Extra EPYC Rome Benchmarks AMD Ryzen 9 3900X 12-Core testing with a Gigabyte X570 AORUS MASTER (F5k BIOS) and Gigabyte NVIDIA GeForce RTX 2080 Ti Rev. A 11GB on Fedora 30 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1908150-SP-1908130AS62&grr .
Extra EPYC Rome Benchmarks Processor Motherboard Chipset Memory Disk Graphics Monitor Network Audio OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2 x AMD EPYC 7502 32-Core @ 2.50GHz (64 Cores / 128 Threads) AMD DAYTONA_X (RDY1001C BIOS) AMD Device 1480 516096MB 280GB INTEL SSDPE21D280GA + 280GB INTEL SSDPED1D280GA + 5 x 3841GB Micron_9300_MTFDHAL3T8TDP + 256GB Micron_1100_MTFD llvmpipe 504GB VE228 2 x Mellanox MT27710 Ubuntu 19.04 5.2.0-050200rc7-generic (x86_64) 20190630 GNOME Shell 3.32.1 X Server 1.20.4 modesetting 1.20.4 3.3 Mesa 19.0.2 (LLVM 8.0 128 bits) GCC 8.3.0 ext4 1920x1080 2 x AMD EPYC 7742 64-Core @ 2.25GHz (128 Cores / 256 Threads) ASPEED 5.3.0-rc1-power (x86_64) 20190728 2 x Intel Xeon Platinum 8280 @ 4.00GHz (56 Cores / 112 Threads) GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS) Intel Sky Lake-E DMI3 Registers 386048MB 280GB INTEL SSDPED1D280GA + 32GB Flash Disk llvmpipe 377GB 2 x Intel X722 for 1GbE + 2 x QLogic FastLinQ QL41000 10/25/40/50GbE 5.2.0-050200rc7-generic (x86_64) 20190630 GNOME Shell 3.32.2 3.3 Mesa 19.0.2 (LLVM 8.0 256 bits) AMD Ryzen 9 3900X 12-Core @ 3.80GHz (12 Cores / 24 Threads) Gigabyte X570 AORUS MASTER (F5k BIOS) AMD Starship/Matisse 16384MB 2 x 2000GB GIGABYTE GP-ASM2NE6200TTTD Gigabyte NVIDIA GeForce RTX 2080 Ti Rev. A 11GB NVIDIA TU102 HD Audio BenQ EL2870U Intel I211 + Realtek Device 8125 + Intel Wi-Fi 6 AX200 Fedora 30 5.2.7-200.fc30.x86_64 (x86_64) X Server GCC 9.1.1 20190503 3840x2160 OpenBenchmarking.org Compiler Details - 2 x EPYC 7502: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v - 2 x EPYC 7742: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v - 2 x Xeon Platinum 8280: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --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 --with-tune=generic --without-cuda-driver -v - Ryzen 3900X: --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-initfini-array --enable-languages=c,c++,fortran,objc,obj-c++,ada,go,d,lto --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-gcc-major-version-only --with-isl --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver Disk Details - 2 x EPYC 7502: NONE / errors=remount-ro,relatime,rw - 2 x EPYC 7742: NONE / errors=remount-ro,relatime,rw - 2 x Xeon Platinum 8280: NONE / errors=remount-ro,relatime,rw - Ryzen 3900X: NONE / relatime,rw,seclabel,stripe=256 / raid0 nvme1n1p3[0] nvme0n1p2[1] Processor Details - 2 x EPYC 7502: Scaling Governor: acpi-cpufreq ondemand - 2 x EPYC 7742: Scaling Governor: acpi-cpufreq ondemand - 2 x Xeon Platinum 8280: Scaling Governor: intel_pstate powersave - Ryzen 3900X: Scaling Governor: acpi-cpufreq ondemand Java Details - 2 x EPYC 7502: OpenJDK Runtime Environment (build 11.0.3+7-Ubuntu-1ubuntu219.04.1) - 2 x EPYC 7742: OpenJDK Runtime Environment (build 11.0.3+7-Ubuntu-1ubuntu219.04.1) - 2 x Xeon Platinum 8280: OpenJDK Runtime Environment (build 11.0.4+11-post-Ubuntu-1ubuntu219.04) Python Details - 2 x EPYC 7502: Python 2.7.16 + Python 3.7.3 - 2 x EPYC 7742: Python 2.7.16 + Python 3.7.3 - 2 x Xeon Platinum 8280: Python 2.7.16 + Python 3.7.3 - Ryzen 3900X: Python 2.7.16 + Python 3.7.4 Security Details - 2 x EPYC 7502: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling - 2 x EPYC 7742: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling - 2 x Xeon Platinum 8280: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling - Ryzen 3900X: SELinux + usercopy/swapgs barriers and __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: always-on RSB filling + SSB disabled via prctl and seccomp
Extra EPYC Rome Benchmarks hpcc: G-HPL spec-jbb2015: SPECjbb2015-Composite critical-jOPS spec-jbb2015: SPECjbb2015-Composite max-jOPS mysqlslap: 256 mkl-dnn: Convolution Batch conv_all - u8s8u8s32 mkl-dnn: Convolution Batch conv_all - u8s8f32s32 mkl-dnn: Convolution Batch conv_all - f32 mkl-dnn: Deconvolution Batch deconv_all - u8s8u8s32 numpy: mbw: Memory Copy, Fixed Block Size - 8192 MiB build-gcc: Time To Compile pts-self-test: Total Time mkl-dnn: Deconvolution Batch deconv_all - f32 mbw: Memory Copy - 8192 MiB parboil: OpenMP MRI Gridding renaissance: Apache Spark ALS dacapobench: Tradesoap mysqlslap: 128 blogbench: Read fftw: Float + SSE - 2D FFT Size 4096 renaissance: Savina Reactors.IO hpcg: cp2k: Fayalite-FIST Data mkl-dnn: Convolution Batch conv_googlenet_v3 - f32 pgbench: Buffer Test - Normal Load - Read Write mkl-dnn: Convolution Batch conv_3d - u8s8u8s32 mkl-dnn: Convolution Batch conv_3d - u8s8f32s32 blender: Barbershop - CPU-Only tinymembench: Standard Memset tinymembench: Standard Memcpy renaissance: Apache Spark PageRank mysqlslap: 64 renaissance: Akka Unbalanced Cobwebbed Tree npb: SP.A fftw: Stock - 2D FFT Size 4096 npb: BT.A pgbench: Buffer Test - Single Thread - Read Only pgbench: Buffer Test - Single Thread - Read Write pgbench: Buffer Test - Normal Load - Read Only mysqlslap: 32 mkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8u8s32 mkl-dnn: Convolution Batch conv_googlenet_v3 - u8s8f32s32 appleseed: Material Tester gmpbench: Total Time mysqlslap: 16 asmfish: 1024 Hash Memory, 26 Depth vpxenc: vpxenc VP9 1080p Video Encode appleseed: Emily blender: Classroom - CPU-Only blender: Barbershop - CPU-Only mkl-dnn: Deconvolution Batch deconv_3d - u8s8f32s32 mkl-dnn: Deconvolution Batch deconv_3d - u8s8u8s32 parboil: OpenMP LBM minion: Quasigroup blender: Pabellon Barcelona - CPU-Only stress-ng: Context Switching namd: ATPase Simulation - 327,506 Atoms fftw: Float + SSE - 2D FFT Size 2048 mysqlslap: 8 clomp: Static OMP Speedup mysqlslap: 4 mkl-dnn: Convolution Batch conv_3d - f32 mkl-dnn: Deconvolution Batch deconv_1d - u8s8f32s32 renaissance: Apache Spark Bayes mkl-dnn: Deconvolution Batch deconv_1d - u8s8u8s32 dacapobench: Tradebeans apache-siege: 250 build-linux-kernel: Time To Compile blender: Fishy Cat - CPU-Only minion: Solitaire fftw: Stock - 2D FFT Size 2048 mkl-dnn: Convolution Batch conv_alexnet - u8s8u8s32 stress-ng: System V Message Passing stress-ng: Glibc C String Functions apache-siege: 200 v-ray: CPU renaissance: In-Memory Database Shootout mkl-dnn: Convolution Batch conv_alexnet - u8s8f32s32 appleseed: Disney Material blender: Pabellon Barcelona - CPU-Only java-scimark2: Composite indigobench: Bedroom renaissance: Scala Dotty npb: LU.C mcperf: Delete indigobench: Supercar sockperf: Latency Under Load stockfish: Total Time mcperf: Set ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping mcperf: Add mbw: Memory Copy, Fixed Block Size - 4096 MiB blender: Classroom - CPU-Only compress-7zip: Compress Speed Test himeno: Poisson Pressure Solver mcperf: Prepend minion: Graceful mcperf: Get blender: BMW27 - CPU-Only redis: LPUSH build-llvm: Time To Compile mkl-dnn: IP Batch All - u8s8f32s32 mkl-dnn: IP Batch All - u8s8u8s32 build-php: Time To Compile mkl-dnn: IP Batch All - f32 stream: Copy redis: LPOP lzbench: XZ 0 - Decompression lzbench: XZ 0 - Compression mbw: Memory Copy - 4096 MiB dcraw: RAW To PPM Image Conversion stress-ng: Semaphores redis: GET redis: SADD blender: Fishy Cat - CPU-Only apache-siege: 100 gimp: auto-levels redis: SET lzbench: Libdeflate 1 - Decompression lzbench: Libdeflate 1 - Compression mkl-dnn: IP Batch 1D - u8s8u8s32 lzbench: Zstd 1 - Decompression lzbench: Zstd 1 - Compression lzbench: Brotli 0 - Decompression lzbench: Brotli 0 - Compression tesseract-ocr: Time To OCR 7 Images npb: FT.B stress-ng: Forking sunflow: Global Illumination + Image Synthesis john-the-ripper: Blowfish sockperf: Throughput stress-ng: CPU Stress stress-ng: Tsearch stress-ng: Memory Copying stress-ng: Lsearch stress-ng: Crypto stress-ng: Glibc Qsort Data Sorting stress-ng: Bsearch stress-ng: Socket Activity stress-ng: Hsearch stress-ng: Matrix Math stress-ng: Vector Math mcperf: Replace gimp: unsharp-mask encode-mp3: WAV To MP3 rodinia: OpenMP Streamcluster coremark: CoreMark Size 666 - Iterations Per Second mcperf: Append fftw: Float + SSE - 2D FFT Size 1024 blender: BMW27 - CPU-Only cython-bench: rodinia: OpenMP CFD Solver mkl-dnn: IP Batch 1D - u8s8f32s32 apache-siege: 50 m-queens: Time To Solve compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 botan: AES-256 - Decrypt botan: AES-256 - Encrypt c-ray: Total Time - 4K, 16 Rays Per Pixel rodinia: OpenMP LavaMD mkl-dnn: Deconvolution Batch deconv_1d - f32 build-imagemagick: Time To Compile botan: Twofish - Decrypt botan: Twofish - Encrypt botan: Blowfish - Decrypt botan: Blowfish - Encrypt botan: CAST-256 - Decrypt botan: CAST-256 - Encrypt openssl: RSA 4096-bit Performance botan: KASUMI - Decrypt botan: KASUMI - Encrypt povray: Trace Time darktable: Server Room - CPU-only ramspeed: Copy - Floating Point ramspeed: Triad - Floating Point ramspeed: Scale - Floating Point dav1d: Summer Nature 4K dacapobench: H2 mencoder: AVI To LAVC gimp: rotate ramspeed: Add - Floating Point mkl-dnn: Convolution Batch conv_alexnet - f32 ramspeed: Average - Floating Point ramspeed: Triad - Integer ramspeed: Copy - Integer ramspeed: Add - Integer ramspeed: Scale - Integer ramspeed: Average - Integer phpbench: PHP Benchmark Suite x265: H.265 1080p Video Encoding compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 mkl-dnn: IP Batch 1D - f32 y-cruncher: Calculating 500M Pi Digits encode-flac: WAV To FLAC x264: H.264 Video Encoding mbw: Memory Copy, Fixed Block Size - 1024 MiB sockperf: Latency Ping Pong npb: LU.A parboil: OpenMP Stencil fftw: Stock - 2D FFT Size 1024 gimp: resize mbw: Memory Copy - 1024 MiB mafft: Multiple Sequence Alignment bork: File Encryption Time osbench: Create Threads apache-siege: 10 dacapobench: Jython fftw: Float + SSE - 1D FFT Size 64 osbench: Launch Programs primesieve: 1e12 Prime Number Generation svt-av1: 1080p 8-bit YUV To AV1 Video Encode fftw: Float + SSE - 1D FFT Size 256 dav1d: Summer Nature 1080p fftw: Float + SSE - 1D FFT Size 4096 fftw: Float + SSE - 2D FFT Size 512 mbw: Memory Copy, Fixed Block Size - 512 MiB encode-ogg: WAV To Ogg tjbench: Decompression Throughput fftw: Float + SSE - 1D FFT Size 128 fftw: Stock - 1D FFT Size 4096 fftw: Stock - 1D FFT Size 32 darktable: Masskrug - CPU-only osbench: Create Files fftw: Float + SSE - 1D FFT Size 2048 mbw: Memory Copy - 512 MiB fftw: Float + SSE - 1D FFT Size 1024 fftw: Float + SSE - 2D FFT Size 256 npb: EP.C osbench: Memory Allocations osbench: Create Processes svt-vp9: 1080p 8-bit YUV To VP9 Video Encode fftw: Stock - 2D FFT Size 512 fftw: Stock - 1D FFT Size 2048 darktable: Boat - CPU-only npb: FT.A fftw: Float + SSE - 1D FFT Size 512 darktable: Server Rack - CPU-only fftw: Stock - 2D FFT Size 256 fftw: Stock - 1D FFT Size 1024 svt-hevc: 1080p 8-bit YUV To HEVC Video Encode fftw: Float + SSE - 2D FFT Size 128 fftw: Stock - 2D FFT Size 64 fftw: Stock - 2D FFT Size 128 fftw: Float + SSE - 1D FFT Size 32 fftw: Stock - 1D FFT Size 64 mkl-dnn: Deconvolution Batch deconv_3d - f32 fftw: Stock - 1D FFT Size 512 fftw: Stock - 1D FFT Size 256 apache-siege: 1 fftw: Float + SSE - 2D FFT Size 32 fftw: Float + SSE - 2D FFT Size 64 ffte: N=256, 1D Complex FFT Routine php: Zend micro_bench fftw: Stock - 2D FFT Size 32 fftw: Stock - 1D FFT Size 128 parboil: OpenMP CUTCP tscp: AI Chess Performance mbw: Memory Copy, Fixed Block Size - 128 MiB mbw: Memory Copy - 128 MiB ctx-clock: Context Switch Time blake2: php: Zend bench pts-self-test: Info Command pts-self-test: Peak Memory Usage pts-self-test: Merge Results pts-self-test: Render Test pts-self-test: List Suites java-scimark2: Jacobi Successive Over-Relaxation java-scimark2: Dense LU Matrix Factorization java-scimark2: Sparse Matrix Multiply java-scimark2: Fast Fourier Transform java-scimark2: Monte Carlo hpcc: Max Ping Pong Bandwidth hpcc: Rand Ring Bandwidth hpcc: Rand Ring Latency hpcc: G-Rand Access hpcc: EP-STREAM Triad hpcc: G-Ptrans hpcc: EP-DGEMM hpcc: G-Ffte hpcc: G-Ffte stream: Add stream: Triad stream: Scale blogbench: Write 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 169.92267 60218 106645 270 20800 20778 621 15307 7216577 9042 746 253 4229 15675 120.92 38778 5601 342 1272056 17004 28404 0.46 580 34.66 21936 2438 2447 236.53 15240 9148 24726 835 30304 396.26 5116 834.43 20388 2689 858392 1757 1752.47 1764.58 284 4519 3713 136257996 145.29 173 122.64 221 1377 1507 25.00 137.89 100.13 62075360 0.41349 26109 6397 0.37 7634 3.67 795.68 3595 881.33 7419 30037 19.74 70.16 82.07 5815 947.22 12216045 4908942 28569 62297 13683 941.69 59.80 102 2389 4.78 7326 33940.81 40803 7.72 22.42 135223710 33580 693 33157 9017 80.60 255492 1004 28279 52.78 53650 47.32 1375853 96.43 778.49 772.47 49.14 81.93 150788 2170359 87 28 15472 51.43 1398972 1897043 1800546 66.52 26655 24.62 1463246 841 190 68.47 932 354 441 383 34.90 3660.37 43021 0.74 109363 484363 19073 1752 9823 708 14306 931 33173 23811 388396 375525 130505 28323 30.60 33.17 13.57 2101299 34850 33251 45.20 28.68 8.87 68.28 21692 12.10 26.97 4640 4585 10.90 6.11 8.11 21.39 299 299 228 231 120 120 14388 75.24 77.84 10.32 3.03 30275 32247 40026 13.24 6911 20.20 16.75 48547 80.93 44275 40110 42739 47552 38465 42023 494910 43.46 11.20 9.47 8.89 10.09 148.94 9093 4.85 31119.46 5.64 6099 11.67 15711 3.42 10.77 33.79 19908 5466 16447 78.17 4.29 94.99 27950 5.69 41618 33879 9126 6.81 169 20960 6709 8741 4.24 15.23 44915 15671 43078 33586 2283.03 96.59 63.31 197.26 6633 6858 3.30 3332.60 37722 0.11 6615 7129 340.94 34549 7300 6788 11848 8082 3.86 7052 7127 4559.86 34572 35897 7639.45 2.66 8824 6766 1.04 1036078 8984 8986 150 7.46 0.54 3.08 503 4.84 69.67 0.74 1506 5270 2211 1624 1334 10781 1.38385 1.77976 0.20233 2.80207 1.36882 22.38970 33.30483 33.30483 164105 166119 155497 43807 9051 15556 182.83 37539 1164067 17865 31115 0.35 611 16031 9131 24412 5214 4589 35.97 0.26589 26661 0.23 4126 7743 5993 2423 7276 22.16 9032 185818 88 28 15436 867 191 961 360 446 383 454195 10.43 33895 9.10 4.33 35526 46521 36196 8226 48530 40960 43023 41105 48253 37700 42272 9153 5.30 4.02 6146 15602 3.41 10.73 5264 16032 27860 43038 34178 9188 21059 6814 8868 45337 15464 43869 34284 6717 7006 38922 6744 7207 35003 7400 6948 11985 8203 7124 7217 35028 37011 8982 6899 0.80 9053 9066 6.60 1529 5364 2246 1622 1356 194680 200966 177029 38226 146.45033 50770 93995 438 1722 1756 382 3712 5262216 5135 892 215 1055 4747 513.31 49161 7404 544 1995239 16099 26058 0.94 1699 21.53 23703 3576 3538 240.94 10920 6933 23691 1286 24373 379.98 4640 834.30 19016 2656 912806 2905 5.77 6.23 185 5098 6347 137371669 94.82 155 135.02 223 1890 2041 22.43 111.21 112.00 26550213 0.35923 21198 10422 0.43 13278 3.30 0.23 3283 0.23 11780 34608 21.45 76.82 84.34 5090 14.23 6551714 4483653 33432 65476 17338 18.95 67.24 117 2182 6.14 6930 32373.02 113029 9.88 35.39 125947505 62021 1010 61510 5053 94.29 232626 2794 70628 53.28 111930 51.15 1656464 107.93 20.48 20.42 48.59 100.80 168490 2475214 95 36 4752 41.77 2098939 2299349 2122727 72.54 33242 22.79 1752860 1016 213 3.64 967 375 519 391 32.79 2556.20 68990 0.86 83578 448019 17955 1091 11263 665 10503 787 35116 37540 301308 380876 120957 71185 28.18 29.97 11.79 1890649 70478 22351 47.45 22.97 7.31 3.55 27915 16.30 22.50 3828 3823 14.40 6.96 0.95 19.67 351 347 275 282 137 137 13753 87.03 89.63 11.57 1.92 32110 28432 25822 18.25 8773 17.69 18.25 28400 48.57 29060 31864 32285 32779 29096 31645 646738 32.73 9.01 11.56 9.12 9.56 120.04 4916 3.98 30471.20 2.60 6340 12.48 4775 5.16 7.10 20.51 26012 4629 16984 54.51 4.82 61.53 36704 9.20 58222 22289 4910 6.12 188 20343 7274 7179 4.77 13.77 63908 4819 64732 24853 2486.05 87.57 45.92 271.63 6355 7392 4.05 2378.80 55983 0.10 6386 7555 260.31 28613 7181 7470 14317 7404 1.11 7793 7200 2606.48 27128 33648 2.06 8898 7024 1.29 1257489 4918 4975 136 3.87 0.45 2.40 503 3.91 59.80 0.71 1585 4793 2156 1392 984 17274 0.83730 1.13356 0.17540 3.00269 15.01817 25.17353 16.90603 16.90603 156319 154125 149490 53516 12.79480 206 294635 293148 18504 202139 260 668 697 31340 443 34.16 4886 589 1298958 20418 12205 321 1021.92 17656 106118 106123 13889 9630 27352 804 12008 6556 33085 1150 302301 914 11526.10 11596.50 161 5922 946 40037020 270 707 59364 59053 168.10 98.29 5721856 1.44442 21767 986 2.45 953 101.70 35857.17 8382 35756.50 24626 47.79 56.55 6573 27767.67 4645426 1262789 24906 5735 27648.10 164.44 370 2924 4891 77722 18.51 40049521 58184 656 46988 10170 320.26 77585 1278 52587 78724 1738434 245.50 2276.91 2265.86 44.68 791.93 14931 2727208 18783 31.15 5724009 2538538 2169218 194 24450 1792791 177.05 63865 1.10 20460 373777 4440 475 9153 178 3520 206 8342 4183 97237 77603 33507 46395 24.38 23.71 579446 46554 41799 138.06 13.14 176.21 28889 46.75 25.32 5896 5910 42.22 20.08 197.67 19.95 391 396 297 299 164 164 3531 101.82 106.29 31.55 23394 27459 23339 22.26 14.92 26957 2393.04 27314 29597 23424 27016 23852 27333 698254 18.96 64.09 21.92 7.53 9963 4.84 16.90 7909 18599 2.08 12.08 14.68 20400 22302 37.85 15.47 39978 7.49 58217 42671 9896 233 29016 8713 11691 17.06 60606 18439 58387 44203 68.01 27.63 9012 8998 56806 8683 9726 48512 9601 8988 15955 8701 38.14 9288 8981 48195 49146 1.80 11466 8996 2.07 1376962 9393 18521 152 8.30 0.40 1.20 500 74.28 35.29 20.17 1716 6944 2674 1280 1716 25023 5.25031 0.59617 0.03249 2.07594 2.49771 1.07655 5.40971 5.40971 16284 16305 15170 32787 OpenBenchmarking.org
HPC Challenge Test / Class: G-HPL OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-HPL 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 169.92 146.45 12.79 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
SPECjbb 2015 SPECjbb2015-Composite critical-jOPS OpenBenchmarking.org jOPS, More Is Better SPECjbb 2015 SPECjbb2015-Composite critical-jOPS 2 x EPYC 7502 2 x Xeon Platinum 8280 13K 26K 39K 52K 65K 60218 50770
SPECjbb 2015 SPECjbb2015-Composite max-jOPS OpenBenchmarking.org jOPS, More Is Better SPECjbb 2015 SPECjbb2015-Composite max-jOPS 2 x EPYC 7502 2 x Xeon Platinum 8280 20K 40K 60K 80K 100K 106645 93995
MariaDB Clients: 256 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 256 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 90 180 270 360 450 SE +/- 1.96, N = 3 SE +/- 0.73, N = 3 SE +/- 5.52, N = 9 270 438 206 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
MKL-DNN Harness: Convolution Batch conv_all - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 60K 120K 180K 240K 300K SE +/- 105.95, N = 3 SE +/- 7.41, N = 3 SE +/- 332.06, N = 3 20800 1722 294635 -O2 - MIN: 293228 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_all - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 60K 120K 180K 240K 300K SE +/- 183.52, N = 3 SE +/- 6.52, N = 3 SE +/- 402.11, N = 3 20778 1756 293148 -O2 - MIN: 292073 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_all - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_all - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 8.02, N = 9 SE +/- 0.44, N = 3 SE +/- 0.66, N = 3 621 382 18504 -O2 - MIN: 18471.7 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_all - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_all - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40K 80K 120K 160K 200K SE +/- 81.72, N = 3 SE +/- 45.90, N = 9 SE +/- 43.36, N = 3 15307 3712 202139 -O2 - MIN: 201928 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Numpy Benchmark OpenBenchmarking.org Nanoseconds, Fewer Is Better Numpy Benchmark 2 x EPYC 7502 2 x Xeon Platinum 8280 1.5M 3M 4.5M 6M 7.5M 7216577 5262216
MBW Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 19.23, N = 3 SE +/- 16.13, N = 3 SE +/- 9.79, N = 3 SE +/- 0.76, N = 3 9042 9051 5135 260 1. (CC) gcc options: -O3 -march=native
Timed GCC Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GCC Compilation 8.2 Time To Compile 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 200 400 600 800 1000 SE +/- 0.64, N = 3 SE +/- 0.37, N = 3 SE +/- 0.57, N = 3 746 892 668
Phoronix Test Suite Self Test Test: Total Time OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Total Time 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 150 300 450 600 750 SE +/- 0.68, N = 3 SE +/- 1.33, N = 3 SE +/- 0.44, N = 2 253 215 697
MKL-DNN Harness: Deconvolution Batch deconv_all - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_all - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 7K 14K 21K 28K 35K SE +/- 89.09, N = 9 SE +/- 6.16, N = 3 SE +/- 58.11, N = 3 4229 1055 31340 -O3 - MIN: 3301.92 -O2 - MIN: 30877.1 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MBW Test: Memory Copy - Array Size: 8192 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 51.91, N = 3 SE +/- 38.55, N = 3 SE +/- 0.90, N = 3 SE +/- 3.95, N = 3 15675 15556 4747 443 1. (CC) gcc options: -O3 -march=native
Parboil Test: OpenMP MRI Gridding OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 110 220 330 440 550 SE +/- 0.54, N = 3 SE +/- 2.09, N = 6 SE +/- 7.08, N = 9 SE +/- 0.01, N = 3 120.92 182.83 513.31 34.16 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Renaissance Test: Apache Spark ALS OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Apache Spark ALS 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 11K 22K 33K 44K 55K SE +/- 351.03, N = 40 SE +/- 1216.19, N = 40 SE +/- 325.62, N = 40 SE +/- 32.09, N = 32 38778 37539 49161 4886
DaCapo Benchmark Java Test: Tradesoap OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap 2 x EPYC 7502 2 x Xeon Platinum 8280 1600 3200 4800 6400 8000 SE +/- 41.94, N = 20 SE +/- 75.42, N = 20 5601 7404
MariaDB Clients: 128 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 128 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 130 260 390 520 650 SE +/- 3.87, N = 3 SE +/- 0.98, N = 3 SE +/- 0.83, N = 3 342 544 589 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
BlogBench Test: Read OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 400K 800K 1200K 1600K 2000K SE +/- 4626.16, N = 3 SE +/- 14055.95, N = 3 SE +/- 20800.00, N = 3 SE +/- 3745.14, N = 3 1272056 1164067 1995239 1298958 1. (CC) gcc options: -O2 -pthread
FFTW Build: Float + SSE - Size: 2D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 214.67, N = 3 SE +/- 63.98, N = 3 SE +/- 82.58, N = 3 SE +/- 251.60, N = 3 17004 17865 16099 20418 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Renaissance Test: Savina Reactors.IO OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Savina Reactors.IO 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 7K 14K 21K 28K 35K SE +/- 733.25, N = 40 SE +/- 831.26, N = 40 SE +/- 297.31, N = 40 SE +/- 144.45, N = 8 28404 31115 26058 12205
High Performance Conjugate Gradient OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.0 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 0.2115 0.423 0.6345 0.846 1.0575 SE +/- 0.01, N = 12 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 0.46 0.35 0.94
CP2K Molecular Dynamics Fayalite-FIST Data OpenBenchmarking.org Seconds, Fewer Is Better CP2K Molecular Dynamics 6.1 Fayalite-FIST Data 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 400 800 1200 1600 2000 580 611 1699 321
MKL-DNN Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 200 400 600 800 1000 SE +/- 0.43, N = 12 SE +/- 0.20, N = 3 SE +/- 0.10, N = 3 34.66 21.53 1021.92 -O2 - MIN: 1017.72 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5K 10K 15K 20K 25K SE +/- 151.66, N = 3 SE +/- 98.31, N = 3 SE +/- 1091.07, N = 12 21936 23703 17656 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
MKL-DNN Harness: Convolution Batch conv_3d - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20K 40K 60K 80K 100K SE +/- 3.79, N = 3 SE +/- 35.12, N = 3 SE +/- 29.28, N = 3 2438 3576 106118 -O2 - MIN: 106044 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_3d - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20K 40K 60K 80K 100K SE +/- 2.13, N = 3 SE +/- 2.31, N = 3 SE +/- 50.84, N = 3 2447 3538 106123 -O2 - MIN: 105997 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79 Blend File: Barbershop - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 50 100 150 200 250 SE +/- 0.50, N = 3 SE +/- 0.08, N = 3 236.53 240.94
Tinymembench Standard Memset OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 16.26, N = 3 SE +/- 17.98, N = 3 SE +/- 67.15, N = 3 SE +/- 28.57, N = 3 15240 16031 10920 13889 1. (CC) gcc options: -O2 -lm
Tinymembench Standard Memcpy OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 4.52, N = 3 SE +/- 4.70, N = 3 SE +/- 51.26, N = 3 SE +/- 5.75, N = 3 9148 9131 6933 9630 1. (CC) gcc options: -O2 -lm
Renaissance Test: Apache Spark PageRank OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Apache Spark PageRank 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 294.87, N = 32 SE +/- 224.85, N = 40 SE +/- 200.36, N = 8 SE +/- 234.52, N = 8 24726 24412 23691 27352
MariaDB Clients: 64 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 64 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 300 600 900 1200 1500 SE +/- 0.93, N = 3 SE +/- 18.28, N = 3 SE +/- 0.99, N = 3 835 1286 804 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
Renaissance Test: Akka Unbalanced Cobwebbed Tree OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Akka Unbalanced Cobwebbed Tree 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 268.51, N = 32 SE +/- 482.12, N = 32 SE +/- 71.27, N = 8 30304 24373 12008
NAS Parallel Benchmarks Test / Class: SP.A OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: SP.A 2 x EPYC 7502 2 x Xeon Platinum 8280 90 180 270 360 450 SE +/- 0.46, N = 3 SE +/- 0.55, N = 3 396.26 379.98 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
FFTW Build: Stock - Size: 2D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1400 2800 4200 5600 7000 SE +/- 2.50, N = 3 SE +/- 10.42, N = 3 SE +/- 31.33, N = 3 SE +/- 18.60, N = 3 5116 5214 4640 6556 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
NAS Parallel Benchmarks Test / Class: BT.A OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: BT.A 2 x EPYC 7502 2 x Xeon Platinum 8280 200 400 600 800 1000 SE +/- 0.50, N = 3 SE +/- 1.67, N = 3 834.43 834.30 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
PostgreSQL pgbench Scaling: Buffer Test - Test: Single Thread - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 7K 14K 21K 28K 35K SE +/- 102.32, N = 3 SE +/- 48.37, N = 3 SE +/- 270.22, N = 3 20388 19016 33085 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Single Thread - Mode: Read Write OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Single Thread - Mode: Read Write 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 600 1200 1800 2400 3000 SE +/- 14.28, N = 3 SE +/- 39.04, N = 3 SE +/- 5.82, N = 3 2689 2656 1150 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Only OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 200K 400K 600K 800K 1000K SE +/- 8471.64, N = 3 SE +/- 3679.97, N = 3 SE +/- 689.91, N = 3 858392 912806 302301 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
MariaDB Clients: 32 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 600 1200 1800 2400 3000 SE +/- 23.57, N = 4 SE +/- 31.65, N = 3 SE +/- 1.21, N = 3 1757 2905 914 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
MKL-DNN Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 14.43, N = 3 SE +/- 0.01, N = 3 SE +/- 101.14, N = 3 1752.47 5.77 11526.10 -O3 - MIN: 1493.25 -O2 - MIN: 11291.8 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_googlenet_v3 - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 14.62, N = 3 SE +/- 0.00, N = 3 SE +/- 34.01, N = 3 1764.58 6.23 11596.50 -O3 - MIN: 1518.91 -O2 - MIN: 11477.4 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Appleseed Scene: Material Tester OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Material Tester 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 60 120 180 240 300 284 185 161
GNU GMP GMPbench Total Time OpenBenchmarking.org GMPbench Score, More Is Better GNU GMP GMPbench 6.1.2 Total Time 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1300 2600 3900 5200 6500 4519 4589 5098 5922 1. (CC) gcc options: -O3 -fomit-frame-pointer -lm
MariaDB Clients: 16 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 16 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 1400 2800 4200 5600 7000 SE +/- 43.08, N = 6 SE +/- 73.23, N = 6 SE +/- 3.79, N = 3 3713 6347 946 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
asmFish 1024 Hash Memory, 26 Depth OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 30M 60M 90M 120M 150M SE +/- 769910.95, N = 3 SE +/- 1715314.92, N = 3 SE +/- 231748.24, N = 3 136257996 137371669 40037020
VP9 libvpx Encoding vpxenc VP9 1080p Video Encode OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode 2 x EPYC 7502 2 x Xeon Platinum 8280 30 60 90 120 150 SE +/- 0.79, N = 3 SE +/- 1.15, N = 3 145.29 94.82 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
Appleseed Scene: Emily OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Emily 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 60 120 180 240 300 173 155 270
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79 Blend File: Classroom - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 30 60 90 120 150 SE +/- 0.99, N = 3 SE +/- 0.34, N = 3 122.64 135.02
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79a Blend File: Barbershop - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 150 300 450 600 750 221 223 707
MKL-DNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 13K 26K 39K 52K 65K SE +/- 2.32, N = 3 SE +/- 1.15, N = 3 SE +/- 70.25, N = 3 1377 1890 59364 -O2 - MIN: 58877.3 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_3d - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 13K 26K 39K 52K 65K SE +/- 6.93, N = 3 SE +/- 2.38, N = 3 SE +/- 163.13, N = 3 1507 2041 59053 -O2 - MIN: 58346.1 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Parboil Test: OpenMP LBM OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 0.29, N = 15 SE +/- 0.30, N = 3 SE +/- 0.52, N = 15 SE +/- 0.04, N = 3 25.00 35.97 22.43 168.10 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Minion Benchmark: Quasigroup OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 30 60 90 120 150 SE +/- 0.20, N = 3 SE +/- 0.23, N = 3 SE +/- 0.32, N = 3 137.89 111.21 98.29 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79 Blend File: Pabellon Barcelona - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 30 60 90 120 150 SE +/- 0.37, N = 3 SE +/- 0.09, N = 3 100.13 112.00
Stress-NG Test: Context Switching OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Context Switching 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 13M 26M 39M 52M 65M SE +/- 2793972.74, N = 13 SE +/- 1626045.16, N = 15 SE +/- 41870.47, N = 3 62075360 26550213 5721856 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
NAMD ATPase Simulation - 327,506 Atoms OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 0.325 0.65 0.975 1.3 1.625 SE +/- 0.02081, N = 15 SE +/- 0.00162, N = 8 SE +/- 0.00011, N = 15 SE +/- 0.00006, N = 3 0.41349 0.26589 0.35923 1.44442
FFTW Build: Float + SSE - Size: 2D FFT Size 2048 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 2048 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 195.72, N = 3 SE +/- 51.68, N = 3 SE +/- 101.12, N = 3 SE +/- 99.90, N = 3 26109 26661 21198 21767 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
MariaDB Clients: 8 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 8 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 49.47, N = 3 SE +/- 140.23, N = 4 SE +/- 1.81, N = 3 6397 10422 986 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
CLOMP Static OMP Speedup OpenBenchmarking.org Speedup, More Is Better CLOMP 3.3 Static OMP Speedup 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 0.5513 1.1026 1.6539 2.2052 2.7565 SE +/- 0.01, N = 25 SE +/- 0.00, N = 5 SE +/- 0.00, N = 12 SE +/- 0.02, N = 5 0.37 0.23 0.43 2.45 1. (CC) gcc options: --openmp -O3 -lm
MariaDB Clients: 4 OpenBenchmarking.org Queries Per Second, More Is Better MariaDB 10.3.8 Clients: 4 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 33.65, N = 3 SE +/- 154.49, N = 3 SE +/- 11.81, N = 3 7634 13278 953 -llzma -lbz2 -lnuma -lpcre -llzma -lbz2 -lnuma -lpcre 1. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
MKL-DNN Harness: Convolution Batch conv_3d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_3d - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20 40 60 80 100 SE +/- 0.10, N = 15 SE +/- 0.01, N = 3 SE +/- 0.34, N = 3 3.67 3.30 101.70 -O2 - MIN: 100.15 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8K 16K 24K 32K 40K SE +/- 0.82, N = 3 SE +/- 0.00, N = 4 SE +/- 116.40, N = 3 795.68 0.23 35857.17 -O2 - MIN: 35696.3 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Renaissance Test: Apache Spark Bayes OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Apache Spark Bayes 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 42.13, N = 40 SE +/- 69.40, N = 40 SE +/- 49.08, N = 32 SE +/- 95.93, N = 8 3595 4126 3283 8382
MKL-DNN Harness: Deconvolution Batch deconv_1d - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8K 16K 24K 32K 40K SE +/- 1.86, N = 3 SE +/- 0.00, N = 3 SE +/- 20.50, N = 3 881.33 0.23 35756.50 -O2 - MIN: 35695.5 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
DaCapo Benchmark Java Test: Tradebeans OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 3K 6K 9K 12K 15K SE +/- 59.83, N = 4 SE +/- 56.27, N = 4 SE +/- 98.03, N = 4 7419 7743 11780
Apache Siege Concurrent Users: 250 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 250 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 7K 14K 21K 28K 35K SE +/- 98.88, N = 3 SE +/- 126.68, N = 3 SE +/- 203.16, N = 3 30037 34608 24626 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 4.18 Time To Compile 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 11 22 33 44 55 SE +/- 0.25, N = 15 SE +/- 0.35, N = 14 SE +/- 0.55, N = 3 19.74 21.45 47.79
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79 Blend File: Fishy Cat - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 20 40 60 80 100 SE +/- 0.10, N = 3 SE +/- 0.04, N = 3 70.16 76.82
Minion Benchmark: Solitaire OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20 40 60 80 100 SE +/- 0.17, N = 3 SE +/- 0.48, N = 3 SE +/- 0.30, N = 3 82.07 84.34 56.55 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
FFTW Build: Stock - Size: 2D FFT Size 2048 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 2048 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1400 2800 4200 5600 7000 SE +/- 71.82, N = 5 SE +/- 9.32, N = 3 SE +/- 75.14, N = 4 SE +/- 4.41, N = 3 5815 5993 5090 6573 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
MKL-DNN Harness: Convolution Batch conv_alexnet - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 0.75, N = 3 SE +/- 0.03, N = 3 SE +/- 213.03, N = 3 947.22 14.23 27767.67 -O2 - MIN: 27421.8 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Stress-NG Test: System V Message Passing OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: System V Message Passing 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 3M 6M 9M 12M 15M SE +/- 99301.92, N = 15 SE +/- 79946.02, N = 3 SE +/- 23189.98, N = 3 12216045 6551714 4645426 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Glibc C String Functions OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Glibc C String Functions 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 1.1M 2.2M 3.3M 4.4M 5.5M SE +/- 54834.48, N = 6 SE +/- 25939.97, N = 3 SE +/- 28284.32, N = 12 4908942 4483653 1262789 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Apache Siege Concurrent Users: 200 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 7K 14K 21K 28K 35K SE +/- 61.21, N = 3 SE +/- 38.45, N = 3 SE +/- 262.82, N = 3 28569 33432 24906 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Chaos Group V-RAY Mode: CPU OpenBenchmarking.org Ksamples, More Is Better Chaos Group V-RAY 4.10.03 Mode: CPU 2 x EPYC 7502 2 x Xeon Platinum 8280 14K 28K 42K 56K 70K SE +/- 755.94, N = 3 SE +/- 89.03, N = 3 62297 65476
Renaissance Test: In-Memory Database Shootout OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: In-Memory Database Shootout 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 120.36, N = 11 SE +/- 158.15, N = 10 SE +/- 35.03, N = 8 13683 17338 5735
MKL-DNN Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 2.17, N = 3 SE +/- 0.05, N = 3 SE +/- 101.98, N = 3 941.69 18.95 27648.10 -O2 - MIN: 27449.3 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Appleseed Scene: Disney Material OpenBenchmarking.org Seconds, Fewer Is Better Appleseed 2.0 Beta Scene: Disney Material 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 59.80 67.24 164.44
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79a Blend File: Pabellon Barcelona - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 80 160 240 320 400 102 117 370
Java SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Composite 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 600 1200 1800 2400 3000 SE +/- 2.65, N = 4 SE +/- 1.16, N = 4 SE +/- 4.49, N = 4 SE +/- 28.76, N = 16 2389 2423 2182 2924
IndigoBench Scene: Bedroom OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.0.64 Scene: Bedroom 2 x EPYC 7502 2 x Xeon Platinum 8280 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 4.78 6.14
Renaissance Test: Scala Dotty OpenBenchmarking.org ms, Fewer Is Better Renaissance 0.9.0 Test: Scala Dotty 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1600 3200 4800 6400 8000 SE +/- 43.69, N = 40 SE +/- 65.40, N = 40 SE +/- 55.49, N = 8 SE +/- 24.44, N = 8 7326 7276 6930 4891
NAS Parallel Benchmarks Test / Class: LU.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: LU.C 2 x EPYC 7502 2 x Xeon Platinum 8280 7K 14K 21K 28K 35K SE +/- 81.55, N = 3 SE +/- 34.81, N = 3 33940.81 32373.02 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Memcached mcperf Method: Delete OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Delete 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20K 40K 60K 80K 100K SE +/- 799.81, N = 15 SE +/- 309.68, N = 3 SE +/- 355.57, N = 3 40803 113029 77722 1. (CC) gcc options: -O2 -lm -rdynamic
IndigoBench Scene: Supercar OpenBenchmarking.org M samples/s, More Is Better IndigoBench 4.0.64 Scene: Supercar 2 x EPYC 7502 2 x Xeon Platinum 8280 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 7.72 9.88
Sockperf Test: Latency Under Load OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Under Load 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 8 16 24 32 40 SE +/- 0.38, N = 25 SE +/- 0.96, N = 25 SE +/- 1.51, N = 25 SE +/- 1.02, N = 15 22.42 22.16 35.39 18.51 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Stockfish Total Time OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 30M 60M 90M 120M 150M SE +/- 810846.41, N = 3 SE +/- 879556.81, N = 3 SE +/- 247195.47, N = 3 135223710 125947505 40049521 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto
Memcached mcperf Method: Set OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Set 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 13K 26K 39K 52K 65K SE +/- 423.60, N = 5 SE +/- 106.71, N = 3 SE +/- 4427.74, N = 12 33580 62021 58184 1. (CC) gcc options: -O2 -lm -rdynamic
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 200 400 600 800 1000 SE +/- 9.93, N = 15 SE +/- 12.66, N = 3 SE +/- 1.32, N = 3 693 1010 656 -lpthread 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++
Memcached mcperf Method: Add OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Add 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 13K 26K 39K 52K 65K SE +/- 537.65, N = 3 SE +/- 153.08, N = 3 SE +/- 1038.10, N = 12 33157 61510 46988 1. (CC) gcc options: -O2 -lm -rdynamic
MBW Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 16.61, N = 3 SE +/- 7.99, N = 3 SE +/- 22.38, N = 3 SE +/- 13.37, N = 3 9017 9032 5053 10170 1. (CC) gcc options: -O3 -march=native
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79a Blend File: Classroom - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 70 140 210 280 350 80.60 94.29 320.26
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 50K 100K 150K 200K 250K SE +/- 3053.66, N = 3 SE +/- 924.31, N = 3 SE +/- 207.54, N = 3 255492 232626 77585 1. (CXX) g++ options: -pipe -lpthread
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 600 1200 1800 2400 3000 SE +/- 4.17, N = 3 SE +/- 4.24, N = 3 SE +/- 19.25, N = 3 1004 2794 1278 1. (CC) gcc options: -O3 -mavx2
Memcached mcperf Method: Prepend OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Prepend 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 15K 30K 45K 60K 75K SE +/- 158.97, N = 3 SE +/- 200.54, N = 3 SE +/- 3176.72, N = 12 28279 70628 52587 1. (CC) gcc options: -O2 -lm -rdynamic
Minion Benchmark: Graceful OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Graceful 2 x EPYC 7502 2 x Xeon Platinum 8280 12 24 36 48 60 SE +/- 0.13, N = 3 SE +/- 0.06, N = 3 52.78 53.28 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Memcached mcperf Method: Get OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Get 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20K 40K 60K 80K 100K SE +/- 464.12, N = 15 SE +/- 712.60, N = 3 SE +/- 1204.09, N = 3 53650 111930 78724 1. (CC) gcc options: -O2 -lm -rdynamic
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79 Blend File: BMW27 - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 12 24 36 48 60 SE +/- 0.21, N = 3 SE +/- 0.05, N = 3 47.32 51.15
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPUSH 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 400K 800K 1200K 1600K 2000K SE +/- 26034.52, N = 15 SE +/- 18930.27, N = 15 SE +/- 25015.51, N = 12 1375853 1656464 1738434 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 6.0.1 Time To Compile 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 50 100 150 200 250 96.43 107.93 245.50
MKL-DNN Harness: IP Batch All - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 500 1000 1500 2000 2500 SE +/- 5.40, N = 3 SE +/- 0.27, N = 3 SE +/- 9.52, N = 3 778.49 20.48 2276.91 -O3 - MIN: 664.03 -O2 - MIN: 2248.79 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
MKL-DNN Harness: IP Batch All - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 500 1000 1500 2000 2500 SE +/- 13.00, N = 3 SE +/- 0.05, N = 3 SE +/- 11.34, N = 3 772.47 20.42 2265.86 -O3 - MIN: 607.02 -O2 - MIN: 2243.09 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.1.9 Time To Compile 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 11 22 33 44 55 SE +/- 0.02, N = 3 SE +/- 0.17, N = 3 SE +/- 0.10, N = 3 49.14 48.59 44.68 1. (CC) gcc options: -O2 -pedantic -ldl -lz -lm
MKL-DNN Harness: IP Batch All - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch All - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 200 400 600 800 1000 SE +/- 0.98, N = 3 SE +/- 1.72, N = 3 SE +/- 0.86, N = 3 81.93 100.80 791.93 -O2 - MIN: 786.54 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Stream Type: Copy OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 40K 80K 120K 160K 200K SE +/- 1580.30, N = 8 SE +/- 645.18, N = 5 SE +/- 4580.61, N = 25 SE +/- 21.78, N = 5 150788 185818 168490 14931 1. (CC) gcc options: -O3 -march=native -fopenmp
Redis Test: LPOP OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPOP 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 600K 1200K 1800K 2400K 3000K SE +/- 80051.04, N = 15 SE +/- 23924.88, N = 15 SE +/- 64125.23, N = 9 2170359 2475214 2727208 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
lzbench Test: XZ 0 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: XZ 0 - Process: Decompression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 20 40 60 80 100 87 88 95 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
lzbench Test: XZ 0 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: XZ 0 - Process: Compression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 8 16 24 32 40 28 28 36 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
MBW Test: Memory Copy - Array Size: 4096 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 23.34, N = 3 SE +/- 51.75, N = 3 SE +/- 2.51, N = 3 SE +/- 12.79, N = 3 15472 15436 4752 18783 1. (CC) gcc options: -O3 -march=native
dcraw RAW To PPM Image Conversion OpenBenchmarking.org Seconds, Fewer Is Better dcraw RAW To PPM Image Conversion 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 12 24 36 48 60 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 51.43 41.77 31.15 1. (CC) gcc options: -lm
Stress-NG Test: Semaphores OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Semaphores 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 1.2M 2.4M 3.6M 4.8M 6M SE +/- 7978.41, N = 3 SE +/- 23663.90, N = 6 SE +/- 15164.55, N = 3 1398972 2098939 5724009 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 500K 1000K 1500K 2000K 2500K SE +/- 43798.91, N = 15 SE +/- 30997.57, N = 15 SE +/- 24305.19, N = 3 1897043 2299349 2538538 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Redis Test: SADD OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SADD 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 500K 1000K 1500K 2000K 2500K SE +/- 26432.00, N = 15 SE +/- 20580.51, N = 15 SE +/- 4715.75, N = 3 1800546 2122727 2169218 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79a Blend File: Fishy Cat - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 66.52 72.54 194.00
Apache Siege Concurrent Users: 100 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 100 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 7K 14K 21K 28K 35K SE +/- 186.18, N = 3 SE +/- 119.32, N = 3 SE +/- 69.47, N = 3 26655 33242 24450 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
GIMP Test: auto-levels OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.8 Test: auto-levels 2 x EPYC 7502 2 x Xeon Platinum 8280 6 12 18 24 30 SE +/- 0.29, N = 6 SE +/- 0.08, N = 3 24.62 22.79
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 400K 800K 1200K 1600K 2000K SE +/- 28243.18, N = 15 SE +/- 21457.62, N = 3 SE +/- 29404.03, N = 12 1463246 1752860 1792791 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
lzbench Test: Libdeflate 1 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Libdeflate 1 - Process: Decompression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 200 400 600 800 1000 SE +/- 8.00, N = 3 841 867 1016 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
lzbench Test: Libdeflate 1 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Libdeflate 1 - Process: Compression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 50 100 150 200 250 SE +/- 1.00, N = 3 SE +/- 0.33, N = 3 190 191 213 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
MKL-DNN Harness: IP Batch 1D - Data Type: u8s8u8s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: u8s8u8s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 0.66, N = 15 SE +/- 0.03, N = 3 SE +/- 0.25, N = 3 68.47 3.64 177.05 -O3 - MIN: 47.15 -O2 - MIN: 172.2 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
lzbench Test: Zstd 1 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Zstd 1 - Process: Decompression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 200 400 600 800 1000 SE +/- 0.33, N = 3 SE +/- 7.88, N = 3 932 961 967 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
lzbench Test: Zstd 1 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Zstd 1 - Process: Compression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 80 160 240 320 400 354 360 375 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
lzbench Test: Brotli 0 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Brotli 0 - Process: Decompression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 110 220 330 440 550 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 441 446 519 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
lzbench Test: Brotli 0 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Brotli 0 - Process: Compression 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 80 160 240 320 400 SE +/- 0.67, N = 3 SE +/- 1.00, N = 3 383 383 391 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Tesseract OCR Time To OCR 7 Images OpenBenchmarking.org Seconds, Fewer Is Better Tesseract OCR 4.0.0 Time To OCR 7 Images 2 x EPYC 7502 2 x Xeon Platinum 8280 8 16 24 32 40 SE +/- 0.20, N = 3 SE +/- 0.03, N = 3 34.90 32.79
NAS Parallel Benchmarks Test / Class: FT.B OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: FT.B 2 x EPYC 7502 2 x Xeon Platinum 8280 800 1600 2400 3200 4000 SE +/- 3.88, N = 3 SE +/- 4.18, N = 3 3660.37 2556.20 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Stress-NG Test: Forking OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Forking 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 15K 30K 45K 60K 75K SE +/- 120.17, N = 3 SE +/- 870.64, N = 4 SE +/- 349.82, N = 3 43021 68990 63865 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Sunflow Rendering System Global Illumination + Image Synthesis OpenBenchmarking.org Seconds, Fewer Is Better Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.2475 0.495 0.7425 0.99 1.2375 SE +/- 0.01, N = 15 SE +/- 0.01, N = 3 SE +/- 0.02, N = 12 0.74 0.86 1.10
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.9.0-jumbo-1 Test: Blowfish 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20K 40K 60K 80K 100K SE +/- 1300.44, N = 3 SE +/- 567.99, N = 3 SE +/- 37.66, N = 3 109363 83578 20460 1. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lgmp -pthread -lm -lz -ldl -lcrypt -lbz2
Sockperf Test: Throughput OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 100K 200K 300K 400K 500K SE +/- 4514.21, N = 5 SE +/- 7584.07, N = 25 SE +/- 4430.66, N = 9 SE +/- 4526.82, N = 5 484363 454195 448019 373777 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Stress-NG Test: CPU Stress OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: CPU Stress 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 14.09, N = 3 SE +/- 121.24, N = 3 SE +/- 4.96, N = 3 19073 17955 4440 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Tsearch OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Tsearch 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 400 800 1200 1600 2000 SE +/- 6.47, N = 3 SE +/- 0.14, N = 3 SE +/- 2.30, N = 3 1752 1091 475 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Memory Copying OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Memory Copying 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 100.69, N = 3 SE +/- 108.22, N = 3 SE +/- 80.33, N = 3 9823 11263 9153 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Lsearch OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Lsearch 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 150 300 450 600 750 SE +/- 1.35, N = 3 SE +/- 1.81, N = 3 SE +/- 0.00, N = 3 708 665 178 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Crypto OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Crypto 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 28.56, N = 3 SE +/- 24.84, N = 3 SE +/- 6.14, N = 3 14306 10503 3520 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Glibc Qsort Data Sorting OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Glibc Qsort Data Sorting 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 200 400 600 800 1000 SE +/- 6.92, N = 3 SE +/- 1.75, N = 3 SE +/- 1.95, N = 3 931 787 206 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Bsearch OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Bsearch 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8K 16K 24K 32K 40K SE +/- 238.75, N = 3 SE +/- 71.90, N = 3 SE +/- 6.88, N = 3 33173 35116 8342 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Socket Activity OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Socket Activity 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8K 16K 24K 32K 40K SE +/- 58.06, N = 3 SE +/- 497.69, N = 3 SE +/- 44.07, N = 3 23811 37540 4183 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Hsearch OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Hsearch 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 80K 160K 240K 320K 400K SE +/- 782.39, N = 3 SE +/- 1743.97, N = 3 SE +/- 522.67, N = 3 388396 301308 97237 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Matrix Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Matrix Math 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 80K 160K 240K 320K 400K SE +/- 512.79, N = 3 SE +/- 1051.08, N = 3 SE +/- 112.22, N = 3 375525 380876 77603 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Vector Math OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Vector Math 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 30K 60K 90K 120K 150K SE +/- 244.96, N = 3 SE +/- 433.67, N = 3 SE +/- 3.94, N = 3 130505 120957 33507 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Memcached mcperf Method: Replace OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Replace 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 15K 30K 45K 60K 75K SE +/- 86.93, N = 3 SE +/- 251.51, N = 3 SE +/- 276.40, N = 3 28323 71185 46395 1. (CC) gcc options: -O2 -lm -rdynamic
GIMP Test: unsharp-mask OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.8 Test: unsharp-mask 2 x EPYC 7502 2 x Xeon Platinum 8280 7 14 21 28 35 SE +/- 0.11, N = 3 SE +/- 0.38, N = 3 30.60 28.18
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8 16 24 32 40 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.24, N = 3 33.17 29.97 24.38 1. (CC) gcc options: -lncurses -lm
Rodinia Test: OpenMP Streamcluster OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 6 12 18 24 30 SE +/- 0.38, N = 15 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 13.57 10.43 11.79 23.71 1. (CXX) g++ options: -O2 -lOpenCL
Coremark CoreMark Size 666 - Iterations Per Second OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 500K 1000K 1500K 2000K 2500K SE +/- 6421.46, N = 3 SE +/- 20125.94, N = 3 SE +/- 280.48, N = 3 2101299 1890649 579446 1. (CC) gcc options: -O2 -lrt" -lrt
Memcached mcperf Method: Append OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Append 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 15K 30K 45K 60K 75K SE +/- 247.66, N = 3 SE +/- 150.54, N = 3 SE +/- 54.84, N = 3 34850 70478 46554 1. (CC) gcc options: -O2 -lm -rdynamic
FFTW Build: Float + SSE - Size: 2D FFT Size 1024 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 1024 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K SE +/- 141.25, N = 3 SE +/- 264.46, N = 3 SE +/- 40.70, N = 3 SE +/- 255.01, N = 3 33251 33895 22351 41799 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79a Blend File: BMW27 - Compute: CPU-Only 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 30 60 90 120 150 45.20 47.45 138.06
Cython benchmark OpenBenchmarking.org Seconds, Fewer Is Better Cython benchmark 0.27 2 x EPYC 7502 2 x Xeon Platinum 8280 7 14 21 28 35 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 28.68 22.97
Rodinia Test: OpenMP CFD Solver OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 3 6 9 12 15 SE +/- 0.13, N = 15 SE +/- 0.13, N = 3 SE +/- 0.10, N = 15 SE +/- 0.01, N = 3 8.87 9.10 7.31 13.14 1. (CXX) g++ options: -O2 -lOpenCL
MKL-DNN Harness: IP Batch 1D - Data Type: u8s8f32s32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: u8s8f32s32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 1.00, N = 4 SE +/- 0.04, N = 8 SE +/- 0.39, N = 3 68.28 3.55 176.21 -O3 - MIN: 48 -O2 - MIN: 171.62 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Apache Siege Concurrent Users: 50 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 50 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 253.10, N = 3 SE +/- 426.87, N = 3 SE +/- 397.01, N = 6 21692 27915 28889 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
m-queens Time To Solve OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 11 22 33 44 55 SE +/- 0.10, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 12.10 16.30 46.75 1. (CXX) g++ options: -fopenmp -O2 -march=native
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6 12 18 24 30 SE +/- 0.13, N = 3 SE +/- 0.19, N = 3 SE +/- 0.04, N = 3 26.97 22.50 25.32 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
Botan Test: AES-256 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: AES-256 - Decrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 1300 2600 3900 5200 6500 SE +/- 4.72, N = 3 SE +/- 1.30, N = 3 SE +/- 11.04, N = 3 4640 3828 5896 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: AES-256 - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: AES-256 - Encrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 1300 2600 3900 5200 6500 SE +/- 1.53, N = 3 SE +/- 1.15, N = 3 SE +/- 4.49, N = 3 4585 3823 5910 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
C-Ray Total Time - 4K, 16 Rays Per Pixel OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 10 20 30 40 50 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 SE +/- 0.02, N = 3 10.90 14.40 42.22 1. (CC) gcc options: -lm -lpthread -O3
Rodinia Test: OpenMP LavaMD OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.07, N = 15 SE +/- 0.06, N = 4 SE +/- 0.12, N = 15 SE +/- 0.11, N = 3 6.11 4.33 6.96 20.08 1. (CXX) g++ options: -O2 -lOpenCL
MKL-DNN Harness: Deconvolution Batch deconv_1d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_1d - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 SE +/- 0.22, N = 3 8.11 0.95 197.67 -O2 - MIN: 196.86 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Timed ImageMagick Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.36, N = 3 SE +/- 0.21, N = 3 SE +/- 0.07, N = 3 21.39 19.67 19.95
Botan Test: Twofish - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Twofish - Decrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 80 160 240 320 400 SE +/- 0.10, N = 3 SE +/- 0.19, N = 3 SE +/- 0.64, N = 3 299 351 391 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: Twofish - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Twofish - Encrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 90 180 270 360 450 SE +/- 0.14, N = 3 SE +/- 0.20, N = 3 SE +/- 0.63, N = 3 299 347 396 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: Blowfish - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Blowfish - Decrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 60 120 180 240 300 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 228 275 297 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: Blowfish - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: Blowfish - Encrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 70 140 210 280 350 SE +/- 0.16, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 231 282 299 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: CAST-256 - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: CAST-256 - Decrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.59, N = 3 120 137 164 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: CAST-256 - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: CAST-256 - Encrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 40 80 120 160 200 SE +/- 0.07, N = 3 SE +/- 0.57, N = 3 SE +/- 0.55, N = 3 120 137 164 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 40.05, N = 3 SE +/- 64.31, N = 3 SE +/- 4.61, N = 3 14388 13753 3531 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Botan Test: KASUMI - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: KASUMI - Decrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.35, N = 3 75.24 87.03 101.82 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: KASUMI - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.8.0 Test: KASUMI - Encrypt 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20 40 60 80 100 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.36, N = 3 77.84 89.63 106.29 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
POV-Ray Trace Time OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 7 14 21 28 35 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 SE +/- 0.03, N = 3 10.32 11.57 31.55 -lSDL -lXpm -lSM -lICE -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lXpm -lSM -lICE -lIlmImf -lImath -lHalf -lIex -lIexMath -lIlmThread -lSDL 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lX11 -lpthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
Darktable Test: Server Room - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.0 Test: Server Room - Acceleration: CPU-only 2 x EPYC 7502 2 x Xeon Platinum 8280 0.6818 1.3636 2.0454 2.7272 3.409 SE +/- 0.17, N = 15 SE +/- 0.02, N = 15 3.03 1.92
RAMspeed SMP Type: Copy - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Floating Point 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 8K 16K 24K 32K 40K 30275 35526 32110 23394 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Triad - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Triad - Benchmark: Floating Point 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 10K 20K 30K 40K 50K 32247 46521 28432 27459 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Scale - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Scale - Benchmark: Floating Point 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K 40026 36196 25822 23339 1. (CC) gcc options: -O3 -march=native
dav1d Video Input: Summer Nature 4K OpenBenchmarking.org Seconds, Fewer Is Better dav1d 0.3 Video Input: Summer Nature 4K 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 13.24 18.25 22.26 1. (CC) gcc options: -pthread
DaCapo Benchmark Java Test: H2 OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: H2 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 2K 4K 6K 8K 10K SE +/- 92.11, N = 4 SE +/- 35.70, N = 4 SE +/- 38.38, N = 4 6911 8226 8773
Mencoder AVI To LAVC OpenBenchmarking.org Seconds, Fewer Is Better Mencoder 1.3.0 AVI To LAVC 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.25, N = 3 20.20 17.69 14.92 -lmad -lvorbisenc -lvorbis -logg -lfreetype -lXext -lX11 -lXxf86vm -lSDL -lGL 1. (CC) gcc options: -ffast-math -fpie -pie -lncurses -lrt -lpng -lz -ljpeg -lasound -ldl -lpthread -lbz2 -rdynamic -lm
GIMP Test: rotate OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.8 Test: rotate 2 x EPYC 7502 2 x Xeon Platinum 8280 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 16.75 18.25
RAMspeed SMP Type: Add - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Add - Benchmark: Floating Point 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 10K 20K 30K 40K 50K 48547 48530 28400 26957 1. (CC) gcc options: -O3 -march=native
MKL-DNN Harness: Convolution Batch conv_alexnet - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Convolution Batch conv_alexnet - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 500 1000 1500 2000 2500 SE +/- 0.89, N = 3 SE +/- 0.18, N = 3 SE +/- 0.24, N = 3 80.93 48.57 2393.04 -O2 - MIN: 2389.74 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
RAMspeed SMP Type: Average - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Average - Benchmark: Floating Point 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K 44275 40960 29060 27314 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Triad - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Triad - Benchmark: Integer 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K 40110 43023 31864 29597 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Copy - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Integer 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K 42739 41105 32285 23424 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Add - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Add - Benchmark: Integer 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 10K 20K 30K 40K 50K 47552 48253 32779 27016 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Scale - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Scale - Benchmark: Integer 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 8K 16K 24K 32K 40K 38465 37700 29096 23852 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Average - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Average - Benchmark: Integer 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K 42023 42272 31645 27333 1. (CC) gcc options: -O3 -march=native
PHPBench PHP Benchmark Suite OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 150K 300K 450K 600K 750K SE +/- 1963.83, N = 3 SE +/- 240.33, N = 3 SE +/- 11875.52, N = 3 494910 646738 698254
x265 H.265 1080p Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x265 3.0 H.265 1080p Video Encoding 2 x EPYC 7502 2 x Xeon Platinum 8280 10 20 30 40 50 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 43.46 32.73 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
Zstd Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.10, N = 6 SE +/- 0.07, N = 3 11.20 9.01 18.96 -llzma -llzma 1. (CC) gcc options: -O3 -pthread -lz
MKL-DNN Harness: IP Batch 1D - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 14 28 42 56 70 SE +/- 0.09, N = 3 SE +/- 0.15, N = 3 SE +/- 0.11, N = 3 9.47 11.56 64.09 -O3 - MIN: 4.89 -O3 - MIN: 6.18 -O2 - MIN: 63.62 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
Y-Cruncher Calculating 500M Pi Digits OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.7 Calculating 500M Pi Digits 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 8.89 9.12 21.92
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 3 6 9 12 15 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 10.09 9.56 7.53 -logg -logg 1. (CXX) g++ options: -O2 -fvisibility=hidden -lm
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-09-25 H.264 Video Encoding 2 x EPYC 7502 2 x Xeon Platinum 8280 30 60 90 120 150 SE +/- 0.99, N = 3 SE +/- 1.20, N = 15 148.94 120.04 1. (CC) gcc options: -ldl -lavformat -lavcodec -lavutil -lswscale -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
MBW Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 4.01, N = 3 SE +/- 3.77, N = 3 SE +/- 12.12, N = 3 SE +/- 12.73, N = 3 9093 9153 4916 9963 1. (CC) gcc options: -O3 -march=native
Sockperf Test: Latency Ping Pong OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1.1925 2.385 3.5775 4.77 5.9625 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.03, N = 5 SE +/- 0.02, N = 5 4.85 5.30 3.98 4.84 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
NAS Parallel Benchmarks Test / Class: LU.A OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: LU.A 2 x EPYC 7502 2 x Xeon Platinum 8280 7K 14K 21K 28K 35K SE +/- 966.94, N = 12 SE +/- 136.93, N = 3 31119.46 30471.20 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Parboil Test: OpenMP Stencil OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.06, N = 15 SE +/- 0.01, N = 3 5.64 4.02 2.60 16.90 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
FFTW Build: Stock - Size: 2D FFT Size 1024 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 1024 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 60.11, N = 3 SE +/- 6.26, N = 3 SE +/- 33.68, N = 3 SE +/- 21.10, N = 3 6099 6146 6340 7909 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
GIMP Test: resize OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.8 Test: resize 2 x EPYC 7502 2 x Xeon Platinum 8280 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 11.67 12.48
MBW Test: Memory Copy - Array Size: 1024 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 1024 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 66.85, N = 3 SE +/- 35.67, N = 3 SE +/- 1.38, N = 3 SE +/- 26.28, N = 3 15711 15602 4775 18599 1. (CC) gcc options: -O3 -march=native
Timed MAFFT Alignment Multiple Sequence Alignment OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.392 Multiple Sequence Alignment 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1.161 2.322 3.483 4.644 5.805 SE +/- 0.03, N = 15 SE +/- 0.10, N = 15 SE +/- 0.07, N = 3 SE +/- 0.02, N = 2 3.42 3.41 5.16 2.08 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
Bork File Encrypter File Encryption Time OpenBenchmarking.org Seconds, Fewer Is Better Bork File Encrypter 1.4 File Encryption Time 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.22, N = 3 10.77 10.73 7.10 12.08
OSBench Test: Create Threads OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8 16 24 32 40 SE +/- 0.37, N = 3 SE +/- 0.06, N = 3 SE +/- 0.35, N = 12 33.79 20.51 14.68 1. (CC) gcc options: -lm
Apache Siege Concurrent Users: 10 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 10 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 86.97, N = 3 SE +/- 326.78, N = 15 SE +/- 238.51, N = 3 19908 26012 20400 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
DaCapo Benchmark Java Test: Jython OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 1200 2400 3600 4800 6000 SE +/- 67.79, N = 4 SE +/- 26.33, N = 4 SE +/- 36.59, N = 4 5466 5264 4629
FFTW Build: Float + SSE - Size: 1D FFT Size 64 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 64 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 5K 10K 15K 20K 25K SE +/- 26.96, N = 3 SE +/- 401.12, N = 15 SE +/- 219.94, N = 13 SE +/- 12.55, N = 3 16447 16032 16984 22302 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OSBench Test: Launch Programs OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20 40 60 80 100 SE +/- 0.82, N = 3 SE +/- 0.52, N = 9 SE +/- 0.19, N = 3 78.17 54.51 37.85 1. (CC) gcc options: -lm
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 7.4 1e12 Prime Number Generation 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 4.29 4.82 15.47 -O3 -O3 -O2 1. (CXX) g++ options: -lpthread
SVT-AV1 1080p 8-bit YUV To AV1 Video Encode OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.5 1080p 8-bit YUV To AV1 Video Encode 2 x EPYC 7502 2 x Xeon Platinum 8280 20 40 60 80 100 SE +/- 0.46, N = 3 SE +/- 0.65, N = 3 94.99 61.53 1. (CXX) g++ options: -O3 -pie -lpthread -lm
FFTW Build: Float + SSE - Size: 1D FFT Size 256 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 256 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K SE +/- 470.01, N = 3 SE +/- 247.84, N = 3 SE +/- 722.12, N = 15 SE +/- 83.95, N = 3 27950 27860 36704 39978 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
dav1d Video Input: Summer Nature 1080p OpenBenchmarking.org Seconds, Fewer Is Better dav1d 0.3 Video Input: Summer Nature 1080p 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 5.69 9.20 7.49 1. (CC) gcc options: -pthread
FFTW Build: Float + SSE - Size: 1D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 4096 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 12K 24K 36K 48K 60K SE +/- 553.18, N = 3 SE +/- 395.93, N = 3 SE +/- 504.24, N = 3 SE +/- 914.81, N = 3 41618 43038 58222 58217 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Float + SSE - Size: 2D FFT Size 512 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 512 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K SE +/- 128.07, N = 3 SE +/- 85.48, N = 3 SE +/- 15.95, N = 3 SE +/- 311.93, N = 3 33879 34178 22289 42671 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
MBW Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 512 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 3.34, N = 3 SE +/- 17.16, N = 3 SE +/- 1.22, N = 3 SE +/- 10.38, N = 3 9126 9188 4910 9896 1. (CC) gcc options: -O3 -march=native
Ogg Encoding WAV To Ogg OpenBenchmarking.org Seconds, Fewer Is Better Ogg Encoding 1.3.3 WAV To Ogg 2 x EPYC 7502 2 x Xeon Platinum 8280 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 6.81 6.12 1. (CC) gcc options: -O2 -ffast-math -fsigned-char -logg
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 2.0.2 Test: Decompression Throughput 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 50 100 150 200 250 SE +/- 0.07, N = 3 SE +/- 0.48, N = 3 SE +/- 0.42, N = 3 169 188 233 1. (CC) gcc options: -O3 -rdynamic
FFTW Build: Float + SSE - Size: 1D FFT Size 128 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 128 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 6K 12K 18K 24K 30K SE +/- 71.04, N = 3 SE +/- 232.04, N = 7 SE +/- 316.85, N = 3 SE +/- 315.54, N = 12 20960 21059 20343 29016 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 4096 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 4096 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 9.28, N = 3 SE +/- 13.44, N = 3 SE +/- 20.22, N = 3 SE +/- 85.21, N = 3 6709 6814 7274 8713 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 32 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 32 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 2.81, N = 3 SE +/- 0.23, N = 3 SE +/- 358.52, N = 15 SE +/- 9.02, N = 3 8741 8868 7179 11691 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Darktable Test: Masskrug - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.0 Test: Masskrug - Acceleration: CPU-only 2 x EPYC 7502 2 x Xeon Platinum 8280 1.0733 2.1466 3.2199 4.2932 5.3665 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 4.24 4.77
OSBench Test: Create Files OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 15.23 13.77 17.06 1. (CC) gcc options: -lm
FFTW Build: Float + SSE - Size: 1D FFT Size 2048 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 2048 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 14K 28K 42K 56K 70K SE +/- 549.65, N = 3 SE +/- 332.33, N = 3 SE +/- 650.59, N = 3 SE +/- 462.69, N = 3 44915 45337 63908 60606 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
MBW Test: Memory Copy - Array Size: 512 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 512 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 59.01, N = 3 SE +/- 48.20, N = 3 SE +/- 2.75, N = 3 SE +/- 25.17, N = 3 15671 15464 4819 18439 1. (CC) gcc options: -O3 -march=native
FFTW Build: Float + SSE - Size: 1D FFT Size 1024 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 1024 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 14K 28K 42K 56K 70K SE +/- 362.20, N = 3 SE +/- 58.50, N = 3 SE +/- 1102.91, N = 3 SE +/- 855.91, N = 3 43078 43869 64732 58387 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Float + SSE - Size: 2D FFT Size 256 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 256 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 9K 18K 27K 36K 45K SE +/- 93.28, N = 3 SE +/- 194.96, N = 3 SE +/- 413.74, N = 3 SE +/- 217.10, N = 3 33586 34284 24853 44203 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
NAS Parallel Benchmarks Test / Class: EP.C OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: EP.C 2 x EPYC 7502 2 x Xeon Platinum 8280 500 1000 1500 2000 2500 SE +/- 1.59, N = 3 SE +/- 22.95, N = 3 2283.03 2486.05 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OSBench Test: Memory Allocations OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 20 40 60 80 100 SE +/- 0.23, N = 3 SE +/- 0.15, N = 3 SE +/- 0.09, N = 3 96.59 87.57 68.01 1. (CC) gcc options: -lm
OSBench Test: Create Processes OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 14 28 42 56 70 SE +/- 0.76, N = 3 SE +/- 0.79, N = 3 SE +/- 0.25, N = 3 63.31 45.92 27.63 1. (CC) gcc options: -lm
SVT-VP9 1080p 8-bit YUV To VP9 Video Encode OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode 2 x EPYC 7502 2 x Xeon Platinum 8280 60 120 180 240 300 SE +/- 1.34, N = 3 SE +/- 3.53, N = 4 197.26 271.63 1. (CC) gcc options: -fPIE -fPIC -O2 -flto -fvisibility=hidden -mavx -pie -rdynamic -lpthread -lrt -lm
FFTW Build: Stock - Size: 2D FFT Size 512 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 512 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 3.11, N = 3 SE +/- 9.03, N = 3 SE +/- 31.31, N = 3 SE +/- 34.09, N = 3 6633 6717 6355 9012 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 2048 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 2048 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 15.82, N = 3 SE +/- 0.67, N = 3 SE +/- 56.60, N = 3 SE +/- 100.27, N = 3 6858 7006 7392 8998 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Darktable Test: Boat - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.0 Test: Boat - Acceleration: CPU-only 2 x EPYC 7502 2 x Xeon Platinum 8280 0.9113 1.8226 2.7339 3.6452 4.5565 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 3.30 4.05
NAS Parallel Benchmarks Test / Class: FT.A OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.3.1 Test / Class: FT.A 2 x EPYC 7502 2 x Xeon Platinum 8280 700 1400 2100 2800 3500 SE +/- 5.92, N = 3 SE +/- 2.56, N = 3 3332.60 2378.80 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
FFTW Build: Float + SSE - Size: 1D FFT Size 512 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 512 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 12K 24K 36K 48K 60K SE +/- 56.26, N = 3 SE +/- 336.45, N = 3 SE +/- 886.61, N = 3 SE +/- 388.57, N = 3 37722 38922 55983 56806 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Darktable Test: Server Rack - Acceleration: CPU-only OpenBenchmarking.org Seconds, Fewer Is Better Darktable 2.6.0 Test: Server Rack - Acceleration: CPU-only 2 x EPYC 7502 2 x Xeon Platinum 8280 0.0248 0.0496 0.0744 0.0992 0.124 SE +/- 0.01, N = 12 SE +/- 0.01, N = 12 0.11 0.10
FFTW Build: Stock - Size: 2D FFT Size 256 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 256 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 13.59, N = 3 SE +/- 18.60, N = 3 SE +/- 7.72, N = 3 SE +/- 16.61, N = 3 6615 6744 6386 8683 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 1024 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 1024 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 2.19, N = 3 SE +/- 22.72, N = 3 SE +/- 81.37, N = 3 SE +/- 10.71, N = 3 7129 7207 7555 9726 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
SVT-HEVC 1080p 8-bit YUV To HEVC Video Encode OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode 2 x EPYC 7502 2 x Xeon Platinum 8280 70 140 210 280 350 SE +/- 2.17, N = 3 SE +/- 3.63, N = 4 340.94 260.31 1. (CC) gcc options: -fPIE -fPIC -O2 -flto -fvisibility=hidden -march=native -pie -rdynamic -lpthread -lrt
FFTW Build: Float + SSE - Size: 2D FFT Size 128 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 128 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 10K 20K 30K 40K 50K SE +/- 72.57, N = 3 SE +/- 152.49, N = 3 SE +/- 356.69, N = 3 SE +/- 891.44, N = 3 34549 35003 28613 48512 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 2D FFT Size 64 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 64 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 2.64, N = 3 SE +/- 5.63, N = 3 SE +/- 55.12, N = 3 SE +/- 30.93, N = 3 7300 7400 7181 9601 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 2D FFT Size 128 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 128 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 40.50, N = 3 SE +/- 15.81, N = 3 SE +/- 108.79, N = 3 SE +/- 26.29, N = 3 6788 6948 7470 8988 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Float + SSE - Size: 1D FFT Size 32 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 1D FFT Size 32 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 3K 6K 9K 12K 15K SE +/- 10.11, N = 3 SE +/- 31.83, N = 3 SE +/- 31.05, N = 3 SE +/- 16.23, N = 3 11848 11985 14317 15955 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 64 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 64 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 0.58, N = 3 SE +/- 3.05, N = 3 SE +/- 5.52, N = 3 SE +/- 61.23, N = 3 8082 8203 7404 8701 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
MKL-DNN Harness: Deconvolution Batch deconv_3d - Data Type: f32 OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: Deconvolution Batch deconv_3d - Data Type: f32 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 9 18 27 36 45 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.06, N = 3 3.86 1.11 38.14 -O2 - MIN: 37.77 1. (CXX) g++ options: -std=c++11 -march=native -mtune=native -fPIC -fopenmp -pie -lmklml_intel -ldl
FFTW Build: Stock - Size: 1D FFT Size 512 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 512 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 10.19, N = 3 SE +/- 15.02, N = 3 SE +/- 9.12, N = 3 SE +/- 9.39, N = 3 7052 7124 7793 9288 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 256 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 256 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 1.71, N = 3 SE +/- 14.81, N = 3 SE +/- 56.78, N = 3 SE +/- 88.12, N = 3 7127 7217 7200 8981 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Apache Siege Concurrent Users: 1 OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 1 2 x EPYC 7502 2 x Xeon Platinum 8280 1000 2000 3000 4000 5000 SE +/- 36.83, N = 3 SE +/- 8.15, N = 3 4559.86 2606.48 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
FFTW Build: Float + SSE - Size: 2D FFT Size 32 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 32 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 10K 20K 30K 40K 50K SE +/- 35.51, N = 3 SE +/- 76.70, N = 3 SE +/- 223.03, N = 3 SE +/- 306.88, N = 3 34572 35028 27128 48195 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Float + SSE - Size: 2D FFT Size 64 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 64 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 11K 22K 33K 44K 55K SE +/- 327.91, N = 3 SE +/- 22.10, N = 3 SE +/- 124.46, N = 3 SE +/- 450.07, N = 3 35897 37011 33648 49146 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTE Test: N=256, 1D Complex FFT Routine OpenBenchmarking.org MFLOPS, More Is Better FFTE 6.0 Test: N=256, 1D Complex FFT Routine 2 x EPYC 7502 1600 3200 4800 6400 8000 SE +/- 1.52, N = 3 7639.45 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
PHP Micro Benchmarks Test: Zend micro_bench OpenBenchmarking.org Seconds, Fewer Is Better PHP Micro Benchmarks Test: Zend micro_bench 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.5985 1.197 1.7955 2.394 2.9925 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 6 2.66 2.06 1.80
FFTW Build: Stock - Size: 2D FFT Size 32 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 32 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 13.25, N = 3 SE +/- 16.76, N = 3 SE +/- 69.98, N = 3 SE +/- 11.50, N = 3 8824 8982 8898 11466 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 1D FFT Size 128 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 128 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 2.34, N = 3 SE +/- 55.49, N = 3 SE +/- 99.21, N = 3 SE +/- 108.32, N = 3 6766 6899 7024 8996 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Parboil Test: OpenMP CUTCP OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 0.4658 0.9316 1.3974 1.8632 2.329 SE +/- 0.02, N = 4 SE +/- 0.01, N = 3 SE +/- 0.02, N = 12 SE +/- 0.01, N = 3 1.04 0.80 1.29 2.07 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 300K 600K 900K 1200K 1500K SE +/- 778.60, N = 5 SE +/- 1283.95, N = 5 SE +/- 2434.15, N = 5 1036078 1257489 1376962 1. (CC) gcc options: -O3 -march=native
MBW Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 128 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2K 4K 6K 8K 10K SE +/- 30.78, N = 3 SE +/- 9.56, N = 3 SE +/- 10.85, N = 3 SE +/- 16.17, N = 3 8984 9053 4918 9393 1. (CC) gcc options: -O3 -march=native
MBW Test: Memory Copy - Array Size: 128 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 128 MiB 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 4K 8K 12K 16K 20K SE +/- 5.56, N = 3 SE +/- 16.75, N = 3 SE +/- 3.38, N = 3 SE +/- 17.09, N = 3 8986 9066 4975 18521 1. (CC) gcc options: -O3 -march=native
ctx_clock Context Switch Time OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 30 60 90 120 150 150 136 152
BLAKE2 OpenBenchmarking.org Cycles Per Byte, Fewer Is Better BLAKE2 20170307 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 7.46 6.60 3.87 8.30 1. (CC) gcc options: -O3 -march=native -lcrypto -lz
PHP Micro Benchmarks Test: Zend bench OpenBenchmarking.org Seconds, Fewer Is Better PHP Micro Benchmarks Test: Zend bench 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.1215 0.243 0.3645 0.486 0.6075 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.54 0.45 0.40
Phoronix Test Suite Self Test Test: Info Command OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Info Command 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.693 1.386 2.079 2.772 3.465 SE +/- 0.05, N = 2 SE +/- 0.00, N = 2 SE +/- 0.01, N = 2 3.08 2.40 1.20
Phoronix Test Suite Self Test Test: Peak Memory Usage OpenBenchmarking.org MB, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Peak Memory Usage 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 110 220 330 440 550 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 503 503 500
Phoronix Test Suite Self Test Test: Merge Results OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Merge Results 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 16 32 48 64 80 SE +/- 0.03, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 4.84 3.91 74.28
Phoronix Test Suite Self Test Test: Render Test OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: Render Test 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 16 32 48 64 80 SE +/- 0.25, N = 2 SE +/- 0.37, N = 2 SE +/- 0.12, N = 2 69.67 59.80 35.29
Phoronix Test Suite Self Test Test: List Suites OpenBenchmarking.org Seconds, Fewer Is Better Phoronix Test Suite Self Test 1.0 Test: List Suites 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5 10 15 20 25 SE +/- 0.01, N = 2 SE +/- 0.00, N = 2 SE +/- 0.00, N = 2 0.74 0.71 20.17
Java SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 400 800 1200 1600 2000 SE +/- 0.00, N = 4 SE +/- 0.15, N = 4 SE +/- 6.63, N = 4 SE +/- 10.90, N = 4 1506 1529 1585 1716
Java SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Dense LU Matrix Factorization 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 1500 3000 4500 6000 7500 SE +/- 9.37, N = 4 SE +/- 9.32, N = 4 SE +/- 15.06, N = 4 SE +/- 34.41, N = 4 5270 5364 4793 6944
Java SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Sparse Matrix Multiply 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 600 1200 1800 2400 3000 SE +/- 1.10, N = 4 SE +/- 0.56, N = 4 SE +/- 0.56, N = 4 SE +/- 15.03, N = 4 2211 2246 2156 2674
Java SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Fast Fourier Transform 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 300 600 900 1200 1500 SE +/- 7.36, N = 4 SE +/- 4.68, N = 4 SE +/- 5.82, N = 4 SE +/- 227.53, N = 4 1624 1622 1392 1280
Java SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Monte Carlo 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 400 800 1200 1600 2000 SE +/- 2.07, N = 4 SE +/- 0.37, N = 4 SE +/- 0.22, N = 4 SE +/- 9.95, N = 4 1334 1356 984 1716
HPC Challenge Test / Class: Max Ping Pong Bandwidth OpenBenchmarking.org MB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Max Ping Pong Bandwidth 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 5K 10K 15K 20K 25K SE +/- 77.41, N = 3 SE +/- 56.79, N = 3 SE +/- 67.88, N = 3 10781 17274 25023 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: Random Ring Bandwidth OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Bandwidth 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 1.1813 2.3626 3.5439 4.7252 5.9065 SE +/- 0.00593, N = 3 SE +/- 0.00164, N = 3 SE +/- 0.08933, N = 3 1.38385 0.83730 5.25031 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: Random Ring Latency OpenBenchmarking.org usecs, Fewer Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Latency 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.4004 0.8008 1.2012 1.6016 2.002 SE +/- 0.00742, N = 3 SE +/- 0.00802, N = 3 SE +/- 0.00148, N = 3 1.77976 1.13356 0.59617 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: G-Random Access OpenBenchmarking.org GUP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Random Access 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.0455 0.091 0.1365 0.182 0.2275 SE +/- 0.00214, N = 3 SE +/- 0.00165, N = 3 SE +/- 0.00049, N = 3 0.20233 0.17540 0.03249 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: EP-STREAM Triad OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: EP-STREAM Triad 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 0.6756 1.3512 2.0268 2.7024 3.378 SE +/- 0.00633, N = 3 SE +/- 0.00024, N = 3 SE +/- 0.00135, N = 3 2.80207 3.00269 2.07594 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: G-Ptrans OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ptrans 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 4 8 12 16 20 SE +/- 0.21006, N = 3 SE +/- 0.01412, N = 3 SE +/- 0.00171, N = 3 1.36882 15.01817 2.49771 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: EP-DGEMM OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: EP-DGEMM 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 6 12 18 24 30 SE +/- 0.11773, N = 3 SE +/- 0.05258, N = 3 SE +/- 0.00010, N = 3 22.38970 25.17353 1.07655 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: G-Ffte OpenBenchmarking.org GFLOP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8 16 24 32 40 SE +/- 0.15951, N = 3 SE +/- 0.05498, N = 3 SE +/- 0.03972, N = 3 33.30483 16.90603 5.40971 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
HPC Challenge Test / Class: G-Ffte OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte 2 x EPYC 7502 2 x Xeon Platinum 8280 Ryzen 3900X 8 16 24 32 40 SE +/- 0.15951, N = 3 SE +/- 0.05498, N = 3 SE +/- 0.03972, N = 3 33.30483 16.90603 5.40971 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. 2 x EPYC 7502: ATLAS + Open MPI 3.1.3 3. 2 x Xeon Platinum 8280: ATLAS + Open MPI 3.1.3 4. Ryzen 3900X: ATLAS + Open MPI 3.1.4
Stream Type: Add OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 40K 80K 120K 160K 200K SE +/- 1387.96, N = 5 SE +/- 713.30, N = 5 SE +/- 4249.65, N = 5 SE +/- 31.89, N = 5 164105 194680 156319 16284 1. (CC) gcc options: -O3 -march=native -fopenmp
Stream Type: Triad OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 40K 80K 120K 160K 200K SE +/- 2801.94, N = 5 SE +/- 624.77, N = 5 SE +/- 5754.74, N = 5 SE +/- 2.30, N = 5 166119 200966 154125 16305 1. (CC) gcc options: -O3 -march=native -fopenmp
Stream Type: Scale OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 40K 80K 120K 160K 200K SE +/- 1282.73, N = 5 SE +/- 986.94, N = 5 SE +/- 682.69, N = 5 SE +/- 22.49, N = 5 155497 177029 149490 15170 1. (CC) gcc options: -O3 -march=native -fopenmp
BlogBench Test: Write OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write 2 x EPYC 7502 2 x EPYC 7742 2 x Xeon Platinum 8280 Ryzen 3900X 11K 22K 33K 44K 55K SE +/- 882.24, N = 3 SE +/- 7.00, N = 2 SE +/- 449.95, N = 3 SE +/- 498.08, N = 3 43807 38226 53516 32787 1. (CC) gcc options: -O2 -pthread
Phoronix Test Suite v10.8.4