557953.597213.262138.5447445770.18840.772137.698714.05947741.3431.931408658.8336.375.73068.2595468.36463830393348129996454008.36123816.767617.78831.82926187030.42753170930.291342031.1113.611202637.360.3811527.5817.64368039714.404419238419124.695818409611.98216346057.295664.39168.8344.85045315.65301338.15799601.6132046.0337.45337.12318.12418643.71316144.02174181.528156.419173.064163.97120.0638833415.97422.58152.2717.7305682.207146.56770.289276844514.58954837138.851236241125.0601.412.142.265.1531.433.4782.2457099408.152157815.6993.9017.981784.48143271.26142619.8455.4028576.816684.4937.5344027904.8937.2434114.587.602145052.142145436.262172804.711142.5327.8228390858.338138.9404928.391126.1465.5662487.3677426947596.782253.572913.253835.8250578426.54840.423637.631914.73918691.4527.731583717.6236.405.70668.2325468.21703835153355959991021008.32193834.243917.95831.65994088529.59086826030.141345598.5913.361211172.130.3809827.4617.69029865014.188058962424123.702889414312.01038078157.313864.67428.8005.00645017.45276352.17895416.0119163.8349.50347.92336.69274645.40958148.25442204.080153.439195.493171.75120.5338572529.56389.92160.7017.6743681.396146.61210.461173382715.85450463144.829220814135.9621.642.412.315.7431.493.7180.7857529201.422148876.03113.6817.966399.04135431.46137051.6955.5026575.733585.6336.4744816394.7440.0854204.347.462126493.292092844.222137964.981144.9827.7121300068.265498.892392.6028.621116.8965.7767485.6842322231596.790253.578213.170936.5649316970.28840.888137.681814.78921701.4431.111441637.6340.115.70768.3456467.315038548737179910051386678.34113824.843118.65832.00976214230.05383480831.741287324.3511.931344749.200.3853427.7317.69591431914.225825227444624.251474530511.95144857757.296763.46687.9344.793OpenBenchmarking.org

OpenVKL

Benchmark: vklBenchmark ISPC

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 1.3.1Benchmark: vklBenchmark ISPCIBPBoffsafe RETsafe RET no microcode100200300400500SE +/- 0.67, N = 3SE +/- 0.58, N = 3SE +/- 0.00, N = 3SE +/- 0.58, N = 3450453453452MIN: 83 / MAX: 2495MIN: 84 / MAX: 2528MIN: 84 / MAX: 2520MIN: 85 / MAX: 2535

TensorFlow

Device: CPU - Batch Size: 64 - Model: ResNet-50

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: ResNet-50IBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 317.4517.7815.6515.56

MariaDB

Clients: 8192

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 8192IBPBoffsafe RETsafe RET no microcode80160240320400SE +/- 0.62, N = 3SE +/- 3.35, N = 3SE +/- 1.18, N = 3SE +/- 0.73, N = 32763553013011. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -lpthread -ldl

Timed Linux Kernel Compilation

Build: allmodconfig

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: allmodconfigIBPBoffsafe RETsafe RET no microcode80160240320400SE +/- 0.72, N = 3SE +/- 0.49, N = 3SE +/- 0.79, N = 3SE +/- 0.90, N = 3352.18289.06338.16344.24

CockroachDB

Workload: KV, 50% Reads - Concurrency: 128

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 50% Reads - Concurrency: 128IBPBoffsafe RETsafe RET no microcode20K40K60K80K100KSE +/- 341.24, N = 3SE +/- 275.86, N = 3SE +/- 948.29, N = 15SE +/- 719.41, N = 1595416.0103635.099601.6100851.4

CockroachDB

Workload: KV, 95% Reads - Concurrency: 128

OpenBenchmarking.orgops/s, More Is BetterCockroachDB 22.2Workload: KV, 95% Reads - Concurrency: 128IBPBoffsafe RETsafe RET no microcode30K60K90K120K150KSE +/- 408.63, N = 3SE +/- 931.05, N = 3SE +/- 1387.70, N = 15SE +/- 1043.12, N = 13119163.8135187.2132046.0131487.0

ClickHouse

100M Rows Hits Dataset, Third Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunIBPBoffsafe RETsafe RET no microcode80160240320400SE +/- 2.22, N = 3SE +/- 2.21, N = 3SE +/- 1.85, N = 3SE +/- 3.27, N = 5349.50362.64337.45329.19MIN: 31.56 / MAX: 5000MIN: 31.5 / MAX: 4285.71MIN: 31.46 / MAX: 4000MIN: 31.32 / MAX: 2857.14

ClickHouse

100M Rows Hits Dataset, Second Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunIBPBoffsafe RETsafe RET no microcode80160240320400SE +/- 2.20, N = 3SE +/- 1.42, N = 3SE +/- 4.86, N = 3SE +/- 2.16, N = 5347.92361.81337.12337.01MIN: 31.85 / MAX: 3750MIN: 31.46 / MAX: 4000MIN: 30.79 / MAX: 4000MIN: 30.49 / MAX: 3529.41

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheIBPBoffsafe RETsafe RET no microcode80160240320400SE +/- 2.84, N = 3SE +/- 0.68, N = 3SE +/- 2.94, N = 3SE +/- 3.38, N = 5336.69349.43318.12323.42MIN: 31.5 / MAX: 4000MIN: 31.06 / MAX: 4285.71MIN: 30.57 / MAX: 3333.33MIN: 30.82 / MAX: 5000

MariaDB

Clients: 4096

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 4096IBPBoffsafe RETsafe RET no microcode130260390520650SE +/- 0.71, N = 3SE +/- 5.48, N = 3SE +/- 3.51, N = 3SE +/- 2.96, N = 32745904184121. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -lpthread -ldl

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution TimeIBPBoffsafe RETsafe RET no microcode140280420560700645.41633.52643.71644.361. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Mesh Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeIBPBoffsafe RETsafe RET no microcode306090120150148.25140.62144.02145.061. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

Timed LLVM Compilation

Build System: Ninja

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaIBPBoffsafe RETsafe RET no microcode4080120160200SE +/- 0.20, N = 3SE +/- 0.11, N = 3SE +/- 0.15, N = 3SE +/- 0.11, N = 3204.08176.37181.53182.17

OSPRay

Benchmark: particle_volume/pathtracer/real_time

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/pathtracer/real_timeIBPBoffsafe RETsafe RET no microcode306090120150SE +/- 0.43, N = 3SE +/- 0.21, N = 3SE +/- 0.07, N = 3SE +/- 1.83, N = 3153.44157.83156.42155.17

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To CompileIBPBoffsafe RETsafe RET no microcode4080120160200SE +/- 0.16, N = 3SE +/- 0.14, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 3195.49164.27173.06172.75

OpenRadioss

Model: INIVOL and Fluid Structure Interaction Drop Container

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: INIVOL and Fluid Structure Interaction Drop ContainerIBPBoffsafe RETsafe RET no microcode4080120160200SE +/- 0.17, N = 3SE +/- 0.16, N = 3SE +/- 0.50, N = 3SE +/- 0.39, N = 3171.75162.13163.97163.02

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode306090120150SE +/- 1.16, N = 8SE +/- 0.95, N = 10SE +/- 0.86, N = 15SE +/- 1.63, N = 5120.53117.97120.06123.61MAX: 4401.37MAX: 4652.25MAX: 4495.21MAX: 4533.33

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode8M16M24M32M40MSE +/- 327739.29, N = 8SE +/- 302926.36, N = 10SE +/- 288707.73, N = 15SE +/- 394126.89, N = 538572529.5639463981.4238833415.9737720117.40

Numpy Benchmark

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkIBPBoffsafe RETsafe RET no microcode100200300400500SE +/- 1.11, N = 3SE +/- 1.76, N = 3SE +/- 0.84, N = 3SE +/- 2.01, N = 3389.92457.23422.58418.95

OpenRadioss

Model: Bird Strike on Windshield

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bird Strike on WindshieldIBPBoffsafe RETsafe RET no microcode4080120160200SE +/- 0.89, N = 3SE +/- 0.07, N = 3SE +/- 0.73, N = 3SE +/- 0.67, N = 3160.70144.83152.27152.91

OSPRay

Benchmark: particle_volume/scivis/real_time

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/scivis/real_timeIBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 317.6717.7517.7317.73

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode150300450600750SE +/- 1.14, N = 3SE +/- 1.21, N = 3SE +/- 0.96, N = 3SE +/- 1.47, N = 3681.40678.93682.21679.58

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode1122334455SE +/- 0.02, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 346.6146.7046.5746.72

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average LatencyIBPBoffsafe RETsafe RET no microcode0.10370.20740.31110.41480.5185SE +/- 0.001, N = 3SE +/- 0.000, N = 3SE +/- 0.004, N = 3SE +/- 0.003, N = 30.4610.2560.2890.2961. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read OnlyIBPBoffsafe RETsafe RET no microcode700K1400K2100K2800K3500KSE +/- 2988.66, N = 3SE +/- 1705.16, N = 3SE +/- 34286.68, N = 3SE +/- 29158.10, N = 317338273128719276844527072801. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average LatencyIBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.02, N = 315.8512.9914.5914.501. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

PostgreSQL

Scaling Factor: 100 - Clients: 800 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 800 - Mode: Read WriteIBPBoffsafe RETsafe RET no microcode13K26K39K52K65KSE +/- 133.40, N = 3SE +/- 418.78, N = 3SE +/- 207.71, N = 3SE +/- 66.28, N = 3504636160454837551751. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Timed MrBayes Analysis

Primate Phylogeny Analysis

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny AnalysisIBPBoffsafe RETsafe RET no microcode306090120150SE +/- 1.03, N = 3SE +/- 0.85, N = 3SE +/- 1.05, N = 3SE +/- 0.66, N = 3144.83136.69138.85137.521. (CC) gcc options: -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm -O3 -std=c99 -pedantic -lm

Apache Cassandra

Test: Writes

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 4.1.3Test: WritesIBPBoffsafe RETsafe RET no microcode50K100K150K200K250KSE +/- 242.75, N = 3SE +/- 413.74, N = 3SE +/- 479.91, N = 3SE +/- 950.59, N = 3220814238741236241233069

Timed Godot Game Engine Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileIBPBoffsafe RETsafe RET no microcode306090120150SE +/- 0.08, N = 3SE +/- 0.19, N = 3SE +/- 0.24, N = 3SE +/- 0.33, N = 3135.96121.95125.06125.66

Apache Spark

Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Time

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test TimeIBPBoffsafe RETsafe RET no microcode0.3690.7381.1071.4761.845SE +/- 0.03, N = 3SE +/- 0.01, N = 15SE +/- 0.01, N = 3SE +/- 0.02, N = 31.641.301.411.39

Apache Spark

Row Count: 1000000 - Partitions: 100 - Inner Join Test Time

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Inner Join Test TimeIBPBoffsafe RETsafe RET no microcode0.54231.08461.62692.16922.7115SE +/- 0.04, N = 3SE +/- 0.02, N = 15SE +/- 0.06, N = 3SE +/- 0.05, N = 32.411.882.142.22

Apache Spark

Row Count: 1000000 - Partitions: 100 - Repartition Test Time

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Repartition Test TimeIBPBoffsafe RETsafe RET no microcode0.53551.0711.60652.1422.6775SE +/- 0.12, N = 3SE +/- 0.04, N = 15SE +/- 0.04, N = 3SE +/- 0.04, N = 32.312.092.262.38

Apache Spark

Row Count: 1000000 - Partitions: 100 - Group By Test Time

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Group By Test TimeIBPBoffsafe RETsafe RET no microcode1.29152.5833.87455.1666.4575SE +/- 0.06, N = 3SE +/- 0.04, N = 15SE +/- 0.07, N = 3SE +/- 0.08, N = 35.744.915.155.17

Apache Spark

Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Calculate Pi BenchmarkIBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.20, N = 3SE +/- 0.12, N = 15SE +/- 0.33, N = 3SE +/- 0.01, N = 331.4931.8431.4332.02

Apache Spark

Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark TimeIBPBoffsafe RETsafe RET no microcode0.83481.66962.50443.33924.174SE +/- 0.05, N = 3SE +/- 0.03, N = 15SE +/- 0.04, N = 3SE +/- 0.04, N = 33.713.393.473.42

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode20406080100SE +/- 0.42, N = 3SE +/- 2.14, N = 3SE +/- 1.29, N = 3SE +/- 0.81, N = 480.7878.9282.2479.19MAX: 2592.69MAX: 1729.94MAX: 3625.32MAX: 5165.86

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode13M26M39M52M65MSE +/- 269354.94, N = 3SE +/- 817020.04, N = 3SE +/- 721225.08, N = 3SE +/- 648692.91, N = 457529201.4258682618.1857099408.1558073516.79

Redis 7.0.12 + memtier_benchmark

Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10IBPBoffsafe RETsafe RET no microcode500K1000K1500K2000K2500KSE +/- 12623.44, N = 3SE +/- 30210.22, N = 3SE +/- 792.70, N = 3SE +/- 16754.20, N = 102148876.032195705.512157815.692154339.261. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenRadioss

Model: Bumper Beam

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Bumper BeamIBPBoffsafe RETsafe RET no microcode306090120150SE +/- 0.23, N = 3SE +/- 0.33, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3113.6887.7293.9093.68

OSPRay

Benchmark: particle_volume/ao/real_time

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/ao/real_timeIBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 317.9718.0217.9818.03

OpenRadioss

Model: Rubber O-Ring Seal Installation

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Rubber O-Ring Seal InstallationIBPBoffsafe RETsafe RET no microcode20406080100SE +/- 0.34, N = 3SE +/- 0.23, N = 3SE +/- 0.24, N = 3SE +/- 0.17, N = 399.0477.4684.4885.04

nginx

Connections: 1000

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000IBPBoffsafe RETsafe RET no microcode40K80K120K160K200KSE +/- 242.54, N = 3SE +/- 362.13, N = 3SE +/- 314.03, N = 3SE +/- 352.89, N = 3135431.46166499.89143271.26140555.981. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

nginx

Connections: 500

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500IBPBoffsafe RETsafe RET no microcode40K80K120K160K200KSE +/- 262.73, N = 3SE +/- 284.72, N = 3SE +/- 251.96, N = 3SE +/- 284.55, N = 3137051.69169583.15142619.84144020.031. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode1224364860SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 355.5055.3955.4055.37

Neural Magic DeepSparse

Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode120240360480600SE +/- 0.94, N = 3SE +/- 0.44, N = 3SE +/- 0.39, N = 3SE +/- 0.39, N = 3575.73576.97576.82577.06

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Pabellon Barcelona - Compute: CPU-OnlyIBPBoffsafe RETsafe RET no microcode20406080100SE +/- 0.05, N = 3SE +/- 0.13, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 385.6384.5084.4984.69

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode918273645SE +/- 0.32, N = 3SE +/- 0.55, N = 15SE +/- 0.52, N = 15SE +/- 0.62, N = 336.4737.7037.5335.10MAX: 808.57MAX: 802.64MAX: 755.16MAX: 728.37

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode10M20M30M40M50MSE +/- 146499.20, N = 3SE +/- 574678.74, N = 15SE +/- 543529.82, N = 15SE +/- 614274.26, N = 344816394.7443665846.2844027904.8946538766.01

Timed Linux Kernel Compilation

Build: defconfig

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigIBPBoffsafe RETsafe RET no microcode918273645SE +/- 0.37, N = 7SE +/- 0.34, N = 5SE +/- 0.37, N = 6SE +/- 0.35, N = 640.0931.1937.2437.62

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUIBPBoffsafe RETsafe RET no microcode9001800270036004500SE +/- 10.56, N = 3SE +/- 14.65, N = 3SE +/- 10.77, N = 3SE +/- 6.87, N = 34204.344092.914114.584124.74MIN: 2302.89 / MAX: 4817.72MIN: 3409.52 / MAX: 4641.43MIN: 2087 / MAX: 5053.62MIN: 2129.26 / MAX: 5016.361. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenVINO

Model: Person Detection FP16 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Person Detection FP16 - Device: CPUIBPBoffsafe RETsafe RET no microcode246810SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 37.467.687.607.581. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Redis 7.0.12 + memtier_benchmark

Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:5

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:5IBPBoffsafe RETsafe RET no microcode500K1000K1500K2000K2500KSE +/- 942.73, N = 3SE +/- 14704.83, N = 3SE +/- 4916.89, N = 3SE +/- 17712.54, N = 32126493.292197287.302145052.142167181.091. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Redis 7.0.12 + memtier_benchmark

Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:5

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:5IBPBoffsafe RETsafe RET no microcode500K1000K1500K2000K2500KSE +/- 21878.46, N = 3SE +/- 11955.97, N = 3SE +/- 1778.76, N = 3SE +/- 31351.12, N = 32092844.222204628.922145436.262218601.791. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Redis 7.0.12 + memtier_benchmark

Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10

OpenBenchmarking.orgOps/sec, More Is BetterRedis 7.0.12 + memtier_benchmark 2.0Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10IBPBoffsafe RETsafe RET no microcode500K1000K1500K2000K2500KSE +/- 13504.65, N = 3SE +/- 14630.02, N = 3SE +/- 2448.62, N = 3SE +/- 17754.58, N = 32137964.982177211.802172804.712173694.771. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUIBPBoffsafe RETsafe RET no microcode2004006008001000SE +/- 0.42, N = 3SE +/- 0.32, N = 3SE +/- 1.09, N = 3SE +/- 0.27, N = 31144.981141.431142.531142.29MIN: 502.04 / MAX: 1175.93MIN: 998.76 / MAX: 1165.45MIN: 999.01 / MAX: 1177.02MIN: 985.75 / MAX: 1168.761. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Face Detection FP16-INT8 - Device: CPUIBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 327.7127.8327.8227.791. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

RocksDB

Test: Read Random Write Random

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Read Random Write RandomIBPBoffsafe RETsafe RET no microcode600K1200K1800K2400K3000KSE +/- 7875.65, N = 3SE +/- 35283.44, N = 4SE +/- 21895.20, N = 3SE +/- 18652.89, N = 321300062951684283908528727651. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OSPRay

Benchmark: gravity_spheres_volume/dim_512/scivis/real_time

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeIBPBoffsafe RETsafe RET no microcode246810SE +/- 0.02088, N = 3SE +/- 0.00864, N = 3SE +/- 0.01059, N = 3SE +/- 0.01659, N = 38.265498.331748.338138.32749

OSPRay

Benchmark: gravity_spheres_volume/dim_512/ao/real_time

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timeIBPBoffsafe RETsafe RET no microcode3691215SE +/- 0.01872, N = 3SE +/- 0.02460, N = 3SE +/- 0.01456, N = 3SE +/- 0.02941, N = 38.892398.960518.940498.96941

Apache Spark

Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using DataframeIBPB0.5851.171.7552.342.925SE +/- 0.08, N = 32.60

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUIBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 328.6228.4028.3928.38MIN: 14.91 / MAX: 49.84MIN: 14.74 / MAX: 48.66MIN: 14.64 / MAX: 50.33MIN: 14.89 / MAX: 51.631. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.3Model: Weld Porosity Detection FP16 - Device: CPUIBPBoffsafe RETsafe RET no microcode2004006008001000SE +/- 0.40, N = 3SE +/- 0.17, N = 3SE +/- 0.13, N = 3SE +/- 0.72, N = 31116.891126.031126.141126.641. (CXX) g++ options: -isystem -fsigned-char -ffunction-sections -fdata-sections -msse4.1 -msse4.2 -O3 -fno-strict-overflow -fwrapv -fPIC -fvisibility=hidden -Os -std=c++11 -MD -MT -MF

Neural Magic DeepSparse

Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode1530456075SE +/- 0.20, N = 3SE +/- 0.15, N = 3SE +/- 0.16, N = 3SE +/- 0.14, N = 365.7865.5965.5765.89

Neural Magic DeepSparse

Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode110220330440550SE +/- 1.48, N = 3SE +/- 1.15, N = 3SE +/- 1.01, N = 3SE +/- 0.96, N = 3485.68487.25487.37485.01

RocksDB

Test: Update Random

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Update RandomIBPBoffsafe RETsafe RET no microcode100K200K300K400K500KSE +/- 110.81, N = 3SE +/- 893.82, N = 3SE +/- 185.49, N = 3SE +/- 426.73, N = 33222314622874269474281121. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Neural Magic DeepSparse

Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode130260390520650SE +/- 0.06, N = 3SE +/- 0.18, N = 3SE +/- 0.11, N = 3SE +/- 0.26, N = 3596.79596.59596.78596.56

Neural Magic DeepSparse

Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode1224364860SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 353.5853.6053.5753.60

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeIBPBoffsafe RETsafe RET no microcode3691215SE +/- 0.00, N = 3SE +/- 0.13, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 313.1713.1413.2513.26

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode918273645SE +/- 0.44, N = 3SE +/- 0.61, N = 3SE +/- 0.49, N = 3SE +/- 0.11, N = 336.5636.5835.8238.54MAX: 2253.21MAX: 2252.73MAX: 3267.55MAX: 3276.77

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode11M22M33M44M55MSE +/- 616490.96, N = 3SE +/- 681823.31, N = 3SE +/- 634314.77, N = 3SE +/- 147114.88, N = 349316970.2849501499.1350578426.5447445770.18

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode2004006008001000SE +/- 0.53, N = 3SE +/- 0.50, N = 3SE +/- 0.34, N = 3SE +/- 0.36, N = 3840.89839.72840.42840.77

Neural Magic DeepSparse

Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode918273645SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.08, N = 3SE +/- 0.07, N = 337.6837.6037.6337.70

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.19, N = 9SE +/- 0.19, N = 3SE +/- 0.21, N = 8SE +/- 0.16, N = 1214.7813.8314.7314.05MAX: 618.06MAX: 596.78MAX: 645.11MAX: 609.96

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode200K400K600K800K1000KSE +/- 7583.18, N = 9SE +/- 8467.91, N = 3SE +/- 7998.38, N = 8SE +/- 6730.71, N = 12921701.44960525.66918691.45947741.34

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.35, N = 3SE +/- 0.29, N = 3SE +/- 0.22, N = 3SE +/- 0.49, N = 331.1131.4927.7331.93MAX: 908.02MAX: 939.96MAX: 938.92MAX: 930.97

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode300K600K900K1200K1500KSE +/- 6687.04, N = 3SE +/- 4294.81, N = 3SE +/- 5073.96, N = 3SE +/- 13029.07, N = 31441637.631415756.331583717.621408658.83

OpenRadioss

Model: Cell Phone Drop Test

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenRadioss 2022.10.13Model: Cell Phone Drop TestIBPBoffsafe RETsafe RET no microcode918273645SE +/- 0.14, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.26, N = 340.1133.1036.4036.37

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2023Implementation: MPI CPU - Input: water_GMX50_bareIBPBoffsafe RETsafe RET no microcode1.28932.57863.86795.15726.4465SE +/- 0.011, N = 3SE +/- 0.012, N = 3SE +/- 0.010, N = 3SE +/- 0.006, N = 35.7075.6805.7065.7301. (CXX) g++ options: -O3

Neural Magic DeepSparse

Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode1530456075SE +/- 0.07, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 368.3568.1568.2368.26

Neural Magic DeepSparse

Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode100200300400500SE +/- 0.43, N = 3SE +/- 0.24, N = 3SE +/- 0.31, N = 3SE +/- 0.42, N = 3467.32468.83468.22468.36

7-Zip Compression

Test: Decompression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression RatingIBPBoffsafe RETsafe RET no microcode80K160K240K320K400KSE +/- 312.85, N = 3SE +/- 845.58, N = 3SE +/- 605.48, N = 3SE +/- 380.69, N = 33854873855853835153830391. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

7-Zip Compression

Test: Compression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingIBPBoffsafe RETsafe RET no microcode80K160K240K320K400KSE +/- 248.93, N = 3SE +/- 1018.75, N = 3SE +/- 435.27, N = 3SE +/- 25.38, N = 33717993843743355953348121. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Algebraic Multi-Grid Benchmark

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2IBPBoffsafe RETsafe RET no microcode200M400M600M800M1000MSE +/- 1724277.85, N = 3SE +/- 839009.73, N = 3SE +/- 367255.40, N = 3SE +/- 575791.94, N = 3100513866710117990009991021009996454001. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

Neural Magic DeepSparse

Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode246810SE +/- 0.0189, N = 3SE +/- 0.0095, N = 3SE +/- 0.0271, N = 3SE +/- 0.0167, N = 38.34118.30788.32198.3612

Neural Magic DeepSparse

Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-StreamIBPBoffsafe RETsafe RET no microcode8001600240032004000SE +/- 7.83, N = 3SE +/- 4.76, N = 3SE +/- 12.28, N = 3SE +/- 7.72, N = 33824.843840.633834.243816.77

Remhos

Test: Sample Remap Example

OpenBenchmarking.orgSeconds, Fewer Is BetterRemhos 1.0Test: Sample Remap ExampleIBPBoffsafe RETsafe RET no microcode510152025SE +/- 0.12, N = 14SE +/- 0.23, N = 3SE +/- 0.19, N = 3SE +/- 0.17, N = 318.6617.3817.9617.791. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

SPECFEM3D

Model: Layered Halfspace

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Layered HalfspaceIBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.21, N = 3SE +/- 0.35, N = 3SE +/- 0.18, N = 3SE +/- 0.23, N = 332.0131.8531.6631.831. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

SPECFEM3D

Model: Water-layered Halfspace

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Water-layered HalfspaceIBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.19, N = 3SE +/- 0.15, N = 3SE +/- 0.35, N = 3SE +/- 0.25, N = 330.0529.7729.5930.431. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode816243240SE +/- 0.42, N = 4SE +/- 0.24, N = 3SE +/- 0.22, N = 3SE +/- 0.02, N = 331.7432.3630.1430.29MAX: 667.18MAX: 646.51MAX: 641.04MAX: 715.01

Apache IoTDB

Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500IBPBoffsafe RETsafe RET no microcode300K600K900K1200K1500KSE +/- 14032.06, N = 4SE +/- 7578.67, N = 3SE +/- 9180.92, N = 3SE +/- 1525.49, N = 31287324.351271946.571345598.591342031.11

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.13, N = 3SE +/- 0.07, N = 3SE +/- 0.19, N = 3SE +/- 0.14, N = 311.9314.0513.3613.61MAX: 855.56MAX: 858.17MAX: 881.3MAX: 854.4

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.1.2Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200IBPBoffsafe RETsafe RET no microcode300K600K900K1200K1500KSE +/- 3166.01, N = 3SE +/- 1566.77, N = 3SE +/- 4253.29, N = 3SE +/- 6553.14, N = 31344749.201176385.351211172.131202637.36

NAMD

ATPase Simulation - 327,506 Atoms

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsIBPBoffsafe RETsafe RET no microcode0.08670.17340.26010.34680.4335SE +/- 0.00026, N = 3SE +/- 0.00017, N = 3SE +/- 0.00028, N = 3SE +/- 0.00029, N = 30.385340.381300.380980.38115

Blender

Blend File: BMW27 - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: BMW27 - Compute: CPU-OnlyIBPBoffsafe RETsafe RET no microcode714212835SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.06, N = 327.7327.3427.4627.58

SPECFEM3D

Model: Homogeneous Halfspace

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Homogeneous HalfspaceIBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.21, N = 3SE +/- 0.20, N = 417.7017.4217.6917.641. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

SPECFEM3D

Model: Tomographic Model

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Tomographic ModelIBPBoffsafe RETsafe RET no microcode48121620SE +/- 0.08, N = 3SE +/- 0.09, N = 3SE +/- 0.20, N = 3SE +/- 0.15, N = 314.2314.1314.1914.401. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

DaCapo Benchmark

Java Test: Jython

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonIBPBoffsafe RETsafe RET no microcode10002000300040005000SE +/- 38.12, N = 20SE +/- 18.07, N = 4SE +/- 49.88, N = 4SE +/- 47.28, N = 44446419342414191

ACES DGEMM

Sustained Floating-Point Rate

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateIBPBoffsafe RETsafe RET no microcode612182430SE +/- 0.09, N = 3SE +/- 0.21, N = 8SE +/- 0.26, N = 5SE +/- 0.34, N = 324.2524.2023.7024.701. (CC) gcc options: -O3 -march=native -fopenmp

DaCapo Benchmark

Java Test: Tradebeans

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansIBPBoffsafe RETsafe RET no microcode11002200330044005500SE +/- 56.17, N = 4SE +/- 42.66, N = 4SE +/- 28.11, N = 4SE +/- 44.47, N = 45305399341434096

SPECFEM3D

Model: Mount St. Helens

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Mount St. HelensIBPBoffsafe RETsafe RET no microcode3691215SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.07, N = 311.9511.8012.0111.981. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

Embree

Binary: Pathtracer ISPC - Model: Crown

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: CrownIBPBoffsafe RETsafe RET no microcode1326395265SE +/- 0.10, N = 3SE +/- 0.08, N = 3SE +/- 0.15, N = 3SE +/- 0.14, N = 357.3057.4257.3157.30MIN: 56.3 / MAX: 58.61MIN: 56.59 / MAX: 58.54MIN: 56.2 / MAX: 58.59MIN: 56.26 / MAX: 58.69

Embree

Binary: Pathtracer ISPC - Model: Asian Dragon

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian DragonIBPBoffsafe RETsafe RET no microcode1428425670SE +/- 0.14, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 363.4764.6064.6764.39MIN: 62.67 / MAX: 65.74MIN: 64.05 / MAX: 66.13MIN: 64.11 / MAX: 66.01MIN: 63.77 / MAX: 66.16

SQLite

Threads / Copies: 16

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 16IBPBoffsafe RETsafe RET no microcode246810SE +/- 0.007, N = 3SE +/- 0.020, N = 3SE +/- 0.052, N = 3SE +/- 0.024, N = 37.9346.2738.8008.8341. (CC) gcc options: -O2 -lz -lm

SQLite

Threads / Copies: 8

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.41.2Threads / Copies: 8IBPBoffsafe RETsafe RET no microcode1.12642.25283.37924.50565.632SE +/- 0.010, N = 3SE +/- 0.013, N = 3SE +/- 0.036, N = 3SE +/- 0.016, N = 34.7933.7555.0064.8501. (CC) gcc options: -O2 -lz -lm

Phoronix Test Suite v10.8.5