2 x AMD EPYC 9684X 96-Core testing with a AMD Titanite_4G (RTI1007B BIOS) and ASPEED on Ubuntu 23.10 via the Phoronix Test Suite.
Linux 6.5 Processor: 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.5.0-060500-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Linux 6.7 OS: Ubuntu 23.10, Kernel: 6.7.0-060700-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Linux 6.6 OS: Ubuntu 23.10, Kernel: 6.6.0-060600-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Linux 6.8 26 Jan OS: Ubuntu 23.10, Kernel: 6.8.0-060800rc1daily20240126-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Currently this test profile is catered to CPU-based testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: Efficientnet_v2_l Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 0.5243 1.0486 1.5729 2.0972 2.6215 SE +/- 0.01, N = 9 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 1.26 1.11 1.18 2.33 MIN: 0.49 / MAX: 2.38 MIN: 0.49 / MAX: 2.12 MIN: 0.47 / MAX: 2.46 MIN: 1.64 / MAX: 2.77
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: Efficientnet_v2_l Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.07, N = 12 SE +/- 0.07, N = 3 SE +/- 0.05, N = 10 5.50 6.55 5.56 6.54 MIN: 2.96 / MAX: 5.62 MIN: 3.01 / MAX: 7 MIN: 3.52 / MAX: 5.71 MIN: 3.15 / MAX: 7.04
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: 512 - Model: ResNet-50 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30 60 90 120 150 SE +/- 0.29, N = 3 SE +/- 0.15, N = 3 SE +/- 0.48, N = 3 SE +/- 0.50, N = 3 145.75 143.21 144.07 142.76
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 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 0.063 0.126 0.189 0.252 0.315 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 SE +/- 0.006, N = 12 SE +/- 0.003, N = 12 0.268 0.268 0.280 0.276 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 800K 1600K 2400K 3200K 4000K SE +/- 40627.80, N = 3 SE +/- 15124.24, N = 3 SE +/- 68594.34, N = 12 SE +/- 35614.94, N = 12 3731648 3729700 3588755 3624464 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
Llama.cpp Llama.cpp is a port of Facebook's LLaMA model in C/C++ developed by Georgi Gerganov. Llama.cpp allows the inference of LLaMA and other supported models in C/C++. For CPU inference Llama.cpp supports AVX2/AVX-512, ARM NEON, and other modern ISAs along with features like OpenBLAS usage. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Tokens Per Second, More Is Better Llama.cpp b1808 Model: llama-2-70b-chat.Q5_0.gguf Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1.206 2.412 3.618 4.824 6.03 SE +/- 0.01, N = 3 SE +/- 0.22, N = 12 SE +/- 0.12, N = 12 SE +/- 0.01, N = 3 5.24 3.57 3.42 5.36 1. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Currently this test profile is catered to CPU-based testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: ResNet-152 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 2 4 6 8 10 SE +/- 0.07, N = 5 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 7.27 8.72 7.20 8.78 MIN: 3.96 / MAX: 7.61 MIN: 4.82 / MAX: 9.24 MIN: 6.97 / MAX: 7.42 MIN: 4.5 / MAX: 9.28
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 While Writing Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5M 10M 15M 20M 25M SE +/- 428949.64, N = 15 SE +/- 531525.96, N = 15 SE +/- 601741.45, N = 15 SE +/- 374386.25, N = 15 15095355 17014088 16760946 21198818 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Speedb Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Speedb 2.7 Test: Read While Writing Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 8M 16M 24M 32M 40M SE +/- 1405761.06, N = 12 SE +/- 840379.83, N = 15 SE +/- 855245.74, N = 12 SE +/- 511361.34, N = 15 33717802 32574249 32475216 39636850 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
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: 1 - Model: ResNet-50 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 0.477 0.954 1.431 1.908 2.385 SE +/- 0.02, N = 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 15 SE +/- 0.03, N = 15 2.01 2.12 2.08 2.04
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Currently this test profile is catered to CPU-based testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-152 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 3 6 9 12 15 SE +/- 0.06, N = 11 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.15, N = 3 8.31 10.33 8.25 10.57 MIN: 3.51 / MAX: 8.85 MIN: 4.78 / MAX: 11.26 MIN: 3.98 / MAX: 8.67 MIN: 4.56 / MAX: 11.43
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5 10 15 20 25 SE +/- 0.19, N = 3 SE +/- 0.06, N = 3 SE +/- 0.12, N = 3 SE +/- 0.14, N = 14 17.43 20.93 17.61 21.08 MIN: 9.78 / MAX: 18.32 MIN: 13.14 / MAX: 22.13 MIN: 11.32 / MAX: 18.41 MIN: 11.86 / MAX: 22.44
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 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5 10 15 20 25 SE +/- 0.13, N = 3 SE +/- 0.04, N = 3 SE +/- 0.40, N = 9 SE +/- 0.20, N = 3 17.82 18.75 17.35 18.45 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 12K 24K 36K 48K 60K SE +/- 395.45, N = 3 SE +/- 125.02, N = 3 SE +/- 1379.59, N = 9 SE +/- 582.81, N = 3 56119 53339 57875 54210 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
Timed Gem5 Compilation This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed Gem5 Compilation 23.0.1 Time To Compile Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30 60 90 120 150 SE +/- 0.81, N = 3 SE +/- 0.54, N = 3 SE +/- 1.03, N = 3 SE +/- 1.63, N = 4 154.44 153.72 153.21 157.44
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 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 40 80 120 160 200 SE +/- 1.33, N = 3 SE +/- 0.70, N = 3 SE +/- 0.20, N = 3 SE +/- 0.09, N = 3 156.23 165.56 161.38 168.09
CacheBench This is a performance test of CacheBench, which is part of LLCbench. CacheBench is designed to test the memory and cache bandwidth performance Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 2K 4K 6K 8K 10K SE +/- 0.19, N = 3 SE +/- 0.04, N = 3 SE +/- 0.36, N = 3 SE +/- 0.11, N = 3 9388.76 9389.42 9389.92 9390.24 MIN: 9387.83 / MAX: 9389.93 MIN: 9387.81 / MAX: 9390.39 MIN: 9376.26 / MAX: 9391.12 MIN: 9389.41 / MAX: 9390.99 1. (CC) gcc options: -O3 -lrt
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Mixed Scheduler Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30K 60K 90K 120K 150K SE +/- 1003.39, N = 15 SE +/- 1093.17, N = 15 SE +/- 1357.21, N = 15 SE +/- 1466.78, N = 4 134318.71 127334.72 133980.52 127531.88 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
CacheBench This is a performance test of CacheBench, which is part of LLCbench. CacheBench is designed to test the memory and cache bandwidth performance Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read / Modify / Write Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 20K 40K 60K 80K 100K SE +/- 77.76, N = 3 SE +/- 145.98, N = 3 SE +/- 36.94, N = 3 SE +/- 39.89, N = 3 107666.91 107670.84 107675.92 107547.07 MIN: 81021.88 / MAX: 111825 MIN: 81031.82 / MAX: 111822.61 MIN: 81045.61 / MAX: 111844.98 MIN: 81047.89 / MAX: 111835.1 1. (CC) gcc options: -O3 -lrt
OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Write Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 12K 24K 36K 48K 60K SE +/- 0.22, N = 3 SE +/- 0.98, N = 3 SE +/- 0.28, N = 3 SE +/- 0.29, N = 3 56268.96 56272.87 56280.20 56278.61 MIN: 56066.44 / MAX: 56333.3 MIN: 56070.22 / MAX: 56338.93 MIN: 56079.62 / MAX: 56345.05 MIN: 56074.59 / MAX: 56343.25 1. (CC) gcc options: -O3 -lrt
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Matrix 3D Math Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5K 10K 15K 20K 25K SE +/- 1628.18, N = 12 SE +/- 1714.00, N = 12 SE +/- 1654.57, N = 12 SE +/- 1665.45, N = 12 21058.81 20769.77 21261.98 21674.06 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: IO_uring Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 4M 8M 12M 16M 20M SE +/- 20817.51, N = 3 SE +/- 1231639.25, N = 12 SE +/- 245563.90, N = 3 SE +/- 2384203.44, N = 12 15943494.88 12307832.19 18217916.69 9895683.75 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
PyTorch This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Currently this test profile is catered to CPU-based testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org batches/sec, More Is Better PyTorch 2.1 Device: CPU - Batch Size: 1 - Model: ResNet-50 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 6 12 18 24 30 SE +/- 0.16, N = 3 SE +/- 0.23, N = 15 SE +/- 0.20, N = 5 SE +/- 0.34, N = 3 19.26 23.47 19.23 23.66 MIN: 11.04 / MAX: 20.23 MIN: 9.72 / MAX: 25.75 MIN: 11.52 / MAX: 20.48 MIN: 13.25 / MAX: 25.04
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Pipe Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 15M 30M 45M 60M 75M SE +/- 812857.42, N = 15 SE +/- 1231347.57, N = 15 SE +/- 829037.53, N = 3 SE +/- 1232934.62, N = 12 66998085.99 61084286.26 61641558.02 71470186.96 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Cloning Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 3K 6K 9K 12K 15K SE +/- 1911.22, N = 12 SE +/- 1195.63, N = 12 SE +/- 1944.59, N = 12 SE +/- 102.00, N = 3 9729.15 9532.62 9747.48 11924.37 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Futex Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 600K 1200K 1800K 2400K 3000K SE +/- 28376.36, N = 15 SE +/- 35379.01, N = 15 SE +/- 24021.26, N = 5 SE +/- 30946.87, N = 3 2337644.18 2243204.99 2410851.80 2996767.15 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Glibc C String Functions Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 40M 80M 120M 160M 200M SE +/- 789114.28, N = 3 SE +/- 1978439.64, N = 15 SE +/- 1989236.60, N = 5 SE +/- 1655443.15, N = 15 196585232.62 204900453.75 201218048.98 207322045.97 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
Llama.cpp Llama.cpp is a port of Facebook's LLaMA model in C/C++ developed by Georgi Gerganov. Llama.cpp allows the inference of LLaMA and other supported models in C/C++. For CPU inference Llama.cpp supports AVX2/AVX-512, ARM NEON, and other modern ISAs along with features like OpenBLAS usage. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Tokens Per Second, More Is Better Llama.cpp b1808 Model: llama-2-13b.Q4_0.gguf Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 4 8 12 16 20 SE +/- 0.47, N = 12 SE +/- 0.05, N = 3 SE +/- 0.58, N = 15 SE +/- 0.10, N = 3 16.49 17.17 15.70 17.15 1. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Forking Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 8K 16K 24K 32K 40K SE +/- 39.55, N = 3 SE +/- 165.68, N = 3 SE +/- 436.72, N = 15 SE +/- 297.13, N = 13 33018.31 37276.20 36748.37 36691.26 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
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 2023.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 0.0765 0.153 0.2295 0.306 0.3825 SE +/- 0.00, N = 3 SE +/- 0.00, N = 6 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.34 0.34 0.34 0.34 MIN: 0.33 / MAX: 44.41 MIN: 0.33 / MAX: 39.24 MIN: 0.33 / MAX: 50.17 MIN: 0.33 / MAX: 35.16 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30K 60K 90K 120K 150K SE +/- 1746.03, N = 3 SE +/- 1446.63, N = 6 SE +/- 294.88, N = 3 SE +/- 1456.84, N = 3 156828.34 156517.91 156137.76 158019.60 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
Llamafile Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.6 Test: llava-v1.5-7b-q4 - Acceleration: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5 10 15 20 25 SE +/- 0.32, N = 15 SE +/- 0.52, N = 15 SE +/- 0.50, N = 15 SE +/- 0.58, N = 15 20.81 21.37 20.85 21.35
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: Barbershop - Compute: CPU-Only Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 16 32 48 64 80 SE +/- 0.25, N = 3 SE +/- 0.06, N = 3 SE +/- 0.18, N = 3 SE +/- 0.12, N = 3 70.11 70.62 69.63 70.76
Llamafile Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.6 Test: mistral-7b-instruct-v0.2.Q8_0 - Acceleration: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 13.52 13.47 13.61 13.53
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: CPU Cache Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200K 400K 600K 800K 1000K SE +/- 894.71, N = 3 SE +/- 1430.34, N = 3 SE +/- 442.06, N = 3 SE +/- 957.96, N = 3 862417.29 840635.00 824396.72 855185.73 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
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 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 7 14 21 28 35 SE +/- 0.21, N = 10 SE +/- 0.26, N = 7 SE +/- 0.22, N = 9 SE +/- 0.28, N = 6 26.88 27.60 27.09 28.28
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 2023.2.dev Model: Face Detection FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 50 100 150 200 250 SE +/- 0.20, N = 3 SE +/- 0.33, N = 3 SE +/- 0.45, N = 3 SE +/- 0.24, N = 3 233.39 233.41 233.10 233.31 MIN: 207.86 / MAX: 270.46 MIN: 206.8 / MAX: 293.22 MIN: 207.01 / MAX: 346.48 MIN: 207.82 / MAX: 247.74 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Face Detection FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 50 100 150 200 250 SE +/- 0.18, N = 3 SE +/- 0.26, N = 3 SE +/- 0.36, N = 3 SE +/- 0.24, N = 3 205.32 205.21 205.45 205.46 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.2.dev Model: Person Detection FP16 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 10 20 30 40 50 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 43.63 43.40 43.61 43.48 MIN: 40.69 / MAX: 93.9 MIN: 40.58 / MAX: 93.21 MIN: 39.44 / MAX: 93.31 MIN: 37.8 / MAX: 109.72 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Person Detection FP16 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200 400 600 800 1000 SE +/- 0.53, N = 3 SE +/- 1.34, N = 3 SE +/- 0.81, N = 3 SE +/- 1.07, N = 3 1099.28 1105.14 1099.92 1103.21 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Malloc Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 70M 140M 210M 280M 350M SE +/- 2654426.38, N = 15 SE +/- 113926.80, N = 3 SE +/- 321332.84, N = 3 SE +/- 583008.10, N = 3 345731944.96 329314591.50 329981338.54 339209442.28 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
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 2023.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 10 20 30 40 50 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 42.96 43.53 43.10 43.39 MIN: 34.02 / MAX: 222.73 MIN: 34.1 / MAX: 246.36 MIN: 33.66 / MAX: 200.61 MIN: 33.85 / MAX: 244.32 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200 400 600 800 1000 SE +/- 1.29, N = 3 SE +/- 1.20, N = 3 SE +/- 1.64, N = 3 SE +/- 1.72, N = 3 1115.48 1101.20 1111.86 1104.56 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.2.dev Model: Face Detection Retail FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1.197 2.394 3.591 4.788 5.985 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 5.32 5.30 5.31 5.31 MIN: 4.83 / MAX: 24.35 MIN: 4.8 / MAX: 26.57 MIN: 4.85 / MAX: 23.09 MIN: 4.87 / MAX: 25.7 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Face Detection Retail FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 8K 16K 24K 32K 40K SE +/- 25.38, N = 3 SE +/- 40.16, N = 3 SE +/- 5.33, N = 3 SE +/- 27.67, N = 3 35914.69 36016.99 35967.57 36011.87 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1.116 2.232 3.348 4.464 5.58 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 4.96 4.96 4.95 4.96 MIN: 4.4 / MAX: 26.65 MIN: 4.48 / MAX: 27.81 MIN: 4.46 / MAX: 20.84 MIN: 4.39 / MAX: 24.18 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 2K 4K 6K 8K 10K SE +/- 7.27, N = 3 SE +/- 2.25, N = 3 SE +/- 4.66, N = 3 SE +/- 11.50, N = 3 9658.44 9648.43 9678.99 9656.03 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: MEMFD Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200 400 600 800 1000 SE +/- 5.11, N = 3 SE +/- 2.46, N = 3 SE +/- 0.33, N = 3 SE +/- 1.98, N = 15 658.67 834.74 840.58 134.12 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
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 2023.2.dev Model: Road Segmentation ADAS FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 11.97 11.93 11.96 11.94 MIN: 11.11 / MAX: 39.94 MIN: 10.66 / MAX: 47.97 MIN: 11.05 / MAX: 41.55 MIN: 10.72 / MAX: 42.19 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Road Segmentation ADAS FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 900 1800 2700 3600 4500 SE +/- 3.26, N = 3 SE +/- 5.04, N = 3 SE +/- 5.04, N = 3 SE +/- 1.53, N = 3 4007.13 4017.92 4008.83 4017.18 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.2.dev Model: Handwritten English Recognition FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 9 18 27 36 45 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.22, N = 3 41.40 41.50 41.52 41.51 MIN: 38.14 / MAX: 58.29 MIN: 38.11 / MAX: 59.33 MIN: 37.94 / MAX: 61.47 MIN: 37.79 / MAX: 60.99 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Handwritten English Recognition FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1000 2000 3000 4000 5000 SE +/- 6.46, N = 3 SE +/- 10.46, N = 3 SE +/- 7.21, N = 3 SE +/- 24.48, N = 3 4633.97 4623.48 4620.79 4622.46 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Context Switching Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 9M 18M 27M 36M 45M SE +/- 161811.56, N = 3 SE +/- 126646.22, N = 3 SE +/- 58903.36, N = 3 SE +/- 735648.32, N = 15 41473670.27 42558881.72 42973852.02 35325682.86 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
Speedb Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Speedb 2.7 Test: Update Random Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 70K 140K 210K 280K 350K SE +/- 263.27, N = 3 SE +/- 3481.85, N = 3 SE +/- 968.55, N = 3 SE +/- 2205.56, N = 3 328490 314786 314332 311661 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 9.14 9.13 9.10 9.09 MIN: 8.24 / MAX: 47.23 MIN: 8.27 / MAX: 37.96 MIN: 8.21 / MAX: 47.59 MIN: 8.25 / MAX: 46.94 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 4K 8K 12K 16K 20K SE +/- 16.20, N = 3 SE +/- 40.35, N = 3 SE +/- 15.09, N = 3 SE +/- 17.98, N = 3 20824.03 20862.69 20905.08 20939.79 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2023.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 0.9113 1.8226 2.7339 3.6452 4.5565 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 4.05 4.05 4.04 4.04 MIN: 3.64 / MAX: 25.83 MIN: 3.62 / MAX: 19.96 MIN: 3.59 / MAX: 25.59 MIN: 3.59 / MAX: 22.24 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
OpenBenchmarking.org FPS, More Is Better OpenVINO 2023.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 3K 6K 9K 12K 15K SE +/- 2.19, N = 3 SE +/- 16.20, N = 3 SE +/- 8.61, N = 3 SE +/- 14.65, N = 3 11826.99 11817.52 11846.98 11840.17 1. (CXX) g++ options: -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -pie
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 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 90K 180K 270K 360K 450K SE +/- 735.51, N = 3 SE +/- 204.01, N = 3 SE +/- 317.62, N = 3 SE +/- 1449.36, N = 3 432923 411177 406964 405246 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Speedb Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Speedb 2.7 Test: Read Random Write Random Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 300K 600K 900K 1200K 1500K SE +/- 13303.55, N = 3 SE +/- 3477.16, N = 3 SE +/- 12600.29, N = 3 SE +/- 8610.85, N = 3 1594058 1585086 1580931 1627225 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
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 Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 300K 600K 900K 1200K 1500K SE +/- 1372.82, N = 3 SE +/- 2394.21, N = 3 SE +/- 1389.79, N = 3 SE +/- 12296.87, N = 3 1581099 1589032 1585552 1594895 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Speedb Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Speedb 2.7 Test: Random Read Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200M 400M 600M 800M 1000M SE +/- 14530585.11, N = 3 SE +/- 650677.38, N = 3 SE +/- 2336203.11, N = 3 SE +/- 3522165.63, N = 3 1102902083 1128431155 1117431090 1119235930 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
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: Random Read Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200M 400M 600M 800M 1000M SE +/- 1037425.37, N = 3 SE +/- 1094497.51, N = 3 SE +/- 2253024.48, N = 3 SE +/- 3926608.50, N = 3 1098560872 1103280876 1101418931 1103782203 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: SENDFILE Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 600K 1200K 1800K 2400K 3000K SE +/- 16147.19, N = 3 SE +/- 24274.27, N = 4 SE +/- 14114.72, N = 3 SE +/- 55032.25, N = 12 2846032.14 2122557.99 2503374.61 1475163.44 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Pthread Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 13K 26K 39K 52K 65K SE +/- 224.09, N = 3 SE +/- 806.76, N = 3 SE +/- 486.25, N = 9 SE +/- 246.97, N = 3 60918.24 60556.77 61798.90 58661.31 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
Llamafile Mozilla's Llamafile allows distributing and running large language models (LLMs) as a single file. Llamafile aims to make open-source LLMs more accessible to developers and users. Llamafile supports a variety of models, CPUs and GPUs, and other options. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Tokens Per Second, More Is Better Llamafile 0.6 Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 6.23 6.21 6.21 6.21
7-Zip Compression This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 300K 600K 900K 1200K 1500K SE +/- 8322.15, N = 3 SE +/- 2795.86, N = 3 SE +/- 5123.49, N = 3 SE +/- 3058.54, N = 3 1355261 1351742 1358817 1360807 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200K 400K 600K 800K 1000K SE +/- 9321.76, N = 3 SE +/- 2678.19, N = 3 SE +/- 8666.07, N = 3 SE +/- 4619.10, N = 3 876753 874679 889667 876040 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Atomic Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 40 80 120 160 200 SE +/- 0.33, N = 3 SE +/- 0.66, N = 3 SE +/- 0.78, N = 3 SE +/- 1.28, N = 3 182.38 187.72 183.27 187.90 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
Llama.cpp Llama.cpp is a port of Facebook's LLaMA model in C/C++ developed by Georgi Gerganov. Llama.cpp allows the inference of LLaMA and other supported models in C/C++. For CPU inference Llama.cpp supports AVX2/AVX-512, ARM NEON, and other modern ISAs along with features like OpenBLAS usage. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Tokens Per Second, More Is Better Llama.cpp b1808 Model: llama-2-7b.Q4_0.gguf Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5 10 15 20 25 SE +/- 0.20, N = 3 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 SE +/- 0.50, N = 12 22.45 22.27 22.23 21.13 1. (CXX) g++ options: -std=c++11 -fPIC -O3 -pthread -march=native -mtune=native -lopenblas
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: MMAP Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1000 2000 3000 4000 5000 SE +/- 6.99, N = 3 SE +/- 4.13, N = 3 SE +/- 1.99, N = 3 SE +/- 11.50, N = 3 4270.84 4685.26 4627.00 4551.91 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: NUMA Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 150 300 450 600 750 SE +/- 3.92, N = 3 SE +/- 1.38, N = 3 SE +/- 3.34, N = 3 SE +/- 5.42, N = 3 696.82 679.34 686.37 691.43 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Matrix Math Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 200K 400K 600K 800K 1000K SE +/- 1530.59, N = 3 SE +/- 2651.09, N = 3 SE +/- 2409.20, N = 3 SE +/- 2446.43, N = 3 947292.95 944251.03 928621.38 943747.15 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Zlib Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 5K 10K 15K 20K 25K SE +/- 36.08, N = 3 SE +/- 35.85, N = 3 SE +/- 11.46, N = 3 SE +/- 17.84, N = 3 25320.45 25301.41 25311.81 25361.45 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Glibc Qsort Data Sorting Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1000 2000 3000 4000 5000 SE +/- 4.07, N = 3 SE +/- 8.41, N = 3 SE +/- 6.43, N = 3 SE +/- 2.91, N = 3 4658.25 4658.15 4664.11 4661.07 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Poll Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 4M 8M 12M 16M 20M SE +/- 32342.64, N = 3 SE +/- 2956.12, N = 3 SE +/- 3875.34, N = 3 SE +/- 2079.04, N = 3 18907852.31 13211019.31 19277506.61 13095189.44 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: AVX-512 VNNI Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 4M 8M 12M 16M 20M SE +/- 33202.31, N = 3 SE +/- 16761.27, N = 3 SE +/- 16941.68, N = 3 SE +/- 34935.06, N = 3 19514410.29 19338319.36 19568145.68 19310087.82 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Shuffle Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30K 60K 90K 120K 150K SE +/- 149.78, N = 3 SE +/- 59.67, N = 3 SE +/- 242.92, N = 3 SE +/- 149.16, N = 3 138787.10 138846.02 138698.52 138701.98 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: System V Message Passing Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 6M 12M 18M 24M 30M SE +/- 17991.29, N = 3 SE +/- 32811.36, N = 3 SE +/- 26751.58, N = 3 SE +/- 34165.06, N = 3 28486279.67 28183544.82 20162835.90 28034733.60 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Floating Point Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 14K 28K 42K 56K 70K SE +/- 42.50, N = 3 SE +/- 86.63, N = 3 SE +/- 67.12, N = 3 SE +/- 56.56, N = 3 65869.48 66160.14 66153.99 66221.40 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Function Call Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30K 60K 90K 120K 150K SE +/- 190.36, N = 3 SE +/- 166.05, N = 3 SE +/- 70.27, N = 3 SE +/- 110.30, N = 3 158008.66 158205.83 158272.40 158411.00 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Mutex Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 10M 20M 30M 40M 50M SE +/- 286875.66, N = 3 SE +/- 560077.05, N = 3 SE +/- 115602.55, N = 3 SE +/- 481173.64, N = 3 48020269.46 44546973.96 47405607.15 46761181.37 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Floating Point Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 100K 200K 300K 400K 500K SE +/- 350.71, N = 3 SE +/- 411.82, N = 3 SE +/- 491.76, N = 3 SE +/- 485.94, N = 3 456329.90 456150.17 456415.88 456920.91 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Crypto Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 110K 220K 330K 440K 550K SE +/- 710.36, N = 3 SE +/- 2866.62, N = 3 SE +/- 3698.19, N = 3 SE +/- 1454.92, N = 3 492474.65 488061.57 488178.36 492021.86 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Semaphores Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 70M 140M 210M 280M 350M SE +/- 762255.36, N = 3 SE +/- 3210083.71, N = 3 SE +/- 2323869.92, N = 3 SE +/- 698011.13, N = 3 321848076.37 294597947.60 316273290.00 301864524.10 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Fused Multiply-Add Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 90M 180M 270M 360M 450M SE +/- 775526.64, N = 3 SE +/- 792348.84, N = 3 SE +/- 1012815.75, N = 3 SE +/- 119630.16, N = 3 441825227.41 441769520.27 441959661.17 441801423.41 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: CPU Stress Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 100K 200K 300K 400K 500K SE +/- 1639.51, N = 3 SE +/- 1508.14, N = 3 SE +/- 1257.93, N = 3 SE +/- 1017.81, N = 3 460093.34 460897.89 458555.64 460475.32 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Wide Vector Math Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 9854.66, N = 3 SE +/- 5891.75, N = 3 SE +/- 5847.18, N = 3 SE +/- 1610.73, N = 3 5884184.30 5893175.32 5886158.59 5890325.79 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Vector Math Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 300K 600K 900K 1200K 1500K SE +/- 1614.72, N = 3 SE +/- 990.33, N = 3 SE +/- 635.24, N = 3 SE +/- 554.48, N = 3 1331024.12 1332249.56 1333950.70 1336243.88 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Memory Copying Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 13K 26K 39K 52K 65K SE +/- 82.66, N = 3 SE +/- 99.29, N = 3 SE +/- 50.44, N = 3 SE +/- 52.89, N = 3 60755.11 60735.60 60600.94 60720.85 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: x86_64 RdRand Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 16M 32M 48M 64M 80M SE +/- 16235.65, N = 3 SE +/- 2207.23, N = 3 SE +/- 12777.82, N = 3 SE +/- 3744.59, N = 3 72193736.32 72323675.39 72276580.38 72401267.36 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Hash Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 9M 18M 27M 36M 45M SE +/- 37473.36, N = 3 SE +/- 49636.88, N = 3 SE +/- 81559.25, N = 3 SE +/- 60053.64, N = 3 42902072.99 42841367.56 42856352.41 42891632.60 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: Socket Activity Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 30K 60K 90K 120K 150K SE +/- 310.90, N = 3 SE +/- 208.73, N = 3 SE +/- 414.89, N = 3 SE +/- 940.39, N = 3 66883.60 124526.05 66923.30 112726.26 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.16.04 Test: AVL Tree Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 800 1600 2400 3200 4000 SE +/- 9.73, N = 3 SE +/- 12.35, N = 3 SE +/- 9.76, N = 3 SE +/- 21.80, N = 3 3643.25 3640.53 3649.23 3637.57 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lmpfr -lpthread -lrt -lz
LZ4 Compression This test measures the time needed to compress/decompress a sample file (silesia archive) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.4 Compression Level: 9 - Decompression Speed Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 900 1800 2700 3600 4500 SE +/- 0.90, N = 3 SE +/- 0.95, N = 3 SE +/- 0.41, N = 3 SE +/- 0.07, N = 3 4265.2 4266.4 4266.8 4270.0 1. (CC) gcc options: -O3
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.4 Compression Level: 9 - Compression Speed Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 8 16 24 32 40 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 34.68 34.65 34.68 34.61 1. (CC) gcc options: -O3
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.4 Compression Level: 3 - Decompression Speed Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 900 1800 2700 3600 4500 SE +/- 0.29, N = 3 SE +/- 0.61, N = 3 SE +/- 0.49, N = 3 SE +/- 0.78, N = 3 4073.6 4071.6 4074.1 4072.8 1. (CC) gcc options: -O3
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.4 Compression Level: 3 - Compression Speed Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 20 40 60 80 100 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.21, N = 3 SE +/- 0.09, N = 3 106.51 106.65 106.46 107.01 1. (CC) gcc options: -O3
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.4 Compression Level: 1 - Decompression Speed Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 1000 2000 3000 4000 5000 SE +/- 0.27, N = 3 SE +/- 0.83, N = 3 SE +/- 0.30, N = 3 SE +/- 0.15, N = 3 4490.4 4491.4 4491.7 4492.5 1. (CC) gcc options: -O3
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.4 Compression Level: 1 - Compression Speed Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 150 300 450 600 750 SE +/- 1.35, N = 3 SE +/- 0.52, N = 3 SE +/- 0.90, N = 3 SE +/- 0.25, N = 3 694.81 694.66 694.14 695.36 1. (CC) gcc options: -O3
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Blender 4.0 Blend File: BMW27 - Compute: CPU-Only Linux 6.8 26 Jan Linux 6.6 Linux 6.7 Linux 6.5 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 7.97 7.99 8.02 7.98
Linux 6.5 Processor: 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.5.0-060500-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 24 January 2024 18:46 by user phoronix.
Linux 6.7 Processor: 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.7.0-060700-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 25 January 2024 12:41 by user phoronix.
Linux 6.6 Processor: 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.6.0-060600-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 25 January 2024 21:35 by user phoronix.
Linux 6.8 26 Jan Processor: 2 x AMD EPYC 9684X 96-Core @ 2.55GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1007B BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3201GB Micron_7450_MTFDKCB3T2TFS, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 23.10, Kernel: 6.8.0-060800rc1daily20240126-generic (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 800x600
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,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-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-XYspKM/gcc-13-13.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113ePython Notes: Python 3.11.6Security 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 Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 26 January 2024 11:27 by user phoronix.