5950x sundat Tests fot a future article. AMD Ryzen 9 5950X 16-Core testing with a ASUS ROG CROSSHAIR VIII HERO (WI-FI) (4402 BIOS) and llvmpipe on Ubuntu 22.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2302262-NE-5950XSUND70&grs&sor .
5950x sundat Processor Motherboard Chipset Memory Disk Graphics Audio Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution a b AMD Ryzen 9 5950X 16-Core @ 3.40GHz (16 Cores / 32 Threads) ASUS ROG CROSSHAIR VIII HERO (WI-FI) (4402 BIOS) AMD Starship/Matisse 32GB 500GB Western Digital WDS500G3X0C-00SJG0 + 2000GB llvmpipe Intel Device 4f92 Realtek RTL8125 2.5GbE + Intel I211 + Intel Wi-Fi 6 AX200 Ubuntu 22.10 5.19.0-23-generic (x86_64) GNOME Shell 43.0 X Server 1.21.1.4 4.5 Mesa 22.2.1 (LLVM 15.0.2 256 bits) 1.3.224 GCC 12.2.0 ext4 3840x2160 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-U8K4Qv/gcc-12-12.2.0/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-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 Processor Details - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201025 Java Details - OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu2) Python Details - Python 3.10.7 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + 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
5950x sundat onnx: super-resolution-10 - CPU - Standard onnx: fcn-resnet101-11 - CPU - Standard onnx: yolov4 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Standard onnx: CaffeNet 12-int8 - CPU - Standard spark: 1000000 - 100 - Broadcast Inner Join Test Time aom-av1: Speed 8 Realtime - Bosphorus 4K cockroach: KV, 10% Reads - 128 spark: 10000000 - 100 - Inner Join Test Time spark: 20000000 - 100 - SHA-512 Benchmark Time spark: 40000000 - 100 - Inner Join Test Time aom-av1: Speed 9 Realtime - Bosphorus 1080p aom-av1: Speed 6 Realtime - Bosphorus 1080p cockroach: KV, 60% Reads - 1024 aom-av1: Speed 10 Realtime - Bosphorus 4K spark: 20000000 - 100 - Repartition Test Time clickhouse: 100M Rows Hits Dataset, First Run / Cold Cache spark: 10000000 - 100 - Broadcast Inner Join Test Time spark: 20000000 - 100 - Inner Join Test Time onnx: CaffeNet 12-int8 - CPU - Parallel cockroach: KV, 60% Reads - 128 compress-zstd: 3 - Decompression Speed spark: 10000000 - 100 - Group By Test Time onnx: GPT-2 - CPU - Standard build-linux-kernel: allmodconfig embree: Pathtracer - Asian Dragon compress-zstd: 3, Long Mode - Decompression Speed aom-av1: Speed 10 Realtime - Bosphorus 1080p onnx: ResNet50 v1-12-int8 - CPU - Standard build-linux-kernel: defconfig aom-av1: Speed 9 Realtime - Bosphorus 4K cockroach: KV, 50% Reads - 128 compress-zstd: 8 - Decompression Speed cockroach: KV, 95% Reads - 128 compress-zstd: 12 - Decompression Speed spark: 1000000 - 100 - Inner Join Test Time compress-zstd: 19 - Decompression Speed spark: 40000000 - 100 - Repartition Test Time cockroach: KV, 95% Reads - 1024 aom-av1: Speed 6 Two-Pass - Bosphorus 4K onnx: bertsquad-12 - CPU - Parallel spark: 1000000 - 100 - Repartition Test Time vpxenc: Speed 5 - Bosphorus 1080p spark: 1000000 - 100 - Group By Test Time compress-zstd: 3, Long Mode - Compression Speed onnx: ResNet50 v1-12-int8 - CPU - Parallel aom-av1: Speed 4 Two-Pass - Bosphorus 4K vvenc: Bosphorus 1080p - Faster vvenc: Bosphorus 1080p - Fast clickhouse: 100M Rows Hits Dataset, Third Run compress-zstd: 8, Long Mode - Compression Speed clickhouse: 100M Rows Hits Dataset, Second Run spark: 20000000 - 100 - Group By Test Time aom-av1: Speed 6 Two-Pass - Bosphorus 1080p aom-av1: Speed 0 Two-Pass - Bosphorus 1080p cockroach: MoVR - 128 compress-zstd: 19 - Compression Speed aom-av1: Speed 6 Realtime - Bosphorus 4K vpxenc: Speed 0 - Bosphorus 1080p uvg266: Bosphorus 4K - Super Fast spark: 40000000 - 100 - Calculate Pi Benchmark kvazaar: Bosphorus 4K - Medium rocksdb: Seq Fill compress-zstd: 19, Long Mode - Decompression Speed spark: 1000000 - 100 - SHA-512 Benchmark Time spark: 40000000 - 100 - Broadcast Inner Join Test Time embree: Pathtracer - Asian Dragon Obj cockroach: KV, 50% Reads - 1024 embree: Pathtracer ISPC - Asian Dragon Obj compress-zstd: 8, Long Mode - Decompression Speed rocksdb: Read While Writing vpxenc: Speed 0 - Bosphorus 4K compress-zstd: 3 - Compression Speed uvg266: Bosphorus 1080p - Slow spark: 20000000 - 100 - Calculate Pi Benchmark Using Dataframe aom-av1: Speed 8 Realtime - Bosphorus 1080p kvazaar: Bosphorus 4K - Very Fast kvazaar: Bosphorus 1080p - Very Fast openvkl: vklBenchmark Scalar aom-av1: Speed 4 Two-Pass - Bosphorus 1080p gromacs: MPI CPU - water_GMX50_bare compress-zstd: 8 - Compression Speed onnx: yolov4 - CPU - Parallel onnx: Faster R-CNN R-50-FPN-int8 - CPU - Standard spark: 1000000 - 100 - Calculate Pi Benchmark Using Dataframe embree: Pathtracer ISPC - Crown onnx: bertsquad-12 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Parallel vvenc: Bosphorus 4K - Fast vpxenc: Speed 5 - Bosphorus 4K uvg266: Bosphorus 1080p - Ultra Fast rocksdb: Update Rand uvg266: Bosphorus 1080p - Medium rocksdb: Rand Fill openvkl: vklBenchmark ISPC uvg266: Bosphorus 4K - Very Fast spark: 10000000 - 100 - SHA-512 Benchmark Time onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallel spark: 40000000 - 100 - Group By Test Time spark: 40000000 - 100 - Calculate Pi Benchmark Using Dataframe cockroach: KV, 10% Reads - 1024 spark: 10000000 - 100 - Calculate Pi Benchmark embree: Pathtracer ISPC - Asian Dragon rocksdb: Read Rand Write Rand kvazaar: Bosphorus 1080p - Super Fast kvazaar: Bosphorus 1080p - Slow vvenc: Bosphorus 4K - Faster kvazaar: Bosphorus 1080p - Ultra Fast embree: Pathtracer - Crown spark: 40000000 - 100 - SHA-512 Benchmark Time compress-zstd: 12 - Compression Speed spark: 20000000 - 100 - Calculate Pi Benchmark kvazaar: Bosphorus 1080p - Medium uvg266: Bosphorus 4K - Slow spark: 10000000 - 100 - Calculate Pi Benchmark Using Dataframe onnx: fcn-resnet101-11 - CPU - Parallel onnx: GPT-2 - CPU - Parallel uvg266: Bosphorus 4K - Medium uvg266: Bosphorus 4K - Ultra Fast rocksdb: Rand Fill Sync onnx: super-resolution-10 - CPU - Parallel uvg266: Bosphorus 1080p - Very Fast brl-cad: VGR Performance Metric spark: 10000000 - 100 - Repartition Test Time spark: 20000000 - 100 - Broadcast Inner Join Test Time rocksdb: Rand Read spark: 1000000 - 100 - Calculate Pi Benchmark kvazaar: Bosphorus 4K - Slow uvg266: Bosphorus 1080p - Super Fast kvazaar: Bosphorus 4K - Super Fast kvazaar: Bosphorus 4K - Ultra Fast cockroach: MoVR - 1024 aom-av1: Speed 0 Two-Pass - Bosphorus 4K compress-zstd: 19, Long Mode - Compression Speed onnx: Faster R-CNN R-50-FPN-int8 - CPU - Standard onnx: Faster R-CNN R-50-FPN-int8 - CPU - Parallel onnx: super-resolution-10 - CPU - Standard onnx: super-resolution-10 - CPU - Parallel onnx: ResNet50 v1-12-int8 - CPU - Standard onnx: ResNet50 v1-12-int8 - CPU - Parallel onnx: ArcFace ResNet-100 - CPU - Standard onnx: ArcFace ResNet-100 - CPU - Parallel onnx: fcn-resnet101-11 - CPU - Standard onnx: fcn-resnet101-11 - CPU - Parallel onnx: CaffeNet 12-int8 - CPU - Standard onnx: CaffeNet 12-int8 - CPU - Parallel onnx: bertsquad-12 - CPU - Standard onnx: bertsquad-12 - CPU - Parallel onnx: yolov4 - CPU - Standard onnx: yolov4 - CPU - Parallel onnx: GPT-2 - CPU - Standard onnx: GPT-2 - CPU - Parallel a b 166.401 1.46417 7.72199 26.6149 712.701 1.40 55.92 50930 11.976279517 31.530187057 45.154175361 104.28 115.45 51077.9 63.05 19.785071582 118.22 11.431434347 21.64143077 691.244 71625.1 1722.4 6.91 119.883 812.174 24.3225 1757.8 119.21 235.487 69.967 63.21 66349.7 1887.7 95581.2 1882.7 1.649008243 1701.6 40.276211423 71285.1 16.98 13.4262 1.98 36.07 3.48 722.2 233.398 9.71 25.962 11.054 134.32 429.5 130.72 10.343822865 40.1 0.91 729.5 18.5 58.91 17.25 32.98 87.075320927 15.46 1454965 1617.9 3.20 44.501980614 21.8939 49144.6 21.2632 1896.5 3594042 8.62 2781.2 41.5 5.09 123.88 34.26 135.84 142 20.32 1.27 432.2 8.37549 41.3454 5.07 22.599 16.9795 27.8609 4.477 20.56 151.22 733550 46.51 1273257 227 30.93 15.855656574 36.0437 30.310931412 5.09 41319.9 86.309113578 24.5096 2510683 175.59 62.18 9.462 218.35 23.7498 58.626368935 128 86.466437618 64.52 9.72 5.05 1.38126 107.977 10.92 38.68 27798 101.529 123.09 273690 10.05 22.020759442 94785075 86.522365613 15.16 131.42 44.11 57.01 737.4 0.31 10 24.1845 27.7424 6.00923 9.84858 4.24607 4.28386 37.5714 35.8917 682.977 723.972 1.40262 1.44574 58.8926 74.4792 129.498 119.394 8.33864 9.25704 103.546 2.13179 10.2354 32.3338 803.47 1.55 51.37 54375.8 11.267809456 29.92 43.05 99.56 110.58 53184.2 65.4 20.49 122.32 11.05 22.33 671.582 73703.7 1768.7 6.73 122.781 831.216 24.8776 1797.9 116.63 240.418 71.399 64.48 67666 1853.1 93880.3 1916.8 1.62 1672 39.59 70079.9 17.27 13.6349 1.95 36.61 3.43 732.1 236.542 9.84 26.307 11.2 136.05 434.8 132.24 10.46 40.55 0.92 737.4 18.7 59.54 17.07 32.64 86.18 15.61 1441186 1633.3 3.17 44.09 22.0976 48708.5 21.4442 1912.1 3623493 8.55 2803.5 41.17 5.05 122.97 34.51 136.82 143 20.46 1.278 429.5 8.32324 41.0978 5.04 22.7331 17.0766 27.7033 4.501 20.67 150.45 737184 46.3 1278893 228 30.8 15.92 36.1885 30.19 5.07 41158.6 86.63 24.5996 2501578 176.19 62.39 9.489 218.96 23.8103 58.764333805 127.7 86.270093043 64.66 9.74 5.06 1.38396 107.776 10.9 38.61 27749 101.354 123.24 274004 10.06 22.00 94703211 86.45 15.17 131.34 44.13 57.02 737.4 0.31 10 24.3301 27.6314 9.65702 9.86564 4.15881 4.22687 30.9262 36.0958 469.087 722.562 1.24424 1.48798 58.5583 73.3398 97.6979 120.143 8.14204 9.27425 OpenBenchmarking.org
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: super-resolution-10 - Device: CPU - Executor: Standard a b 40 80 120 160 200 166.40 103.55 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard b a 0.4797 0.9594 1.4391 1.9188 2.3985 2.13179 1.46417 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: yolov4 - Device: CPU - Executor: Standard b a 3 6 9 12 15 10.23540 7.72199 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard b a 8 16 24 32 40 32.33 26.61 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard b a 200 400 600 800 1000 803.47 712.70 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Apache Spark Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Time a b 0.3488 0.6976 1.0464 1.3952 1.744 1.40 1.55
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K a b 13 26 39 52 65 55.92 51.37 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
CockroachDB Workload: KV, 10% Reads - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 128 b a 12K 24K 36K 48K 60K 54375.8 50930.0
Apache Spark Row Count: 10000000 - Partitions: 100 - Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - Inner Join Test Time b a 3 6 9 12 15 11.27 11.98
Apache Spark Row Count: 20000000 - Partitions: 100 - SHA-512 Benchmark Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - SHA-512 Benchmark Time b a 7 14 21 28 35 29.92 31.53
Apache Spark Row Count: 40000000 - Partitions: 100 - Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Inner Join Test Time b a 10 20 30 40 50 43.05 45.15
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p a b 20 40 60 80 100 104.28 99.56 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p a b 30 60 90 120 150 115.45 110.58 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
CockroachDB Workload: KV, 60% Reads - Concurrency: 1024 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 60% Reads - Concurrency: 1024 b a 11K 22K 33K 44K 55K 53184.2 51077.9
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K b a 15 30 45 60 75 65.40 63.05 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Apache Spark Row Count: 20000000 - Partitions: 100 - Repartition Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Repartition Test Time a b 5 10 15 20 25 19.79 20.49
ClickHouse 100M Rows Hits Dataset, First Run / Cold Cache OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache b a 30 60 90 120 150 122.32 118.22 MIN: 5.84 / MAX: 6666.67 MIN: 5.57 / MAX: 7500
Apache Spark Row Count: 10000000 - Partitions: 100 - Broadcast Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - Broadcast Inner Join Test Time b a 3 6 9 12 15 11.05 11.43
Apache Spark Row Count: 20000000 - Partitions: 100 - Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Inner Join Test Time a b 5 10 15 20 25 21.64 22.33
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel a b 150 300 450 600 750 691.24 671.58 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
CockroachDB Workload: KV, 60% Reads - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 60% Reads - Concurrency: 128 b a 16K 32K 48K 64K 80K 73703.7 71625.1
Zstd Compression Compression Level: 3 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3 - Decompression Speed b a 400 800 1200 1600 2000 1768.7 1722.4 1. (CC) gcc options: -O3 -pthread -lz -llzma
Apache Spark Row Count: 10000000 - Partitions: 100 - Group By Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - Group By Test Time b a 2 4 6 8 10 6.73 6.91
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: GPT-2 - Device: CPU - Executor: Standard b a 30 60 90 120 150 122.78 119.88 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Timed Linux Kernel Compilation Build: allmodconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: allmodconfig a b 200 400 600 800 1000 812.17 831.22
Embree Binary: Pathtracer - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0 Binary: Pathtracer - Model: Asian Dragon b a 6 12 18 24 30 24.88 24.32 MIN: 24.75 / MAX: 25.43 MIN: 24.2 / MAX: 24.72
Zstd Compression Compression Level: 3, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3, Long Mode - Decompression Speed b a 400 800 1200 1600 2000 1797.9 1757.8 1. (CC) gcc options: -O3 -pthread -lz -llzma
AOM AV1 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p a b 30 60 90 120 150 119.21 116.63 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard b a 50 100 150 200 250 240.42 235.49 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Timed Linux Kernel Compilation Build: defconfig OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 6.1 Build: defconfig a b 16 32 48 64 80 69.97 71.40
AOM AV1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K b a 14 28 42 56 70 64.48 63.21 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
CockroachDB Workload: KV, 50% Reads - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 128 b a 14K 28K 42K 56K 70K 67666.0 66349.7
Zstd Compression Compression Level: 8 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8 - Decompression Speed a b 400 800 1200 1600 2000 1887.7 1853.1 1. (CC) gcc options: -O3 -pthread -lz -llzma
CockroachDB Workload: KV, 95% Reads - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 128 a b 20K 40K 60K 80K 100K 95581.2 93880.3
Zstd Compression Compression Level: 12 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Decompression Speed b a 400 800 1200 1600 2000 1916.8 1882.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
Apache Spark Row Count: 1000000 - Partitions: 100 - Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Inner Join Test Time b a 0.371 0.742 1.113 1.484 1.855 1.620000000 1.649008243
Zstd Compression Compression Level: 19 - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed a b 400 800 1200 1600 2000 1701.6 1672.0 1. (CC) gcc options: -O3 -pthread -lz -llzma
Apache Spark Row Count: 40000000 - Partitions: 100 - Repartition Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Repartition Test Time b a 9 18 27 36 45 39.59 40.28
CockroachDB Workload: KV, 95% Reads - Concurrency: 1024 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 1024 a b 15K 30K 45K 60K 75K 71285.1 70079.9
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K b a 4 8 12 16 20 17.27 16.98 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: bertsquad-12 - Device: CPU - Executor: Parallel b a 4 8 12 16 20 13.63 13.43 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Apache Spark Row Count: 1000000 - Partitions: 100 - Repartition Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Repartition Test Time b a 0.4455 0.891 1.3365 1.782 2.2275 1.95 1.98
VP9 libvpx Encoding Speed: Speed 5 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 1080p b a 8 16 24 32 40 36.61 36.07 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Apache Spark Row Count: 1000000 - Partitions: 100 - Group By Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Group By Test Time b a 0.783 1.566 2.349 3.132 3.915 3.43 3.48
Zstd Compression Compression Level: 3, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3, Long Mode - Compression Speed b a 160 320 480 640 800 732.1 722.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel b a 50 100 150 200 250 236.54 233.40 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K b a 3 6 9 12 15 9.84 9.71 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
VVenC Video Input: Bosphorus 1080p - Video Preset: Faster OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 1080p - Video Preset: Faster b a 6 12 18 24 30 26.31 25.96 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
VVenC Video Input: Bosphorus 1080p - Video Preset: Fast OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 1080p - Video Preset: Fast b a 3 6 9 12 15 11.20 11.05 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
ClickHouse 100M Rows Hits Dataset, Third Run OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run b a 30 60 90 120 150 136.05 134.32 MIN: 6.35 / MAX: 8571.43 MIN: 6.63 / MAX: 8571.43
Zstd Compression Compression Level: 8, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8, Long Mode - Compression Speed b a 90 180 270 360 450 434.8 429.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
ClickHouse 100M Rows Hits Dataset, Second Run OpenBenchmarking.org Queries Per Minute, Geo Mean, More Is Better ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run b a 30 60 90 120 150 132.24 130.72 MIN: 8.21 / MAX: 6666.67 MIN: 8.12 / MAX: 6666.67
Apache Spark Row Count: 20000000 - Partitions: 100 - Group By Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Group By Test Time a b 3 6 9 12 15 10.34 10.46
AOM AV1 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p b a 9 18 27 36 45 40.55 40.10 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p b a 0.207 0.414 0.621 0.828 1.035 0.92 0.91 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
CockroachDB Workload: MoVR - Concurrency: 128 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: MoVR - Concurrency: 128 b a 160 320 480 640 800 737.4 729.5
Zstd Compression Compression Level: 19 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed b a 5 10 15 20 25 18.7 18.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
AOM AV1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K b a 13 26 39 52 65 59.54 58.91 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
VP9 libvpx Encoding Speed: Speed 0 - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 1080p a b 4 8 12 16 20 17.25 17.07 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
uvg266 Video Input: Bosphorus 4K - Video Preset: Super Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast a b 8 16 24 32 40 32.98 32.64
Apache Spark Row Count: 40000000 - Partitions: 100 - Calculate Pi Benchmark OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Calculate Pi Benchmark b a 20 40 60 80 100 86.18 87.08
Kvazaar Video Input: Bosphorus 4K - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium b a 4 8 12 16 20 15.61 15.46 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
RocksDB Test: Sequential Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Sequential Fill a b 300K 600K 900K 1200K 1500K 1454965 1441186 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Zstd Compression Compression Level: 19, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed b a 400 800 1200 1600 2000 1633.3 1617.9 1. (CC) gcc options: -O3 -pthread -lz -llzma
Apache Spark Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Time b a 0.72 1.44 2.16 2.88 3.6 3.17 3.20
Apache Spark Row Count: 40000000 - Partitions: 100 - Broadcast Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Broadcast Inner Join Test Time b a 10 20 30 40 50 44.09 44.50
Embree Binary: Pathtracer - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0 Binary: Pathtracer - Model: Asian Dragon Obj b a 5 10 15 20 25 22.10 21.89 MIN: 21.99 / MAX: 22.41 MIN: 21.71 / MAX: 22.21
CockroachDB Workload: KV, 50% Reads - Concurrency: 1024 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 1024 a b 11K 22K 33K 44K 55K 49144.6 48708.5
Embree Binary: Pathtracer ISPC - Model: Asian Dragon Obj OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0 Binary: Pathtracer ISPC - Model: Asian Dragon Obj b a 5 10 15 20 25 21.44 21.26 MIN: 21.31 / MAX: 21.79 MIN: 21.15 / MAX: 21.83
Zstd Compression Compression Level: 8, Long Mode - Decompression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8, Long Mode - Decompression Speed b a 400 800 1200 1600 2000 1912.1 1896.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
RocksDB Test: Read While Writing OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Read While Writing b a 800K 1600K 2400K 3200K 4000K 3623493 3594042 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
VP9 libvpx Encoding Speed: Speed 0 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 0 - Input: Bosphorus 4K a b 2 4 6 8 10 8.62 8.55 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
Zstd Compression Compression Level: 3 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 3 - Compression Speed b a 600 1200 1800 2400 3000 2803.5 2781.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
uvg266 Video Input: Bosphorus 1080p - Video Preset: Slow OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Slow a b 9 18 27 36 45 41.50 41.17
Apache Spark Row Count: 20000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe b a 1.1453 2.2906 3.4359 4.5812 5.7265 5.05 5.09
AOM AV1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p a b 30 60 90 120 150 123.88 122.97 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Kvazaar Video Input: Bosphorus 4K - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Very Fast b a 8 16 24 32 40 34.51 34.26 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Very Fast b a 30 60 90 120 150 136.82 135.84 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenVKL Benchmark: vklBenchmark Scalar OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 1.3.1 Benchmark: vklBenchmark Scalar b a 30 60 90 120 150 143 142 MIN: 14 / MAX: 2679 MIN: 14 / MAX: 2649
AOM AV1 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p b a 5 10 15 20 25 20.46 20.32 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
GROMACS Implementation: MPI CPU - Input: water_GMX50_bare OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2023 Implementation: MPI CPU - Input: water_GMX50_bare b a 0.2876 0.5752 0.8628 1.1504 1.438 1.278 1.270 1. (CXX) g++ options: -O3
Zstd Compression Compression Level: 8 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 8 - Compression Speed a b 90 180 270 360 450 432.2 429.5 1. (CC) gcc options: -O3 -pthread -lz -llzma
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: yolov4 - Device: CPU - Executor: Parallel a b 2 4 6 8 10 8.37549 8.32324 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard a b 9 18 27 36 45 41.35 41.10 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Apache Spark Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe b a 1.1408 2.2816 3.4224 4.5632 5.704 5.04 5.07
Embree Binary: Pathtracer ISPC - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0 Binary: Pathtracer ISPC - Model: Crown b a 5 10 15 20 25 22.73 22.60 MIN: 22.52 / MAX: 23.11 MIN: 22.37 / MAX: 22.99
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: bertsquad-12 - Device: CPU - Executor: Standard b a 4 8 12 16 20 17.08 16.98 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel a b 7 14 21 28 35 27.86 27.70 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
VVenC Video Input: Bosphorus 4K - Video Preset: Fast OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Fast b a 1.0127 2.0254 3.0381 4.0508 5.0635 4.501 4.477 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
VP9 libvpx Encoding Speed: Speed 5 - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.13 Speed: Speed 5 - Input: Bosphorus 4K b a 5 10 15 20 25 20.67 20.56 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=gnu++11
uvg266 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast a b 30 60 90 120 150 151.22 150.45
RocksDB Test: Update Random OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Update Random b a 160K 320K 480K 640K 800K 737184 733550 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
uvg266 Video Input: Bosphorus 1080p - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Medium a b 11 22 33 44 55 46.51 46.30
RocksDB Test: Random Fill OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Fill b a 300K 600K 900K 1200K 1500K 1278893 1273257 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
OpenVKL Benchmark: vklBenchmark ISPC OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC b a 50 100 150 200 250 228 227 MIN: 30 / MAX: 2850 MIN: 30 / MAX: 2846
uvg266 Video Input: Bosphorus 4K - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast a b 7 14 21 28 35 30.93 30.80
Apache Spark Row Count: 10000000 - Partitions: 100 - SHA-512 Benchmark Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - SHA-512 Benchmark Time a b 4 8 12 16 20 15.86 15.92
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel b a 8 16 24 32 40 36.19 36.04 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Apache Spark Row Count: 40000000 - Partitions: 100 - Group By Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Group By Test Time b a 7 14 21 28 35 30.19 30.31
Apache Spark Row Count: 40000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe b a 1.1453 2.2906 3.4359 4.5812 5.7265 5.07 5.09
CockroachDB Workload: KV, 10% Reads - Concurrency: 1024 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 1024 a b 9K 18K 27K 36K 45K 41319.9 41158.6
Apache Spark Row Count: 10000000 - Partitions: 100 - Calculate Pi Benchmark OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - Calculate Pi Benchmark a b 20 40 60 80 100 86.31 86.63
Embree Binary: Pathtracer ISPC - Model: Asian Dragon OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0 Binary: Pathtracer ISPC - Model: Asian Dragon b a 6 12 18 24 30 24.60 24.51 MIN: 24.42 / MAX: 25.04 MIN: 24.39 / MAX: 24.8
RocksDB Test: Read Random Write Random OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Read Random Write Random a b 500K 1000K 1500K 2000K 2500K 2510683 2501578 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Super Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Super Fast b a 40 80 120 160 200 176.19 175.59 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Slow OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Slow b a 14 28 42 56 70 62.39 62.18 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
VVenC Video Input: Bosphorus 4K - Video Preset: Faster OpenBenchmarking.org Frames Per Second, More Is Better VVenC 1.7 Video Input: Bosphorus 4K - Video Preset: Faster b a 3 6 9 12 15 9.489 9.462 1. (CXX) g++ options: -O3 -flto=auto -fno-fat-lto-objects
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast b a 50 100 150 200 250 218.96 218.35 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Embree Binary: Pathtracer - Model: Crown OpenBenchmarking.org Frames Per Second, More Is Better Embree 4.0 Binary: Pathtracer - Model: Crown b a 6 12 18 24 30 23.81 23.75 MIN: 23.57 / MAX: 24.24 MIN: 23.53 / MAX: 24.25
Apache Spark Row Count: 40000000 - Partitions: 100 - SHA-512 Benchmark Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 40000000 - Partitions: 100 - SHA-512 Benchmark Time a b 13 26 39 52 65 58.63 58.76
Zstd Compression Compression Level: 12 - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 12 - Compression Speed a b 30 60 90 120 150 128.0 127.7 1. (CC) gcc options: -O3 -pthread -lz -llzma
Apache Spark Row Count: 20000000 - Partitions: 100 - Calculate Pi Benchmark OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Calculate Pi Benchmark b a 20 40 60 80 100 86.27 86.47
Kvazaar Video Input: Bosphorus 1080p - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 1080p - Video Preset: Medium b a 14 28 42 56 70 64.66 64.52 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
uvg266 Video Input: Bosphorus 4K - Video Preset: Slow OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow b a 3 6 9 12 15 9.74 9.72
Apache Spark Row Count: 10000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframe a b 1.1385 2.277 3.4155 4.554 5.6925 5.05 5.06
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel b a 0.3114 0.6228 0.9342 1.2456 1.557 1.38396 1.38126 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: GPT-2 - Device: CPU - Executor: Parallel a b 20 40 60 80 100 107.98 107.78 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
uvg266 Video Input: Bosphorus 4K - Video Preset: Medium OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium a b 3 6 9 12 15 10.92 10.90
uvg266 Video Input: Bosphorus 4K - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast a b 9 18 27 36 45 38.68 38.61
RocksDB Test: Random Fill Sync OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Fill Sync a b 6K 12K 18K 24K 30K 27798 27749 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inferences Per Second, More Is Better ONNX Runtime 1.14 Model: super-resolution-10 - Device: CPU - Executor: Parallel a b 20 40 60 80 100 101.53 101.35 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
uvg266 Video Input: Bosphorus 1080p - Video Preset: Very Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast b a 30 60 90 120 150 123.24 123.09
BRL-CAD VGR Performance Metric OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.34 VGR Performance Metric b a 60K 120K 180K 240K 300K 274004 273690 1. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6
Apache Spark Row Count: 10000000 - Partitions: 100 - Repartition Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 10000000 - Partitions: 100 - Repartition Test Time a b 3 6 9 12 15 10.05 10.06
Apache Spark Row Count: 20000000 - Partitions: 100 - Broadcast Inner Join Test Time OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 20000000 - Partitions: 100 - Broadcast Inner Join Test Time b a 5 10 15 20 25 22.00 22.02
RocksDB Test: Random Read OpenBenchmarking.org Op/s, More Is Better RocksDB 7.9.2 Test: Random Read a b 20M 40M 60M 80M 100M 94785075 94703211 1. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread
Apache Spark Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark OpenBenchmarking.org Seconds, Fewer Is Better Apache Spark 3.3 Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark b a 20 40 60 80 100 86.45 86.52
Kvazaar Video Input: Bosphorus 4K - Video Preset: Slow OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Slow b a 4 8 12 16 20 15.17 15.16 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
uvg266 Video Input: Bosphorus 1080p - Video Preset: Super Fast OpenBenchmarking.org Frames Per Second, More Is Better uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast a b 30 60 90 120 150 131.42 131.34
Kvazaar Video Input: Bosphorus 4K - Video Preset: Super Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Super Fast b a 10 20 30 40 50 44.13 44.11 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
Kvazaar Video Input: Bosphorus 4K - Video Preset: Ultra Fast OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Ultra Fast b a 13 26 39 52 65 57.02 57.01 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
CockroachDB Workload: MoVR - Concurrency: 1024 OpenBenchmarking.org ops/s, More Is Better CockroachDB 22.2 Workload: MoVR - Concurrency: 1024 b a 160 320 480 640 800 737.4 737.4
AOM AV1 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.6 Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4K b a 0.0698 0.1396 0.2094 0.2792 0.349 0.31 0.31 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Zstd Compression Compression Level: 19, Long Mode - Compression Speed OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed b a 3 6 9 12 15 10 10 1. (CC) gcc options: -O3 -pthread -lz -llzma
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard a b 6 12 18 24 30 24.18 24.33 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel b a 7 14 21 28 35 27.63 27.74 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: super-resolution-10 - Device: CPU - Executor: Standard a b 3 6 9 12 15 6.00923 9.65702 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: super-resolution-10 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: super-resolution-10 - Device: CPU - Executor: Parallel a b 3 6 9 12 15 9.84858 9.86564 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Standard b a 0.9554 1.9108 2.8662 3.8216 4.777 4.15881 4.24607 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel b a 0.9639 1.9278 2.8917 3.8556 4.8195 4.22687 4.28386 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard b a 9 18 27 36 45 30.93 37.57 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel a b 8 16 24 32 40 35.89 36.10 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard b a 150 300 450 600 750 469.09 682.98 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel b a 160 320 480 640 800 722.56 723.97 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard b a 0.3156 0.6312 0.9468 1.2624 1.578 1.24424 1.40262 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: CaffeNet 12-int8 - Device: CPU - Executor: Parallel a b 0.3348 0.6696 1.0044 1.3392 1.674 1.44574 1.48798 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: bertsquad-12 - Device: CPU - Executor: Standard b a 13 26 39 52 65 58.56 58.89 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: bertsquad-12 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: bertsquad-12 - Device: CPU - Executor: Parallel b a 20 40 60 80 100 73.34 74.48 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: yolov4 - Device: CPU - Executor: Standard b a 30 60 90 120 150 97.70 129.50 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: yolov4 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: yolov4 - Device: CPU - Executor: Parallel a b 30 60 90 120 150 119.39 120.14 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Standard OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: GPT-2 - Device: CPU - Executor: Standard b a 2 4 6 8 10 8.14204 8.33864 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
ONNX Runtime Model: GPT-2 - Device: CPU - Executor: Parallel OpenBenchmarking.org Inference Time Cost (ms), Fewer Is Better ONNX Runtime 1.14 Model: GPT-2 - Device: CPU - Executor: Parallel a b 3 6 9 12 15 9.25704 9.27425 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto=auto -fno-fat-lto-objects -ldl -lrt
Phoronix Test Suite v10.8.4