Benchmarks by Michael Larabel for a future article looking at AMD Inception impact.
off Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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 Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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 Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa0011d1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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 Processor: AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 256GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710
OS: Ubuntu 22.04, Kernel: 6.5.0-rc5-phx-tues (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa0011d1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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_overflow OpenBenchmarking.org Phoronix Test Suite AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads) AMD DAYTONA_X (RYM1009B BIOS) AMD Starship/Matisse 256GB 800GB INTEL SSDPF21Q800GB ASPEED VE228 2 x Mellanox MT27710 Ubuntu 22.04 6.5.0-rc5-phx-tues (x86_64) GNOME Shell 42.5 X Server 1.21.1.3 1.3.224 GCC 11.3.0 + LLVM 14.0.0 ext4 1920x1080 Processor Motherboard Chipset Memory Disk Graphics Monitor Network OS Kernel Desktop Display Server Vulkan Compiler File-System Screen Resolution AMD EPYC 7763 1P Spec_rstack_overflow Benchmarks System Logs - Transparent Huge Pages: madvise - --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 - NONE / errors=remount-ro,relatime,rw / Block Size: 4096 - 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: 0xa0011d1 - OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04) - Python 3.10.6 - 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
off safe RET no microcode safe RET IBPB Result Overview Phoronix Test Suite 100% 117% 133% 150% 166% MariaDB PostgreSQL RocksDB SQLite Timed Linux Kernel Compilation nginx Timed Node.js Compilation OpenRadioss Numpy Benchmark Timed LLVM Compilation Apache Spark DaCapo Benchmark TensorFlow Timed Godot Game Engine Compilation CockroachDB ClickHouse Apache Cassandra 7-Zip Compression Remhos Timed MrBayes Analysis ACES DGEMM OpenFOAM Redis 7.0.12 + memtier_benchmark Apache IoTDB Blender SPECFEM3D OpenVINO Algebraic Multi-Grid Benchmark NAMD Embree GROMACS OSPRay OpenVKL Neural Magic DeepSparse
AMD EPYC 7763 1P spec_rstack_overflow openvkl: vklBenchmark ISPC tensorflow: CPU - 64 - ResNet-50 mysqlslap: 8192 build-linux-kernel: allmodconfig cockroach: KV, 50% Reads - 128 cockroach: KV, 95% Reads - 128 clickhouse: 100M Rows Hits Dataset, Third Run clickhouse: 100M Rows Hits Dataset, Second Run clickhouse: 100M Rows Hits Dataset, First Run / Cold Cache mysqlslap: 4096 openfoam: drivaerFastback, Medium Mesh Size - Execution Time openfoam: drivaerFastback, Medium Mesh Size - Mesh Time build-llvm: Ninja ospray: particle_volume/pathtracer/real_time build-nodejs: Time To Compile openradioss: INIVOL and Fluid Structure Interaction Drop Container apache-iotdb: 200 - 100 - 500 apache-iotdb: 200 - 100 - 500 numpy: openradioss: Bird Strike on Windshield ospray: particle_volume/scivis/real_time deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream pgbench: 100 - 800 - Read Only - Average Latency pgbench: 100 - 800 - Read Only pgbench: 100 - 800 - Read Write - Average Latency pgbench: 100 - 800 - Read Write mrbayes: Primate Phylogeny Analysis cassandra: Writes build-godot: Time To Compile spark: 1000000 - 100 - Broadcast Inner Join Test Time spark: 1000000 - 100 - Inner Join Test Time spark: 1000000 - 100 - Repartition Test Time spark: 1000000 - 100 - Group By Test Time spark: 1000000 - 100 - Calculate Pi Benchmark spark: 1000000 - 100 - SHA-512 Benchmark Time apache-iotdb: 500 - 100 - 500 apache-iotdb: 500 - 100 - 500 memtier-benchmark: Redis - 100 - 1:10 openradioss: Bumper Beam ospray: particle_volume/ao/real_time openradioss: Rubber O-Ring Seal Installation nginx: 1000 nginx: 500 deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream blender: Pabellon Barcelona - CPU-Only apache-iotdb: 200 - 100 - 200 apache-iotdb: 200 - 100 - 200 build-linux-kernel: defconfig openvino: Person Detection FP16 - CPU openvino: Person Detection FP16 - CPU memtier-benchmark: Redis - 100 - 1:5 memtier-benchmark: Redis - 50 - 1:5 memtier-benchmark: Redis - 50 - 1:10 openvino: Face Detection FP16-INT8 - CPU openvino: Face Detection FP16-INT8 - CPU rocksdb: Read Rand Write Rand ospray: gravity_spheres_volume/dim_512/scivis/real_time ospray: gravity_spheres_volume/dim_512/ao/real_time spark: 1000000 - 100 - Calculate Pi Benchmark Using Dataframe openvino: Weld Porosity Detection FP16 - CPU openvino: Weld Porosity Detection FP16 - CPU deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream deepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream rocksdb: Update Rand deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream ospray: gravity_spheres_volume/dim_512/pathtracer/real_time apache-iotdb: 500 - 100 - 200 apache-iotdb: 500 - 100 - 200 deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream apache-iotdb: 200 - 1 - 200 apache-iotdb: 200 - 1 - 200 apache-iotdb: 500 - 1 - 500 apache-iotdb: 500 - 1 - 500 openradioss: Cell Phone Drop Test gromacs: MPI CPU - water_GMX50_bare deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream compress-7zip: Decompression Rating compress-7zip: Compression Rating amg: deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream remhos: Sample Remap Example specfem3d: Layered Halfspace specfem3d: Water-layered Halfspace apache-iotdb: 200 - 1 - 500 apache-iotdb: 200 - 1 - 500 apache-iotdb: 500 - 1 - 200 apache-iotdb: 500 - 1 - 200 namd: ATPase Simulation - 327,506 Atoms blender: BMW27 - CPU-Only specfem3d: Homogeneous Halfspace specfem3d: Tomographic Model dacapobench: Jython mt-dgemm: Sustained Floating-Point Rate dacapobench: Tradebeans specfem3d: Mount St. Helens embree: Pathtracer ISPC - Crown embree: Pathtracer ISPC - Asian Dragon sqlite: 16 sqlite: 8 off safe RET no microcode safe RET IBPB 453 17.78 355 289.063 103635.0 135187.2 362.64 361.81 349.43 590 633.51902 140.61562 176.374 157.829 164.268 162.13 117.97 39463981.42 457.23 144.83 17.7511 678.9330 46.7020 0.256 3128719 12.988 61604 136.686 238741 121.948 1.30 1.88 2.09 4.91 31.84 3.39 78.92 58682618.18 2195705.51 87.72 18.0226 77.46 166499.89 169583.15 55.3922 576.9722 84.50 37.70 43665846.28 31.192 4092.91 7.68 2197287.30 2204628.92 2177211.80 1141.43 27.83 2951684 8.33174 8.96051 28.40 1126.03 65.5911 487.2450 462287 596.5915 53.5968 13.1355 36.58 49501499.13 839.7201 37.6037 13.83 960525.66 31.49 1415756.33 33.10 5.680 68.1483 468.8306 385585 384374 1011799000 8.3078 3840.6307 17.375 31.845630424 29.772535386 32.36 1271946.57 14.05 1176385.35 0.38130 27.34 17.417120933 14.134265606 4193 24.200551 3993 11.801238732 57.4229 64.5964 6.273 3.755 452 15.56 301 344.242 100851.4 131487.0 329.19 337.01 323.42 412 644.36223 145.06069 182.169 155.165 172.749 163.02 123.61 37720117.40 418.95 152.91 17.7305 679.5766 46.7239 0.296 2707280 14.499 55175 137.518 233069 125.663 1.39 2.22 2.38 5.17 32.02 3.42 79.19 58073516.79 2154339.26 93.68 18.0288 85.04 140555.98 144020.03 55.3747 577.0589 84.69 35.10 46538766.01 37.623 4124.74 7.58 2167181.09 2218601.79 2173694.77 1142.29 27.79 2872765 8.32749 8.96941 28.38 1126.64 65.8910 485.0069 428112 596.5579 53.5972 13.2621 38.54 47445770.18 840.7721 37.6987 14.05 947741.34 31.93 1408658.83 36.37 5.730 68.2595 468.3646 383039 334812 999645400 8.3612 3816.7676 17.788 31.829261870 30.427531709 30.29 1342031.11 13.61 1202637.36 0.38115 27.58 17.643680397 14.404419238 4191 24.695818 4096 11.982163460 57.2956 64.3916 8.834 4.850 453 15.65 301 338.157 99601.6 132046.0 337.45 337.12 318.12 418 643.71316 144.02174 181.528 156.419 173.064 163.97 120.06 38833415.97 422.58 152.27 17.7305 682.2071 46.5677 0.289 2768445 14.589 54837 138.851 236241 125.060 1.41 2.14 2.26 5.15 31.43 3.47 82.24 57099408.15 2157815.69 93.90 17.9817 84.48 143271.26 142619.84 55.4028 576.8166 84.49 37.53 44027904.89 37.243 4114.58 7.60 2145052.14 2145436.26 2172804.71 1142.53 27.82 2839085 8.33813 8.94049 28.39 1126.14 65.5662 487.3677 426947 596.7822 53.5729 13.2538 35.82 50578426.54 840.4236 37.6319 14.73 918691.45 27.73 1583717.62 36.40 5.706 68.2325 468.2170 383515 335595 999102100 8.3219 3834.2439 17.958 31.659940885 29.590868260 30.14 1345598.59 13.36 1211172.13 0.38098 27.46 17.690298650 14.188058962 4241 23.702889 4143 12.010380781 57.3138 64.6742 8.800 5.006 450 17.45 276 352.178 95416.0 119163.8 349.50 347.92 336.69 274 645.40958 148.25442 204.080 153.439 195.493 171.75 120.53 38572529.56 389.92 160.70 17.6743 681.3961 46.6121 0.461 1733827 15.854 50463 144.829 220814 135.962 1.64 2.41 2.31 5.74 31.49 3.71 80.78 57529201.42 2148876.03 113.68 17.9663 99.04 135431.46 137051.69 55.5026 575.7335 85.63 36.47 44816394.74 40.085 4204.34 7.46 2126493.29 2092844.22 2137964.98 1144.98 27.71 2130006 8.26549 8.89239 2.60 28.62 1116.89 65.7767 485.6842 322231 596.7902 53.5782 13.1709 36.56 49316970.28 840.8881 37.6818 14.78 921701.44 31.11 1441637.63 40.11 5.707 68.3456 467.3150 385487 371799 1005138667 8.3411 3824.8431 18.658 32.009762142 30.053834808 31.74 1287324.35 11.93 1344749.20 0.38534 27.73 17.695914319 14.225825227 4446 24.251474 5305 11.951448577 57.2967 63.4668 7.934 4.793 OpenBenchmarking.org
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.org Items / Sec, More Is Better OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC IBPB off safe RET safe RET no microcode 100 200 300 400 500 SE +/- 0.67, N = 3 SE +/- 0.58, N = 3 SE +/- 0.00, N = 3 SE +/- 0.58, N = 3 450 453 453 452 MIN: 83 / MAX: 2495 MIN: 84 / MAX: 2528 MIN: 84 / MAX: 2520 MIN: 85 / MAX: 2535
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 17.45 17.78 15.65 15.56
MariaDB This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 8192 IBPB off safe RET safe RET no microcode 80 160 240 320 400 SE +/- 0.62, N = 3 SE +/- 3.35, N = 3 SE +/- 1.18, N = 3 SE +/- 0.73, N = 3 276 355 301 301 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -lpthread -ldl
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: allmodconfig IBPB off safe RET safe RET no microcode 80 160 240 320 400 SE +/- 0.72, N = 3 SE +/- 0.49, N = 3 SE +/- 0.79, N = 3 SE +/- 0.90, N = 3 352.18 289.06 338.16 344.24
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.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 128 IBPB off safe RET safe RET no microcode 20K 40K 60K 80K 100K SE +/- 341.24, N = 3 SE +/- 275.86, N = 3 SE +/- 948.29, N = 15 SE +/- 719.41, N = 15 95416.0 103635.0 99601.6 100851.4
OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 128 IBPB off safe RET safe RET no microcode 30K 60K 90K 120K 150K SE +/- 408.63, N = 3 SE +/- 931.05, N = 3 SE +/- 1387.70, N = 15 SE +/- 1043.12, N = 13 119163.8 135187.2 132046.0 131487.0
ClickHouse ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run IBPB off safe RET safe RET no microcode 80 160 240 320 400 SE +/- 2.22, N = 3 SE +/- 2.21, N = 3 SE +/- 1.85, N = 3 SE +/- 3.27, N = 5 349.50 362.64 337.45 329.19 MIN: 31.56 / MAX: 5000 MIN: 31.5 / MAX: 4285.71 MIN: 31.46 / MAX: 4000 MIN: 31.32 / MAX: 2857.14
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run IBPB off safe RET safe RET no microcode 80 160 240 320 400 SE +/- 2.20, N = 3 SE +/- 1.42, N = 3 SE +/- 4.86, N = 3 SE +/- 2.16, N = 5 347.92 361.81 337.12 337.01 MIN: 31.85 / MAX: 3750 MIN: 31.46 / MAX: 4000 MIN: 30.79 / MAX: 4000 MIN: 30.49 / MAX: 3529.41
OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache IBPB off safe RET safe RET no microcode 80 160 240 320 400 SE +/- 2.84, N = 3 SE +/- 0.68, N = 3 SE +/- 2.94, N = 3 SE +/- 3.38, N = 5 336.69 349.43 318.12 323.42 MIN: 31.5 / MAX: 4000 MIN: 31.06 / MAX: 4285.71 MIN: 30.57 / MAX: 3333.33 MIN: 30.82 / MAX: 5000
MariaDB This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 11.0.1 Clients: 4096 IBPB off safe RET safe RET no microcode 130 260 390 520 650 SE +/- 0.71, N = 3 SE +/- 5.48, N = 3 SE +/- 3.51, N = 3 SE +/- 2.96, N = 3 274 590 418 412 1. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -lpthread -ldl
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.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time IBPB off safe RET safe RET no microcode 140 280 420 560 700 645.41 633.52 643.71 644.36 1. (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
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time IBPB off safe RET safe RET no microcode 30 60 90 120 150 148.25 140.62 144.02 145.06 1. (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
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/pathtracer/real_time IBPB off safe RET safe RET no microcode 30 60 90 120 150 SE +/- 0.43, N = 3 SE +/- 0.21, N = 3 SE +/- 0.07, N = 3 SE +/- 1.83, N = 3 153.44 157.83 156.42 155.17
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.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: INIVOL and Fluid Structure Interaction Drop Container IBPB off safe RET safe RET no microcode 40 80 120 160 200 SE +/- 0.17, N = 3 SE +/- 0.16, N = 3 SE +/- 0.50, N = 3 SE +/- 0.39, N = 3 171.75 162.13 163.97 163.02
Apache IoTDB OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 30 60 90 120 150 SE +/- 1.16, N = 8 SE +/- 0.95, N = 10 SE +/- 0.86, N = 15 SE +/- 1.63, N = 5 120.53 117.97 120.06 123.61 MAX: 4401.37 MAX: 4652.25 MAX: 4495.21 MAX: 4533.33
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 8M 16M 24M 32M 40M SE +/- 327739.29, N = 8 SE +/- 302926.36, N = 10 SE +/- 288707.73, N = 15 SE +/- 394126.89, N = 5 38572529.56 39463981.42 38833415.97 37720117.40
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.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Bird Strike on Windshield IBPB off safe RET safe RET no microcode 40 80 120 160 200 SE +/- 0.89, N = 3 SE +/- 0.07, N = 3 SE +/- 0.73, N = 3 SE +/- 0.67, N = 3 160.70 144.83 152.27 152.91
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/scivis/real_time IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 17.67 17.75 17.73 17.73
Neural Magic DeepSparse OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 150 300 450 600 750 SE +/- 1.14, N = 3 SE +/- 1.21, N = 3 SE +/- 0.96, N = 3 SE +/- 1.47, N = 3 681.40 678.93 682.21 679.58
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 11 22 33 44 55 SE +/- 0.02, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 46.61 46.70 46.57 46.72
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.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency IBPB off safe RET safe RET no microcode 0.1037 0.2074 0.3111 0.4148 0.5185 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.004, N = 3 SE +/- 0.003, N = 3 0.461 0.256 0.289 0.296 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Only IBPB off safe RET safe RET no microcode 700K 1400K 2100K 2800K 3500K SE +/- 2988.66, N = 3 SE +/- 1705.16, N = 3 SE +/- 34286.68, N = 3 SE +/- 29158.10, N = 3 1733827 3128719 2768445 2707280 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 15.85 12.99 14.59 14.50 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL 15 Scaling Factor: 100 - Clients: 800 - Mode: Read Write IBPB off safe RET safe RET no microcode 13K 26K 39K 52K 65K SE +/- 133.40, N = 3 SE +/- 418.78, N = 3 SE +/- 207.71, N = 3 SE +/- 66.28, N = 3 50463 61604 54837 55175 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
Timed MrBayes Analysis This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed MrBayes Analysis 3.2.7 Primate Phylogeny Analysis IBPB off safe RET safe RET no microcode 30 60 90 120 150 SE +/- 1.03, N = 3 SE +/- 0.85, N = 3 SE +/- 1.05, N = 3 SE +/- 0.66, N = 3 144.83 136.69 138.85 137.52 1. (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 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.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Time IBPB off safe RET safe RET no microcode 0.369 0.738 1.107 1.476 1.845 SE +/- 0.03, N = 3 SE +/- 0.01, N = 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 1.64 1.30 1.41 1.39
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Inner Join Test Time IBPB off safe RET safe RET no microcode 0.5423 1.0846 1.6269 2.1692 2.7115 SE +/- 0.04, N = 3 SE +/- 0.02, N = 15 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 2.41 1.88 2.14 2.22
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Repartition Test Time IBPB off safe RET safe RET no microcode 0.5355 1.071 1.6065 2.142 2.6775 SE +/- 0.12, N = 3 SE +/- 0.04, N = 15 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 2.31 2.09 2.26 2.38
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Group By Test Time IBPB off safe RET safe RET no microcode 1.2915 2.583 3.8745 5.166 6.4575 SE +/- 0.06, N = 3 SE +/- 0.04, N = 15 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 5.74 4.91 5.15 5.17
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.20, N = 3 SE +/- 0.12, N = 15 SE +/- 0.33, N = 3 SE +/- 0.01, N = 3 31.49 31.84 31.43 32.02
OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time IBPB off safe RET safe RET no microcode 0.8348 1.6696 2.5044 3.3392 4.174 SE +/- 0.05, N = 3 SE +/- 0.03, N = 15 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 3.71 3.39 3.47 3.42
Apache IoTDB OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 20 40 60 80 100 SE +/- 0.42, N = 3 SE +/- 2.14, N = 3 SE +/- 1.29, N = 3 SE +/- 0.81, N = 4 80.78 78.92 82.24 79.19 MAX: 2592.69 MAX: 1729.94 MAX: 3625.32 MAX: 5165.86
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 13M 26M 39M 52M 65M SE +/- 269354.94, N = 3 SE +/- 817020.04, N = 3 SE +/- 721225.08, N = 3 SE +/- 648692.91, N = 4 57529201.42 58682618.18 57099408.15 58073516.79
Redis 7.0.12 + memtier_benchmark Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ops/sec, More Is Better Redis 7.0.12 + memtier_benchmark 2.0 Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:10 IBPB off safe RET safe RET no microcode 500K 1000K 1500K 2000K 2500K SE +/- 12623.44, N = 3 SE +/- 30210.22, N = 3 SE +/- 792.70, N = 3 SE +/- 16754.20, N = 10 2148876.03 2195705.51 2157815.69 2154339.26 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
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.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Bumper Beam IBPB off safe RET safe RET no microcode 30 60 90 120 150 SE +/- 0.23, N = 3 SE +/- 0.33, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 113.68 87.72 93.90 93.68
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: particle_volume/ao/real_time IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 17.97 18.02 17.98 18.03
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.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Rubber O-Ring Seal Installation IBPB off safe RET safe RET no microcode 20 40 60 80 100 SE +/- 0.34, N = 3 SE +/- 0.23, N = 3 SE +/- 0.24, N = 3 SE +/- 0.17, N = 3 99.04 77.46 84.48 85.04
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.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 IBPB off safe RET safe RET no microcode 40K 80K 120K 160K 200K SE +/- 242.54, N = 3 SE +/- 362.13, N = 3 SE +/- 314.03, N = 3 SE +/- 352.89, N = 3 135431.46 166499.89 143271.26 140555.98 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 IBPB off safe RET safe RET no microcode 40K 80K 120K 160K 200K SE +/- 262.73, N = 3 SE +/- 284.72, N = 3 SE +/- 251.96, N = 3 SE +/- 284.55, N = 3 137051.69 169583.15 142619.84 144020.03 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
Neural Magic DeepSparse OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 12 24 36 48 60 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 55.50 55.39 55.40 55.37
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 120 240 360 480 600 SE +/- 0.94, N = 3 SE +/- 0.44, N = 3 SE +/- 0.39, N = 3 SE +/- 0.39, N = 3 575.73 576.97 576.82 577.06
Blender OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: Pabellon Barcelona - Compute: CPU-Only IBPB off safe RET safe RET no microcode 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 85.63 84.50 84.49 84.69
Apache IoTDB OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 9 18 27 36 45 SE +/- 0.32, N = 3 SE +/- 0.55, N = 15 SE +/- 0.52, N = 15 SE +/- 0.62, N = 3 36.47 37.70 37.53 35.10 MAX: 808.57 MAX: 802.64 MAX: 755.16 MAX: 728.37
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 100 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 10M 20M 30M 40M 50M SE +/- 146499.20, N = 3 SE +/- 574678.74, N = 15 SE +/- 543529.82, N = 15 SE +/- 614274.26, N = 3 44816394.74 43665846.28 44027904.89 46538766.01
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.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: defconfig IBPB off safe RET safe RET no microcode 9 18 27 36 45 SE +/- 0.37, N = 7 SE +/- 0.34, N = 5 SE +/- 0.37, N = 6 SE +/- 0.35, N = 6 40.09 31.19 37.24 37.62
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.org ms, Fewer Is Better OpenVINO 2022.3 Model: Person Detection FP16 - Device: CPU IBPB off safe RET safe RET no microcode 900 1800 2700 3600 4500 SE +/- 10.56, N = 3 SE +/- 14.65, N = 3 SE +/- 10.77, N = 3 SE +/- 6.87, N = 3 4204.34 4092.91 4114.58 4124.74 MIN: 2302.89 / MAX: 4817.72 MIN: 3409.52 / MAX: 4641.43 MIN: 2087 / MAX: 5053.62 MIN: 2129.26 / MAX: 5016.36 1. (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
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.3 Model: Person Detection FP16 - Device: CPU IBPB off safe RET safe RET no microcode 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 7.46 7.68 7.60 7.58 1. (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 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.org Ops/sec, More Is Better Redis 7.0.12 + memtier_benchmark 2.0 Protocol: Redis - Clients: 100 - Set To Get Ratio: 1:5 IBPB off safe RET safe RET no microcode 500K 1000K 1500K 2000K 2500K SE +/- 942.73, N = 3 SE +/- 14704.83, N = 3 SE +/- 4916.89, N = 3 SE +/- 17712.54, N = 3 2126493.29 2197287.30 2145052.14 2167181.09 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.org Ops/sec, More Is Better Redis 7.0.12 + memtier_benchmark 2.0 Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:5 IBPB off safe RET safe RET no microcode 500K 1000K 1500K 2000K 2500K SE +/- 21878.46, N = 3 SE +/- 11955.97, N = 3 SE +/- 1778.76, N = 3 SE +/- 31351.12, N = 3 2092844.22 2204628.92 2145436.26 2218601.79 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.org Ops/sec, More Is Better Redis 7.0.12 + memtier_benchmark 2.0 Protocol: Redis - Clients: 50 - Set To Get Ratio: 1:10 IBPB off safe RET safe RET no microcode 500K 1000K 1500K 2000K 2500K SE +/- 13504.65, N = 3 SE +/- 14630.02, N = 3 SE +/- 2448.62, N = 3 SE +/- 17754.58, N = 3 2137964.98 2177211.80 2172804.71 2173694.77 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
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.org ms, Fewer Is Better OpenVINO 2022.3 Model: Face Detection FP16-INT8 - Device: CPU IBPB off safe RET safe RET no microcode 200 400 600 800 1000 SE +/- 0.42, N = 3 SE +/- 0.32, N = 3 SE +/- 1.09, N = 3 SE +/- 0.27, N = 3 1144.98 1141.43 1142.53 1142.29 MIN: 502.04 / MAX: 1175.93 MIN: 998.76 / MAX: 1165.45 MIN: 999.01 / MAX: 1177.02 MIN: 985.75 / MAX: 1168.76 1. (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
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.3 Model: Face Detection FP16-INT8 - Device: CPU IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 27.71 27.83 27.82 27.79 1. (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 This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Read Random Write Random IBPB off safe RET safe RET no microcode 600K 1200K 1800K 2400K 3000K SE +/- 7875.65, N = 3 SE +/- 35283.44, N = 4 SE +/- 21895.20, N = 3 SE +/- 18652.89, N = 3 2130006 2951684 2839085 2872765 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/scivis/real_time IBPB off safe RET safe RET no microcode 2 4 6 8 10 SE +/- 0.02088, N = 3 SE +/- 0.00864, N = 3 SE +/- 0.01059, N = 3 SE +/- 0.01659, N = 3 8.26549 8.33174 8.33813 8.32749
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/ao/real_time IBPB off safe RET safe RET no microcode 3 6 9 12 15 SE +/- 0.01872, N = 3 SE +/- 0.02460, N = 3 SE +/- 0.01456, N = 3 SE +/- 0.02941, N = 3 8.89239 8.96051 8.94049 8.96941
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.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe IBPB 0.585 1.17 1.755 2.34 2.925 SE +/- 0.08, N = 3 2.60
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.org ms, Fewer Is Better OpenVINO 2022.3 Model: Weld Porosity Detection FP16 - Device: CPU IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 28.62 28.40 28.39 28.38 MIN: 14.91 / MAX: 49.84 MIN: 14.74 / MAX: 48.66 MIN: 14.64 / MAX: 50.33 MIN: 14.89 / MAX: 51.63 1. (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
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.3 Model: Weld Porosity Detection FP16 - Device: CPU IBPB off safe RET safe RET no microcode 200 400 600 800 1000 SE +/- 0.40, N = 3 SE +/- 0.17, N = 3 SE +/- 0.13, N = 3 SE +/- 0.72, N = 3 1116.89 1126.03 1126.14 1126.64 1. (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 OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 15 30 45 60 75 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 65.78 65.59 65.57 65.89
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 110 220 330 440 550 SE +/- 1.48, N = 3 SE +/- 1.15, N = 3 SE +/- 1.01, N = 3 SE +/- 0.96, N = 3 485.68 487.25 487.37 485.01
RocksDB This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better RocksDB 8.0 Test: Update Random IBPB off safe RET safe RET no microcode 100K 200K 300K 400K 500K SE +/- 110.81, N = 3 SE +/- 893.82, N = 3 SE +/- 185.49, N = 3 SE +/- 426.73, N = 3 322231 462287 426947 428112 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Neural Magic DeepSparse OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 130 260 390 520 650 SE +/- 0.06, N = 3 SE +/- 0.18, N = 3 SE +/- 0.11, N = 3 SE +/- 0.26, N = 3 596.79 596.59 596.78 596.56
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 12 24 36 48 60 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 53.58 53.60 53.57 53.60
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time IBPB off safe RET safe RET no microcode 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.13, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 13.17 13.14 13.25 13.26
Apache IoTDB OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 9 18 27 36 45 SE +/- 0.44, N = 3 SE +/- 0.61, N = 3 SE +/- 0.49, N = 3 SE +/- 0.11, N = 3 36.56 36.58 35.82 38.54 MAX: 2253.21 MAX: 2252.73 MAX: 3267.55 MAX: 3276.77
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 11M 22M 33M 44M 55M SE +/- 616490.96, N = 3 SE +/- 681823.31, N = 3 SE +/- 634314.77, N = 3 SE +/- 147114.88, N = 3 49316970.28 49501499.13 50578426.54 47445770.18
Neural Magic DeepSparse OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 200 400 600 800 1000 SE +/- 0.53, N = 3 SE +/- 0.50, N = 3 SE +/- 0.34, N = 3 SE +/- 0.36, N = 3 840.89 839.72 840.42 840.77
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 9 18 27 36 45 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.07, N = 3 37.68 37.60 37.63 37.70
Apache IoTDB OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.19, N = 9 SE +/- 0.19, N = 3 SE +/- 0.21, N = 8 SE +/- 0.16, N = 12 14.78 13.83 14.73 14.05 MAX: 618.06 MAX: 596.78 MAX: 645.11 MAX: 609.96
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 200K 400K 600K 800K 1000K SE +/- 7583.18, N = 9 SE +/- 8467.91, N = 3 SE +/- 7998.38, N = 8 SE +/- 6730.71, N = 12 921701.44 960525.66 918691.45 947741.34
OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.35, N = 3 SE +/- 0.29, N = 3 SE +/- 0.22, N = 3 SE +/- 0.49, N = 3 31.11 31.49 27.73 31.93 MAX: 908.02 MAX: 939.96 MAX: 938.92 MAX: 930.97
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 300K 600K 900K 1200K 1500K SE +/- 6687.04, N = 3 SE +/- 4294.81, N = 3 SE +/- 5073.96, N = 3 SE +/- 13029.07, N = 3 1441637.63 1415756.33 1583717.62 1408658.83
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.org Seconds, Fewer Is Better OpenRadioss 2022.10.13 Model: Cell Phone Drop Test IBPB off safe RET safe RET no microcode 9 18 27 36 45 SE +/- 0.14, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.26, N = 3 40.11 33.10 36.40 36.37
GROMACS The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2023 Implementation: MPI CPU - Input: water_GMX50_bare IBPB off safe RET safe RET no microcode 1.2893 2.5786 3.8679 5.1572 6.4465 SE +/- 0.011, N = 3 SE +/- 0.012, N = 3 SE +/- 0.010, N = 3 SE +/- 0.006, N = 3 5.707 5.680 5.706 5.730 1. (CXX) g++ options: -O3
Neural Magic DeepSparse OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 15 30 45 60 75 SE +/- 0.07, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 68.35 68.15 68.23 68.26
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 100 200 300 400 500 SE +/- 0.43, N = 3 SE +/- 0.24, N = 3 SE +/- 0.31, N = 3 SE +/- 0.42, N = 3 467.32 468.83 468.22 468.36
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating IBPB off safe RET safe RET no microcode 80K 160K 240K 320K 400K SE +/- 248.93, N = 3 SE +/- 1018.75, N = 3 SE +/- 435.27, N = 3 SE +/- 25.38, N = 3 371799 384374 335595 334812 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Algebraic Multi-Grid Benchmark AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 IBPB off safe RET safe RET no microcode 200M 400M 600M 800M 1000M SE +/- 1724277.85, N = 3 SE +/- 839009.73, N = 3 SE +/- 367255.40, N = 3 SE +/- 575791.94, N = 3 1005138667 1011799000 999102100 999645400 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
Neural Magic DeepSparse OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 2 4 6 8 10 SE +/- 0.0189, N = 3 SE +/- 0.0095, N = 3 SE +/- 0.0271, N = 3 SE +/- 0.0167, N = 3 8.3411 8.3078 8.3219 8.3612
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.5 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream IBPB off safe RET safe RET no microcode 800 1600 2400 3200 4000 SE +/- 7.83, N = 3 SE +/- 4.76, N = 3 SE +/- 12.28, N = 3 SE +/- 7.72, N = 3 3824.84 3840.63 3834.24 3816.77
Remhos Remhos (REMap High-Order Solver) is a miniapp that solves the pure advection equations that are used to perform monotonic and conservative discontinuous field interpolation (remap) as part of the Eulerian phase in Arbitrary Lagrangian Eulerian (ALE) simulations. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Remhos 1.0 Test: Sample Remap Example IBPB off safe RET safe RET no microcode 5 10 15 20 25 SE +/- 0.12, N = 14 SE +/- 0.23, N = 3 SE +/- 0.19, N = 3 SE +/- 0.17, N = 3 18.66 17.38 17.96 17.79 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
SPECFEM3D simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Layered Halfspace IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.21, N = 3 SE +/- 0.35, N = 3 SE +/- 0.18, N = 3 SE +/- 0.23, N = 3 32.01 31.85 31.66 31.83 1. (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
OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Water-layered Halfspace IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.19, N = 3 SE +/- 0.15, N = 3 SE +/- 0.35, N = 3 SE +/- 0.25, N = 3 30.05 29.77 29.59 30.43 1. (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 OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 8 16 24 32 40 SE +/- 0.42, N = 4 SE +/- 0.24, N = 3 SE +/- 0.22, N = 3 SE +/- 0.02, N = 3 31.74 32.36 30.14 30.29 MAX: 667.18 MAX: 646.51 MAX: 641.04 MAX: 715.01
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 200 - Batch Size Per Write: 1 - Sensor Count: 500 IBPB off safe RET safe RET no microcode 300K 600K 900K 1200K 1500K SE +/- 14032.06, N = 4 SE +/- 7578.67, N = 3 SE +/- 9180.92, N = 3 SE +/- 1525.49, N = 3 1287324.35 1271946.57 1345598.59 1342031.11
OpenBenchmarking.org Average Latency, Fewer Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.13, N = 3 SE +/- 0.07, N = 3 SE +/- 0.19, N = 3 SE +/- 0.14, N = 3 11.93 14.05 13.36 13.61 MAX: 855.56 MAX: 858.17 MAX: 881.3 MAX: 854.4
OpenBenchmarking.org point/sec, More Is Better Apache IoTDB 1.1.2 Device Count: 500 - Batch Size Per Write: 1 - Sensor Count: 200 IBPB off safe RET safe RET no microcode 300K 600K 900K 1200K 1500K SE +/- 3166.01, N = 3 SE +/- 1566.77, N = 3 SE +/- 4253.29, N = 3 SE +/- 6553.14, N = 3 1344749.20 1176385.35 1211172.13 1202637.36
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms IBPB off safe RET safe RET no microcode 0.0867 0.1734 0.2601 0.3468 0.4335 SE +/- 0.00026, N = 3 SE +/- 0.00017, N = 3 SE +/- 0.00028, N = 3 SE +/- 0.00029, N = 3 0.38534 0.38130 0.38098 0.38115
Blender OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.6 Blend File: BMW27 - Compute: CPU-Only IBPB off safe RET safe RET no microcode 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 27.73 27.34 27.46 27.58
SPECFEM3D simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Homogeneous Halfspace IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.11, N = 3 SE +/- 0.07, N = 3 SE +/- 0.21, N = 3 SE +/- 0.20, N = 4 17.70 17.42 17.69 17.64 1. (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
OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Tomographic Model IBPB off safe RET safe RET no microcode 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 14.23 14.13 14.19 14.40 1. (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 simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SPECFEM3D 4.0 Model: Mount St. Helens IBPB off safe RET safe RET no microcode 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 11.95 11.80 12.01 11.98 1. (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 Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Crown IBPB off safe RET safe RET no microcode 13 26 39 52 65 SE +/- 0.10, N = 3 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 SE +/- 0.14, N = 3 57.30 57.42 57.31 57.30 MIN: 56.3 / MAX: 58.61 MIN: 56.59 / MAX: 58.54 MIN: 56.2 / MAX: 58.59 MIN: 56.26 / MAX: 58.69
OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.1 Binary: Pathtracer ISPC - Model: Asian Dragon IBPB off safe RET safe RET no microcode 14 28 42 56 70 SE +/- 0.14, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 63.47 64.60 64.67 64.39 MIN: 62.67 / MAX: 65.74 MIN: 64.05 / MAX: 66.13 MIN: 64.11 / MAX: 66.01 MIN: 63.77 / MAX: 66.16
SQLite This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database with a variable number of concurrent repetitions -- up to the maximum number of CPU threads available. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.41.2 Threads / Copies: 16 IBPB off safe RET safe RET no microcode 2 4 6 8 10 SE +/- 0.007, N = 3 SE +/- 0.020, N = 3 SE +/- 0.052, N = 3 SE +/- 0.024, N = 3 7.934 6.273 8.800 8.834 1. (CC) gcc options: -O2 -lz -lm
OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.41.2 Threads / Copies: 8 IBPB off safe RET safe RET no microcode 1.1264 2.2528 3.3792 4.5056 5.632 SE +/- 0.010, N = 3 SE +/- 0.013, N = 3 SE +/- 0.036, N = 3 SE +/- 0.016, N = 3 4.793 3.755 5.006 4.850 1. (CC) gcc options: -O2 -lz -lm
off Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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
Testing initiated at 10 August 2023 05:20 by user phoronix.
safe RET no microcode Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001173Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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
Testing initiated at 9 August 2023 18:46 by user phoronix.
safe RET Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa0011d1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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
Testing initiated at 10 August 2023 12:35 by user phoronix.
IBPB Processor: AMD EPYC 7763 64-Core @ 2.45GHz (64 Cores / 128 Threads), Motherboard: AMD DAYTONA_X (RYM1009B BIOS), Chipset: AMD Starship/Matisse, Memory: 256GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VE228, Network: 2 x Mellanox MT27710
OS: Ubuntu 22.04, Kernel: 6.5.0-rc5-phx-tues (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3, Vulkan: 1.3.224, Compiler: GCC 11.3.0 + LLVM 14.0.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa0011d1Java Notes: OpenJDK Runtime Environment (build 11.0.20+8-post-Ubuntu-1ubuntu122.04)Python Notes: Python 3.10.6Security Notes: 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
Testing initiated at 10 August 2023 20:55 by user phoronix.