extra tests

2 x AMD EPYC 9124 16-Core testing with a Supermicro H13DSH (1.5 BIOS) and astdrmfb on AlmaLinux 9.2 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2308259-NE-EXTRATEST30
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 9124 2P
August 25 2023
  3 Hours, 7 Minutes
b
August 25 2023
  10 Hours, 34 Minutes
c
August 25 2023
  3 Hours, 12 Minutes
d
August 25 2023
  3 Hours, 8 Minutes
Invert Behavior (Only Show Selected Data)
  5 Hours

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


extra testsOpenBenchmarking.orgPhoronix Test Suite2 x AMD EPYC 9124 16-Core @ 3.00GHz (32 Cores / 64 Threads)Supermicro H13DSH (1.5 BIOS)24 x 32 GB DDR5-4800MT/s Samsung M321R4GA3BB6-CQKET2 x 1920GB SAMSUNG MZQL21T9HCJR-00A07astdrmfbAlmaLinux 9.25.14.0-284.25.1.el9_2.x86_64 (x86_64)GCC 11.3.1 20221121ext41024x768ProcessorMotherboardMemoryDiskGraphicsOSKernelCompilerFile-SystemScreen ResolutionExtra Tests PerformanceSystem Logs- Transparent Huge Pages: always- --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-host-bind-now --enable-host-pie --enable-initfini-array --enable-languages=c,c++,fortran,lto --enable-link-serialization=1 --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-arch_64=x86-64-v2 --with-build-config=bootstrap-lto --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver --without-isl - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa10113e - OpenJDK Runtime Environment (Red_Hat-11.0.20.0.8-1) (build 11.0.20+8-LTS)- Python 3.9.16- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected

EPYC 9124 2PbcdResult OverviewPhoronix Test Suite100%101%102%103%NCNNStress-NGApache CassandraDragonflydbSVT-AV1RemhosKripkeTimed Linux Kernel CompilationSPECFEM3DnekRSBRL-CADIntel Open Image DenoiseVVenCLaghosNeural Magic DeepSparseLiquid-DSPEmbreeOSPRayBlender

extra testsstress-ng: Cryptostress-ng: Semaphoresdragonflydb: 50 - 1:100svt-av1: Preset 12 - Bosphorus 4Kdragonflydb: 20 - 1:10dragonflydb: 20 - 1:100ncnn: CPU - yolov4-tinyncnn: CPU - shufflenet-v2ncnn: CPU - regnety_400mdragonflydb: 10 - 1:10stress-ng: Futexncnn: CPU - blazefacesvt-av1: Preset 12 - Bosphorus 1080pstress-ng: Matrix 3D Mathliquid-dsp: 16 - 256 - 57ncnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - vision_transformerncnn: CPU - efficientnet-b0ncnn: CPU - squeezenet_ssdspecfem3d: Water-layered Halfspacedragonflydb: 10 - 1:100liquid-dsp: 32 - 256 - 57svt-av1: Preset 13 - Bosphorus 1080pstress-ng: Glibc C String Functionsncnn: CPU - mobilenetstress-ng: Pipecassandra: Writesncnn: CPU - googlenetstress-ng: Context Switchingliquid-dsp: 8 - 256 - 512liquid-dsp: 64 - 256 - 57remhos: Sample Remap Examplekripke: specfem3d: Homogeneous Halfspacenekrs: Kershawdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Streambuild-linux-kernel: defconfigspecfem3d: Tomographic Modelstress-ng: Cloningliquid-dsp: 4 - 256 - 57deepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamdeepsparse: ResNet-50, Sparse INT8 - Synchronous Single-Streamsvt-av1: Preset 8 - Bosphorus 1080pliquid-dsp: 8 - 256 - 57svt-av1: Preset 8 - Bosphorus 4Ksvt-av1: Preset 4 - Bosphorus 1080pncnn: CPU - resnet50specfem3d: Mount St. Helensncnn: CPU - alexnetvvenc: Bosphorus 1080p - Fastervvenc: Bosphorus 4K - Fastsvt-av1: Preset 13 - Bosphorus 4Kncnn: CPU - mnasnetstress-ng: CPU Stressliquid-dsp: 4 - 256 - 512ncnn: CPU - resnet18stress-ng: AVL Treedeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Streamdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Streamliquid-dsp: 2 - 256 - 512oidn: RTLightmap.hdr.4096x4096 - CPU-Onlyliquid-dsp: 8 - 256 - 32embree: Pathtracer - Crownncnn: CPU - vgg16stress-ng: MMAPstress-ng: Mixed Schedulerliquid-dsp: 16 - 256 - 512deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Streamstress-ng: Vector Floating Pointvvenc: Bosphorus 1080p - Fastdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streambrl-cad: VGR Performance Metricdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Synchronous Single-Streamliquid-dsp: 1 - 256 - 512specfem3d: Layered Halfspacestress-ng: Atomicembree: Pathtracer ISPC - Crownembree: Pathtracer - Asian Dragon Objstress-ng: MEMFDospray: gravity_spheres_volume/dim_512/ao/real_timestress-ng: Mallocnekrs: TurboPipe Periodicdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Streamstress-ng: Function Callstress-ng: SENDFILEsvt-av1: Preset 4 - Bosphorus 4Kdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamoidn: RT.ldr_alb_nrm.3840x2160 - CPU-Onlyoidn: RT.hdr_alb_nrm.3840x2160 - CPU-Onlyblender: BMW27 - CPU-Onlyliquid-dsp: 64 - 256 - 512liquid-dsp: 32 - 256 - 512deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamliquid-dsp: 32 - 256 - 32deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Streamospray: gravity_spheres_volume/dim_512/scivis/real_timeblender: Fishy Cat - CPU-Onlydeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamembree: Pathtracer ISPC - Asian Dragon Objvvenc: Bosphorus 4K - Fasterliquid-dsp: 4 - 256 - 32deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamliquid-dsp: 64 - 256 - 32liquid-dsp: 16 - 256 - 32deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamlaghos: Triple Point Problemdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Streamblender: Pabellon Barcelona - CPU-Onlyembree: Pathtracer - Asian Dragondeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: ResNet-50, Baseline - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Streamdeepsparse: ResNet-50, Baseline - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Streamdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamlaghos: Sedov Blast Wave, ube_922_hex.meshdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Synchronous Single-Streamstress-ng: Glibc Qsort Data Sortingdeepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Streamdeepsparse: BERT-Large, NLP Question Answering - Synchronous Single-Streamstress-ng: Forkingdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamdeepsparse: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Streamospray: particle_volume/pathtracer/real_timestress-ng: Vector Shuffledeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamstress-ng: Zlibospray: gravity_spheres_volume/dim_512/pathtracer/real_timedeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamblender: Barbershop - CPU-Onlydeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Streamstress-ng: Mutexstress-ng: System V Message Passingliquid-dsp: 1 - 256 - 32blender: Classroom - CPU-Onlyospray: particle_volume/ao/real_timestress-ng: Floating Pointdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamdeepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Streamstress-ng: Wide Vector Mathliquid-dsp: 1 - 256 - 57ospray: particle_volume/scivis/real_timedeepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Streamstress-ng: AVX-512 VNNIdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamstress-ng: Memory Copyingdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamembree: Pathtracer ISPC - Asian Dragonstress-ng: Pollstress-ng: Fused Multiply-Adddeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Streamstress-ng: Matrix Mathstress-ng: Vector Mathdeepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Streamdeepsparse: CV Detection, YOLOv5s COCO - Synchronous Single-Streamliquid-dsp: 2 - 256 - 32deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamstress-ng: Hashstress-ng: x86_64 RdRandstress-ng: Pthreadncnn: CPU - FastestDetstress-ng: Socket Activitystress-ng: CPU Cachestress-ng: NUMAdragonflydb: 50 - 1:10liquid-dsp: 2 - 256 - 57build-linux-kernel: allmodconfigEPYC 9124 2Pbcd79760.8281543324.9215294187.64188.69613016172.3815292696.235.6314.5933.7512095393.844322676.825.8443.52811756.3560847000011.6311.2757.2315.6524.934.64914764311365230.141205600000522.52141243730.2722.9820487917.5422207529.4918252120.5298607000180900000020.35637625870018.9246984191146960000019.312651.758191.534610.911235.1714.9491478031173.141781100001.1021904.4208139.3133879000070.77415.15426.2914.7135915678.0633.6556.763199.36410.8989602.514962700013.53410.957.2056138.6214246800000.6527581000037.966328.311142.0135532.5619779000070.955814.0918107730.418.942204.81445459524.87721263700038.09607639235.3539.119438.3359920.910.1003136350929.27741085000070.832514.116127757.95852899.855.068106.78641.371.3637.66501070000388330000149.549108100000048.316820.68659.8768249.25115.002440.821612.755136890000138.820924.855420562000005506100005.9715196.44140.14032673.221122.4942.636242.718561.7898140.81537.0918374.0267754.6362113.9236258.667521.1701793.7775264.82140.68347.098998.957610.0975917.0858.588317.06541008.85356.328744.8589177.27324987.4619.927185.607661.9852934.2411.9287257.7738438.1881373.9336.2719438082.36868467.373528800094.8210.831711911.0324.538540.73221539140.955285000010.8472640.6253675272.5119.794413430.33186.599448.51044334100.7532101622.5414.6568.2224173872.46234249.8167.690314.76036882400094.6839168.7591794.1887217477.212012881.5116.629605.29774009.018.8916386566.1710565000079687.8578539814.9015714147.26194.50314116849.9014344471.8435.6415.3234.7511029528.704311036.055.82457.78511593.6863896666711.3811.9058.5215.6525.1336.43785876311311943.101203966667545.78741635360.2723.6721069388.8922679029.6318094107.2498769000180736666720.31037403360019.2393593461149590000019.122052.274991.260210.944734.36915.0690917941143.641749526671.0913913.3288139.40233122000071.84914.99226.6314.6761860768.2234.1956.816199.27710.8989306.284919533313.66413.897.1484139.7381250403330.6427363666737.713328.481143.5135867.3319800666771.436613.9971107171.3419.029202.89455398324.92291252833338.391611888234.8139.310438.2082913.4010.1463137065291.04741774000071.402214.003727681.20859483.205.092106.68401.361.3637.85503286667388873333149.7098107886666748.573920.57749.8287849.21114.475840.812712.705137223333139.442824.794120495333335493033335.9740196.03140.53122670.2118122.3742.699942.634261.9786141.07857.0789374.8172757.1128113.5859257.886721.1000793.8161264.90140.73487.096399.279510.0652917.3158.720317.02751011.36355.577744.9310176.98225048.7919.854585.301162.02312940.0611.8874257.6865437.9078373.0636.2924437829.056872106.483521366795.1010.861311937.7224.528640.74791542414.025290666710.8482641.23953671132.6519.798813421.50186.927248.54274331008.9932117609.1814.643968.2516174068.70234493.0767.727214.75286881366794.6615168.7549793.81677218980.8712013496.3770031.3017.643050.45763301.0117.6516274465.5910206286779781.6694057409.4614906385.78170.9514637785.3913357787.1135.515.3836.812032750.263946891.516.15426.75411673.1163430000012.0411.9858.1415.1824.9335.29838747611871319.971180800000539.10639897113.4623.520689455.3922884629.7518545533.3699528000183200000020.76636566090019.3885463481120640000019.104352.323992.506210.797535.28215.2707416131168.881787000001.1081899.5957136.77833444000071.42615.29926.8414.7974863668.1433.6856.888196.67910.7588082.894999700013.6407.357.0929140.8233250700000.6527497000037.405728.561128.4635991.4119733000071.25914.0319106411.7419.083204.53035426214.88311266600038.368702202235.5639.480938.2233920.2910.0585137440336.06746029000071.277914.02827539.19857316.415.071106.95821.371.3737.75501110000386130000149.2727108000000048.436820.63519.8580649.26114.529440.74412.71137280000139.340524.767720450000005476800005.9629196.34140.87982675.008512342.855742.851562.0239140.37377.1143372.9577753.7418113.366257.748421.191792.2274265.03140.127.127499.374910.0557920.5358.662217.04391008.05356.909444.7637177.24425080.3319.927185.445261.90052940.2711.9023258.2072436.7701373.5536.3468436953.866858428.663518300095.0210.863511904.7324.640.63011541832.295283700010.8212640.08753674818.5919.833313405.8186.821448.46634335334.8732144757.814.656168.1941173848.98234217.5467.751414.74716886300094.632168.8102794.07917218080.712011361.1119.879611.34786282.1118.4516263996.16105450000107643.4992019395.1317583489.38199.52315179164.8614250093.7135.7513.9633.511686474.264022496.075.63419.85310869.4965041000011.2911.6455.0514.7624.2234.82222748211860591.651234500000545.19540991307.9523.5520249488.6123040229.0317931974.9296449000177860000020.90437207200018.8538288761152090000019.635650.908990.075611.086835.28314.8777103461170.311742800001.1189890.9448136.07133120000072.35314.97426.5314.5056452538.1333.5726.768195.82910.788466.544988400013.65410.897.1456139.7934249000000.6527173000037.67728.691131.5535640.3619551000071.836213.9192106715.8519.166202.50195446484.93171265200038.489779649236.9939.384137.9876912.6210.0774137476652.93747049000071.3514.013927731.16857324.895.053106.14671.371.3637.93504660000387460000150.276107380000048.251220.7149.8131848.95114.28240.572612.678136500000139.612724.715320456000005493900005.9947195.40140.57672661.2255122.742.764142.833762.1006140.45687.1102373.0279757.477113.6578257.423721.0894795.852263.85140.20147.122899.122210.0813916.7458.827916.99611007.46355.887744.9148176.61425006.7319.876385.485461.80532944.5411.9002258.5599436.9749372.7236.3849438280.166851573.23519500094.8810.863411939.4124.53640.73451543291.915276500010.8453641.60513667964.1519.8113428.64186.648248.47464337787.7232151217.0314.663268.1621173878.7234239.667.765714.74546883400094.6197168.7247794.03877218607.1212013602.0670043.78169504.38776708.7218.3515253621.53105630000OpenBenchmarking.org

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CryptodcEPYC 9124 2Pb20K40K60K80K100KSE +/- 23.57, N = 3107643.4979781.6679760.8279687.851. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: SemaphorescdEPYC 9124 2Pb20M40M60M80M100MSE +/- 972856.35, N = 1594057409.4692019395.1381543324.9278539814.901. (CXX) g++ options: -O2 -std=gnu99 -lc

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 1.6.2Clients Per Thread: 50 - Set To Get Ratio: 1:100dbEPYC 9124 2Pc4M8M12M16M20MSE +/- 219976.25, N = 1517583489.3815714147.2615294187.6414906385.781. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 12 - Input: Bosphorus 4KdbEPYC 9124 2Pc4080120160200SE +/- 1.55, N = 3199.52194.50188.70170.951. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 1.6.2Clients Per Thread: 20 - Set To Get Ratio: 1:10dcbEPYC 9124 2P3M6M9M12M15MSE +/- 184904.56, N = 1515179164.8614637785.3914116849.9013016172.381. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 1.6.2Clients Per Thread: 20 - Set To Get Ratio: 1:100EPYC 9124 2Pbdc3M6M9M12M15MSE +/- 164469.07, N = 1515292696.2014344471.8414250093.7113357787.111. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: yolov4-tinycEPYC 9124 2Pbd918273645SE +/- 0.25, N = 335.5035.6335.6435.75MIN: 34.27 / MAX: 39.94MIN: 34.31 / MAX: 39.09MIN: 34.1 / MAX: 39.46MIN: 34.32 / MAX: 45.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: shufflenet-v2dEPYC 9124 2Pbc48121620SE +/- 0.07, N = 313.9614.5915.3215.38MIN: 13.65 / MAX: 18.71MIN: 14.26 / MAX: 19.1MIN: 14.99 / MAX: 19.69MIN: 14.6 / MAX: 86.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: regnety_400mdEPYC 9124 2Pbc816243240SE +/- 0.30, N = 333.5033.7534.7536.80MIN: 33.26 / MAX: 37.92MIN: 33.36 / MAX: 38.06MIN: 32.99 / MAX: 39.74MIN: 33.81 / MAX: 451.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 1.6.2Clients Per Thread: 10 - Set To Get Ratio: 1:10EPYC 9124 2Pcdb3M6M9M12M15MSE +/- 28474.68, N = 312095393.8412032750.2611686474.2611029528.701. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: FutexEPYC 9124 2Pbdc900K1800K2700K3600K4500KSE +/- 13871.52, N = 34322676.824311036.054022496.073946891.511. (CXX) g++ options: -O2 -std=gnu99 -lc

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: blazefacedEPYC 9124 2Pbc246810SE +/- 0.12, N = 35.635.805.826.15MIN: 5.53 / MAX: 5.95MIN: 5.61 / MAX: 6.11MIN: 5.44 / MAX: 10.1MIN: 5.85 / MAX: 10.71. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 12 - Input: Bosphorus 1080pbEPYC 9124 2Pcd100200300400500SE +/- 5.17, N = 3457.79443.53426.75419.851. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix 3D MathEPYC 9124 2Pcbd3K6K9K12K15KSE +/- 31.82, N = 311756.3511673.1111593.6810869.491. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 16 - Buffer Length: 256 - Filter Length: 57dbcEPYC 9124 2P140M280M420M560M700MSE +/- 2539254.05, N = 36504100006389666676343000006084700001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v2-v2 - Model: mobilenet-v2dbEPYC 9124 2Pc3691215SE +/- 0.10, N = 311.2911.3811.6312.04MIN: 10.83 / MAX: 11.91MIN: 10.59 / MAX: 46.25MIN: 11.11 / MAX: 16.06MIN: 10.97 / MAX: 124.421. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU-v3-v3 - Model: mobilenet-v3EPYC 9124 2Pdbc3691215SE +/- 0.03, N = 311.2711.6411.9011.98MIN: 10.84 / MAX: 15.54MIN: 11.48 / MAX: 15.48MIN: 11.07 / MAX: 16.04MIN: 11.57 / MAX: 16.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vision_transformerdEPYC 9124 2Pcb1326395265SE +/- 1.11, N = 355.0557.2358.1458.52MIN: 54.24 / MAX: 72.37MIN: 56.51 / MAX: 61.47MIN: 56.8 / MAX: 80.65MIN: 56.14 / MAX: 303.741. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: efficientnet-b0dcEPYC 9124 2Pb48121620SE +/- 0.09, N = 314.7615.1815.6515.65MIN: 14.47 / MAX: 29.12MIN: 14.78 / MAX: 24.12MIN: 15.38 / MAX: 19.76MIN: 15.22 / MAX: 20.881. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: squeezenet_ssddEPYC 9124 2Pcb612182430SE +/- 0.09, N = 324.2224.9024.9325.13MIN: 23.8 / MAX: 28.97MIN: 24.53 / MAX: 28.94MIN: 24.54 / MAX: 29.12MIN: 24.57 / MAX: 30.591. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Water-layered HalfspaceEPYC 9124 2Pdcb816243240SE +/- 0.39, N = 534.6534.8235.3036.441. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 1.6.2Clients Per Thread: 10 - Set To Get Ratio: 1:100cdEPYC 9124 2Pb3M6M9M12M15MSE +/- 109712.47, N = 311871319.9711860591.6511365230.1411311943.101. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 32 - Buffer Length: 256 - Filter Length: 57dEPYC 9124 2Pbc300M600M900M1200M1500MSE +/- 10114401.17, N = 312345000001205600000120396666711808000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 13 - Input: Bosphorus 1080pbdcEPYC 9124 2P120240360480600SE +/- 4.09, N = 15545.79545.20539.11522.521. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Glibc C String FunctionsbEPYC 9124 2Pdc9M18M27M36M45MSE +/- 540521.68, N = 341635360.2741243730.2740991307.9539897113.461. (CXX) g++ options: -O2 -std=gnu99 -lc

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mobilenetEPYC 9124 2Pcdb612182430SE +/- 0.17, N = 322.9823.5023.5523.67MIN: 22.68 / MAX: 27.26MIN: 23.22 / MAX: 27.65MIN: 23.18 / MAX: 27.8MIN: 22.96 / MAX: 27.91. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: PipebcEPYC 9124 2Pd5M10M15M20M25MSE +/- 263402.50, N = 1521069388.8920689455.3920487917.5420249488.611. (CXX) g++ options: -O2 -std=gnu99 -lc

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 4.1.3Test: WritesdcbEPYC 9124 2P50K100K150K200K250KSE +/- 1083.03, N = 3230402228846226790222075

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: googlenetdEPYC 9124 2Pbc714212835SE +/- 0.14, N = 329.0329.4929.6329.75MIN: 27.97 / MAX: 36.49MIN: 28.8 / MAX: 33.78MIN: 28.91 / MAX: 37.12MIN: 29.08 / MAX: 39.281. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Context SwitchingcEPYC 9124 2Pbd4M8M12M16M20MSE +/- 108301.82, N = 318545533.3618252120.5218094107.2417931974.921. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 8 - Buffer Length: 256 - Filter Length: 512cbEPYC 9124 2Pd20M40M60M80M100MSE +/- 197811.36, N = 3995280009876900098607000964490001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 57cEPYC 9124 2Pbd400M800M1200M1600M2000MSE +/- 7859884.08, N = 318320000001809000000180736666717786000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Remhos

Remhos (REMap High-Order Solver) is a miniapp that solves the pure advection equations that are used to perform monotonic and conservative discontinuous field interpolation (remap) as part of the Eulerian phase in Arbitrary Lagrangian Eulerian (ALE) simulations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRemhos 1.0Test: Sample Remap ExamplebEPYC 9124 2Pcd510152025SE +/- 0.20, N = 320.3120.3620.7720.901. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

Kripke

Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.6EPYC 9124 2Pbdc80M160M240M320M400MSE +/- 2419934.15, N = 33762587003740336003720720003656609001. (CXX) g++ options: -O3 -fopenmp -ldl

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Homogeneous HalfspacedEPYC 9124 2Pbc510152025SE +/- 0.18, N = 718.8518.9219.2419.391. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi

nekRS

nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming on smaller systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgflops/rank, More Is BetternekRS 23.0Input: KershawdbEPYC 9124 2Pc2000M4000M6000M8000M10000MSE +/- 49802643.84, N = 3115209000001149590000011469600000112064000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-StreamcbEPYC 9124 2Pd510152025SE +/- 0.03, N = 319.1019.1219.3119.64

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-StreamcbEPYC 9124 2Pd1224364860SE +/- 0.07, N = 352.3252.2751.7650.91

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-StreamcEPYC 9124 2Pbd20406080100SE +/- 0.31, N = 392.5191.5391.2690.08

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-StreamcEPYC 9124 2Pbd3691215SE +/- 0.04, N = 310.8010.9110.9411.09

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigbEPYC 9124 2Pcd816243240SE +/- 0.49, N = 334.3735.1735.2835.28

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Tomographic ModeldEPYC 9124 2Pbc48121620SE +/- 0.10, N = 314.8814.9515.0715.271. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CloningEPYC 9124 2Pdcb30060090012001500SE +/- 10.16, N = 31173.141170.311168.881143.641. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 4 - Buffer Length: 256 - Filter Length: 57cEPYC 9124 2Pbd40M80M120M160M200MSE +/- 2041803.93, N = 151787000001781100001749526671742800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-StreambEPYC 9124 2Pcd0.25180.50360.75541.00721.259SE +/- 0.0016, N = 31.09131.10211.10811.1189

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-StreambEPYC 9124 2Pcd2004006008001000SE +/- 1.35, N = 3913.33904.42899.60890.94

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 8 - Input: Bosphorus 1080pbEPYC 9124 2Pcd306090120150SE +/- 0.50, N = 3139.40139.31136.78136.071. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 8 - Buffer Length: 256 - Filter Length: 57EPYC 9124 2Pcbd70M140M210M280M350MSE +/- 2830995.11, N = 33387900003344400003312200003312000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 8 - Input: Bosphorus 4KdbcEPYC 9124 2P1632486480SE +/- 0.21, N = 372.3571.8571.4370.771. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 4 - Input: Bosphorus 1080pcEPYC 9124 2Pbd48121620SE +/- 0.08, N = 315.3015.1514.9914.971. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet50EPYC 9124 2Pdbc612182430SE +/- 0.32, N = 326.2926.5326.6326.84MIN: 25.96 / MAX: 30.61MIN: 26.13 / MAX: 30.95MIN: 25.98 / MAX: 30.72MIN: 26.51 / MAX: 31.181. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Mount St. HelensdbEPYC 9124 2Pc48121620SE +/- 0.06, N = 314.5114.6814.7114.801. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: alexnetEPYC 9124 2Pdcb246810SE +/- 0.23, N = 38.068.138.148.22MIN: 7.76 / MAX: 12.3MIN: 7.95 / MAX: 8.44MIN: 7.75 / MAX: 12.48MIN: 7.66 / MAX: 12.541. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.9Video Input: Bosphorus 1080p - Video Preset: FasterbcEPYC 9124 2Pd816243240SE +/- 0.11, N = 334.2033.6933.6633.571. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.9Video Input: Bosphorus 4K - Video Preset: FastcbdEPYC 9124 2P246810SE +/- 0.051, N = 36.8886.8166.7686.7631. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 13 - Input: Bosphorus 4KEPYC 9124 2Pbcd4080120160200SE +/- 1.95, N = 3199.36199.28196.68195.831. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: mnasnetdcEPYC 9124 2Pb3691215SE +/- 0.11, N = 310.7010.7510.8910.89MIN: 10.22 / MAX: 14.66MIN: 10.55 / MAX: 20.58MIN: 10.64 / MAX: 15.33MIN: 10.59 / MAX: 11.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CPU StressEPYC 9124 2Pbdc20K40K60K80K100KSE +/- 346.97, N = 389602.5189306.2888466.5488082.891. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 4 - Buffer Length: 256 - Filter Length: 512cdEPYC 9124 2Pb11M22M33M44M55MSE +/- 220678.75, N = 3499970004988400049627000491953331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: resnet18EPYC 9124 2Pcdb48121620SE +/- 0.18, N = 313.5313.6013.6513.66MIN: 13.23 / MAX: 13.98MIN: 13.23 / MAX: 17.34MIN: 13.29 / MAX: 15.11MIN: 13.02 / MAX: 18.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVL TreebEPYC 9124 2Pdc90180270360450SE +/- 5.37, N = 3413.89410.95410.89407.351. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-StreamcdbEPYC 9124 2P246810SE +/- 0.0129, N = 37.09297.14567.14847.2056

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-StreamcdbEPYC 9124 2P306090120150SE +/- 0.26, N = 3140.82139.79139.74138.62

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 2 - Buffer Length: 256 - Filter Length: 512cbdEPYC 9124 2P5M10M15M20M25MSE +/- 138974.02, N = 3250700002504033324900000246800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.0Run: RTLightmap.hdr.4096x4096 - Device: CPU-OnlydcEPYC 9124 2Pb0.14630.29260.43890.58520.7315SE +/- 0.00, N = 30.650.650.650.64

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 8 - Buffer Length: 256 - Filter Length: 32EPYC 9124 2Pcbd60M120M180M240M300MSE +/- 1505958.54, N = 32758100002749700002736366672717300001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer - Model: CrownEPYC 9124 2Pbdc918273645SE +/- 0.01, N = 337.9737.7137.6837.41MIN: 37.52 / MAX: 39.47MIN: 37.2 / MAX: 38.96MIN: 37.26 / MAX: 39.44MIN: 36.79 / MAX: 38.87

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: vgg16EPYC 9124 2Pbcd714212835SE +/- 0.61, N = 328.3128.4828.5628.57MIN: 27.15 / MAX: 33.46MIN: 26.51 / MAX: 33.89MIN: 25.87 / MAX: 151.12MIN: 27.98 / MAX: 32.461. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: MMAPbEPYC 9124 2Pdc2004006008001000SE +/- 10.51, N = 31143.511142.011131.551128.461. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Mixed SchedulercbdEPYC 9124 2P8K16K24K32K40KSE +/- 132.19, N = 335991.4135867.3335640.3635532.561. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 16 - Buffer Length: 256 - Filter Length: 512bEPYC 9124 2Pcd40M80M120M160M200MSE +/- 401718.53, N = 31980066671977900001973300001955100001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamEPYC 9124 2Pcbd1632486480SE +/- 0.25, N = 370.9671.2671.4471.84

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-StreamEPYC 9124 2Pcbd48121620SE +/- 0.05, N = 314.0914.0314.0013.92

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector Floating PointEPYC 9124 2Pbdc20K40K60K80K100KSE +/- 414.14, N = 3107730.40107171.34106715.85106411.741. (CXX) g++ options: -O2 -std=gnu99 -lc

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.9Video Input: Bosphorus 1080p - Video Preset: FastdcbEPYC 9124 2P510152025SE +/- 0.03, N = 319.1719.0819.0318.941. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-StreamEPYC 9124 2Pcbd4080120160200SE +/- 0.98, N = 3204.81204.53202.89202.50

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.36VGR Performance MetricEPYC 9124 2Pdcb120K240K360K480K600K5459525446485426215398321. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-StreamEPYC 9124 2Pcbd1.10962.21923.32884.43845.548SE +/- 0.0237, N = 34.87724.88314.92294.9317

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 512cdEPYC 9124 2Pb3M6M9M12M15MSE +/- 143667.83, N = 3126660001265200012637000125283331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSPECFEM3D 4.0Model: Layered HalfspaceEPYC 9124 2Pcbd918273645SE +/- 0.26, N = 338.1038.3738.3938.491. (F9X) gfortran options: -O2 -fopenmp -std=f2003 -fimplicit-none -fmax-errors=10 -pedantic -pedantic-errors -O3 -finline-functions -lmpi_usempif08 -lmpi_mpifh -lmpi

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AtomicdcEPYC 9124 2Pb50100150200250SE +/- 0.55, N = 3236.99235.56235.35234.811. (CXX) g++ options: -O2 -std=gnu99 -lc

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: CrowncdbEPYC 9124 2P918273645SE +/- 0.04, N = 339.4839.3839.3139.12MIN: 39 / MAX: 40.94MIN: 38.86 / MAX: 40.97MIN: 38.66 / MAX: 40.8MIN: 38.49 / MAX: 40.34

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer - Model: Asian Dragon ObjEPYC 9124 2Pcbd918273645SE +/- 0.02, N = 338.3438.2238.2137.99MIN: 37.99 / MAX: 38.77MIN: 37.93 / MAX: 38.58MIN: 37.93 / MAX: 38.75MIN: 37.71 / MAX: 38.4

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: MEMFDEPYC 9124 2Pcbd2004006008001000SE +/- 2.16, N = 3920.90920.29913.40912.621. (CXX) g++ options: -O2 -std=gnu99 -lc

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/ao/real_timebEPYC 9124 2Pdc3691215SE +/- 0.01, N = 310.1510.1010.0810.06

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: MallocdcbEPYC 9124 2P30M60M90M120M150MSE +/- 86125.58, N = 3137476652.93137440336.06137065291.04136350929.271. (CXX) g++ options: -O2 -std=gnu99 -lc

nekRS

nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming on smaller systems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgflops/rank, More Is BetternekRS 23.0Input: TurboPipe PeriodicdcbEPYC 9124 2P1600M3200M4800M6400M8000MSE +/- 22116600.85, N = 374704900007460290000741774000074108500001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-StreamEPYC 9124 2Pcdb1632486480SE +/- 0.19, N = 370.8371.2871.3571.40

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-StreamEPYC 9124 2Pcdb48121620SE +/- 0.04, N = 314.1214.0314.0114.00

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Function CallEPYC 9124 2Pdbc6K12K18K24K30KSE +/- 27.79, N = 327757.9527731.1627681.2027539.191. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: SENDFILEbdcEPYC 9124 2P200K400K600K800K1000KSE +/- 459.45, N = 3859483.20857324.89857316.41852899.851. (CXX) g++ options: -O2 -std=gnu99 -lc

SVT-AV1

This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.6Encoder Mode: Preset 4 - Input: Bosphorus 4KbcEPYC 9124 2Pd1.14572.29143.43714.58285.7285SE +/- 0.021, N = 35.0925.0715.0685.0531. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd20406080100SE +/- 0.09, N = 3106.96106.79106.68106.15

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.0Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-OnlydcEPYC 9124 2Pb0.30830.61660.92491.23321.5415SE +/- 0.00, N = 31.371.371.371.36

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 2.0Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-OnlycdbEPYC 9124 2P0.30830.61660.92491.23321.5415SE +/- 0.00, N = 31.371.361.361.36

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: BMW27 - Compute: CPU-OnlyEPYC 9124 2Pcbd918273645SE +/- 0.02, N = 337.6637.7537.8537.93

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 512dbcEPYC 9124 2P110M220M330M440M550MSE +/- 219266.45, N = 35046600005032866675011100005010700001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 32 - Buffer Length: 256 - Filter Length: 512bEPYC 9124 2Pdc80M160M240M320M400MSE +/- 468946.81, N = 33888733333883300003874600003861300001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd306090120150SE +/- 0.18, N = 3149.27149.55149.71150.28

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 32 - Buffer Length: 256 - Filter Length: 32EPYC 9124 2Pcbd200M400M600M800M1000MSE +/- 1146492.23, N = 310810000001080000000107886666710738000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-StreambcEPYC 9124 2Pd1122334455SE +/- 0.15, N = 348.5748.4448.3248.25

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-StreambcEPYC 9124 2Pd510152025SE +/- 0.06, N = 320.5820.6420.6920.71

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/scivis/real_timeEPYC 9124 2Pcbd3691215SE +/- 0.01904, N = 39.876829.858069.828789.81318

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Fishy Cat - Compute: CPU-OnlydbEPYC 9124 2Pc1122334455SE +/- 0.13, N = 348.9549.2149.2549.26

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pcbd306090120150SE +/- 0.13, N = 3115.00114.53114.48114.28

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian Dragon ObjEPYC 9124 2Pbcd918273645SE +/- 0.04, N = 340.8240.8140.7440.57MIN: 40.51 / MAX: 41.28MIN: 40.43 / MAX: 41.85MIN: 40.34 / MAX: 41.4MIN: 40.28 / MAX: 41.19

VVenC

VVenC is the Fraunhofer Versatile Video Encoder as a fast/efficient H.266/VVC encoder. The vvenc encoder makes use of SIMD Everywhere (SIMDe). The vvenc software is published under the Clear BSD License. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.9Video Input: Bosphorus 4K - Video Preset: FasterEPYC 9124 2Pcbd3691215SE +/- 0.01, N = 312.7612.7112.7112.681. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 4 - Buffer Length: 256 - Filter Length: 32cbEPYC 9124 2Pd30M60M90M120M150MSE +/- 301182.85, N = 31372800001372233331368900001365000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pcbd306090120150SE +/- 0.13, N = 3138.82139.34139.44139.61

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pbcd612182430SE +/- 0.04, N = 324.8624.7924.7724.72

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 32EPYC 9124 2Pbdc400M800M1200M1600M2000MSE +/- 1319511.69, N = 320562000002049533333204560000020450000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 16 - Buffer Length: 256 - Filter Length: 32EPYC 9124 2Pdbc120M240M360M480M600MSE +/- 1392172.56, N = 35506100005493900005493033335476800001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd1.34882.69764.04645.39526.744SE +/- 0.0057, N = 35.96295.97155.97405.9947

Laghos

Laghos (LAGrangian High-Order Solver) is a miniapp that solves the time-dependent Euler equations of compressible gas dynamics in a moving Lagrangian frame using unstructured high-order finite element spatial discretization and explicit high-order time-stepping. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMajor Kernels Total Rate, More Is BetterLaghos 3.1Test: Triple Point ProblemEPYC 9124 2Pcbd4080120160200SE +/- 0.76, N = 3196.44196.34196.03195.401. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pbdc306090120150SE +/- 0.10, N = 3140.14140.53140.58140.88

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd6001200180024003000SE +/- 2.37, N = 32675.012673.222670.212661.23

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Pabellon Barcelona - Compute: CPU-OnlybEPYC 9124 2Pdc306090120150SE +/- 0.41, N = 3122.37122.49122.70123.00

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer - Model: Asian DragoncdbEPYC 9124 2P1020304050SE +/- 0.04, N = 342.8642.7642.7042.64MIN: 42.63 / MAX: 43.37MIN: 42.48 / MAX: 43.18MIN: 42.4 / MAX: 43.23MIN: 42.35 / MAX: 43.05

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-StreambEPYC 9124 2Pdc1020304050SE +/- 0.03, N = 342.6342.7242.8342.85

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pbcd1428425670SE +/- 0.05, N = 361.7961.9862.0262.10

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Synchronous Single-StreambEPYC 9124 2Pdc306090120150SE +/- 0.14, N = 3141.08140.82140.46140.37

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Synchronous Single-StreambEPYC 9124 2Pdc246810SE +/- 0.0070, N = 37.07897.09187.11027.1143

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-StreambEPYC 9124 2Pdc80160240320400SE +/- 0.27, N = 3374.82374.03373.03372.96

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-StreamdbEPYC 9124 2Pc160320480640800SE +/- 1.99, N = 3757.48757.11754.64753.74

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pdbc306090120150SE +/- 0.06, N = 3113.92113.66113.59113.37

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pbcd60120180240300SE +/- 0.30, N = 3258.67257.89257.75257.42

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-StreamdbEPYC 9124 2Pc510152025SE +/- 0.06, N = 321.0921.1021.1721.19

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd2004006008001000SE +/- 0.18, N = 3792.23793.78793.82795.85

Laghos

Laghos (LAGrangian High-Order Solver) is a miniapp that solves the time-dependent Euler equations of compressible gas dynamics in a moving Lagrangian frame using unstructured high-order finite element spatial discretization and explicit high-order time-stepping. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMajor Kernels Total Rate, More Is BetterLaghos 3.1Test: Sedov Blast Wave, ube_922_hex.meshcbEPYC 9124 2Pd60120180240300SE +/- 0.70, N = 3265.03264.90264.82263.851. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-StreambEPYC 9124 2Pdc306090120150SE +/- 0.05, N = 3140.73140.68140.20140.12

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-StreambEPYC 9124 2Pdc246810SE +/- 0.0029, N = 37.09637.09897.12287.1274

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-StreamcbdEPYC 9124 2P20406080100SE +/- 0.23, N = 399.3799.2899.1298.96

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-StreamcbdEPYC 9124 2P3691215SE +/- 0.02, N = 310.0610.0710.0810.10

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Glibc Qsort Data SortingcbEPYC 9124 2Pd2004006008001000SE +/- 0.71, N = 3920.53917.31917.08916.741. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-StreamEPYC 9124 2Pcbd1326395265SE +/- 0.15, N = 358.5958.6658.7258.83

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Synchronous Single-StreamEPYC 9124 2Pcbd48121620SE +/- 0.04, N = 317.0717.0417.0317.00

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: ForkingbEPYC 9124 2Pcd2004006008001000SE +/- 5.44, N = 31011.361008.851008.051007.461. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pdb80160240320400SE +/- 0.15, N = 3356.91356.33355.89355.58

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pdb1020304050SE +/- 0.01, N = 344.7644.8644.9144.93

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/pathtracer/real_timeEPYC 9124 2Pcbd4080120160200SE +/- 0.36, N = 3177.27177.24176.98176.61

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector ShufflecbdEPYC 9124 2P5K10K15K20K25KSE +/- 29.34, N = 325080.3325048.7925006.7324987.461. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pdb510152025SE +/- 0.02, N = 319.9319.9319.8819.85

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pdcb20406080100SE +/- 0.04, N = 385.6185.4985.4585.30

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamdcEPYC 9124 2Pb1428425670SE +/- 0.16, N = 361.8161.9061.9962.02

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: ZlibdcbEPYC 9124 2P6001200180024003000SE +/- 1.86, N = 32944.542940.272940.062934.241. (CXX) g++ options: -O2 -std=gnu99 -lc

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_timeEPYC 9124 2Pcdb3691215SE +/- 0.00, N = 311.9311.9011.9011.89

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-StreamdcEPYC 9124 2Pb60120180240300SE +/- 0.66, N = 3258.56258.21257.77257.69

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamcdbEPYC 9124 2P90180270360450SE +/- 0.15, N = 3436.77436.97437.91438.19

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Barbershop - Compute: CPU-OnlydbcEPYC 9124 2P80160240320400SE +/- 0.81, N = 3372.72373.06373.55373.93

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-StreamdcbEPYC 9124 2P816243240SE +/- 0.02, N = 336.3836.3536.2936.27

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: MutexdEPYC 9124 2Pbc90K180K270K360K450KSE +/- 421.09, N = 3438280.16438082.30437829.05436953.861. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: System V Message PassingbEPYC 9124 2Pcd1.5M3M4.5M6M7.5MSE +/- 5858.93, N = 36872106.486868467.376858428.666851573.201. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 32EPYC 9124 2Pbdc8M16M24M32M40MSE +/- 26660.42, N = 3352880003521366735195000351830001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.6Blend File: Classroom - Compute: CPU-OnlyEPYC 9124 2Pdcb20406080100SE +/- 0.08, N = 394.8294.8895.0295.10

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/ao/real_timecdbEPYC 9124 2P3691215SE +/- 0.00, N = 310.8610.8610.8610.83

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Floating PointdbEPYC 9124 2Pc3K6K9K12K15KSE +/- 1.29, N = 311939.4111937.7211911.0311904.731. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-StreamcEPYC 9124 2Pdb612182430SE +/- 0.01, N = 324.6024.5424.5424.53

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-StreamcEPYC 9124 2Pdb918273645SE +/- 0.02, N = 340.6340.7340.7340.75

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Wide Vector MathdbcEPYC 9124 2P300K600K900K1200K1500KSE +/- 1152.35, N = 31543291.911542414.021541832.291539140.951. (CXX) g++ options: -O2 -std=gnu99 -lc

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 1 - Buffer Length: 256 - Filter Length: 57bEPYC 9124 2Pcd11M22M33M44M55MSE +/- 24037.01, N = 3529066675285000052837000527650001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OSPRay

Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems Per Second, More Is BetterOSPRay 2.12Benchmark: particle_volume/scivis/real_timebEPYC 9124 2Pdc3691215SE +/- 0.01, N = 310.8510.8510.8510.82

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: BERT-Large, NLP Question Answering - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd140280420560700SE +/- 0.58, N = 3640.09640.63641.24641.61

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: AVX-512 VNNIEPYC 9124 2Pcbd800K1600K2400K3200K4000KSE +/- 4464.99, N = 33675272.513674818.593671132.653667964.151. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-StreamcdbEPYC 9124 2P510152025SE +/- 0.06, N = 319.8319.8119.8019.79

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Memory CopyingEPYC 9124 2Pdbc3K6K9K12K15KSE +/- 2.68, N = 313430.3313428.6413421.5013405.801. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-StreamEPYC 9124 2Pdcb4080120160200SE +/- 0.01, N = 3186.60186.65186.82186.93

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.1Binary: Pathtracer ISPC - Model: Asian DragonbEPYC 9124 2Pdc1122334455SE +/- 0.05, N = 348.5448.5148.4748.47MIN: 48.1 / MAX: 49.38MIN: 48.2 / MAX: 49.13MIN: 48.1 / MAX: 49.28MIN: 48.15 / MAX: 49.27

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: PolldcEPYC 9124 2Pb900K1800K2700K3600K4500KSE +/- 1071.05, N = 34337787.724335334.874334100.754331008.991. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Fused Multiply-AdddcbEPYC 9124 2P7M14M21M28M35MSE +/- 5730.93, N = 332151217.0332144757.8032117609.1832101622.541. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-StreambEPYC 9124 2Pcd48121620SE +/- 0.00, N = 314.6414.6514.6614.66

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-StreambEPYC 9124 2Pcd1530456075SE +/- 0.01, N = 368.2568.2268.1968.16

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Matrix MathbdEPYC 9124 2Pc40K80K120K160K200KSE +/- 1.07, N = 3174068.70173878.70173872.46173848.981. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Vector MathbEPYC 9124 2Pdc50K100K150K200K250KSE +/- 5.36, N = 3234493.07234249.81234239.60234217.541. (CXX) g++ options: -O2 -std=gnu99 -lc

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-StreamdcbEPYC 9124 2P1530456075SE +/- 0.02, N = 367.7767.7567.7367.69

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-StreamdcbEPYC 9124 2P48121620SE +/- 0.00, N = 314.7514.7514.7514.76

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 2 - Buffer Length: 256 - Filter Length: 32cdEPYC 9124 2Pb15M30M45M60M75MSE +/- 21231.53, N = 3688630006883400068824000688136671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-StreamdcbEPYC 9124 2P20406080100SE +/- 0.10, N = 394.6294.6394.6694.68

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.5Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-StreamcEPYC 9124 2Pbd4080120160200SE +/- 0.19, N = 3168.81168.76168.75168.72

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.5Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-StreambdcEPYC 9124 2P2004006008001000SE +/- 0.67, N = 3793.82794.04794.08794.19

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: HashbdcEPYC 9124 2P1.5M3M4.5M6M7.5MSE +/- 1736.87, N = 37218980.877218607.127218080.707217477.201. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: x86_64 RdRanddbEPYC 9124 2Pc3M6M9M12M15MSE +/- 82.64, N = 312013602.0612013496.3712012881.5112011361.111. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Pthreaddb15K30K45K60K75KSE +/- 71.08, N = 270043.7870031.301. (CXX) g++ options: -O2 -std=gnu99 -lc

Test: Pthread

c: The test quit with a non-zero exit status.

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20230517Target: CPU - Model: FastestDetdEPYC 9124 2Pbc510152025SE +/- 1.24, N = 316.0016.6217.6419.87MIN: 15.86 / MAX: 20.49MIN: 16.27 / MAX: 18.12MIN: 15.65 / MAX: 24.31MIN: 19.45 / MAX: 29.361. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: Socket ActivitycEPYC 9124 2Pdb2K4K6K8K10KSE +/- 941.54, N = 159611.349605.299504.383050.451. (CXX) g++ options: -O2 -std=gnu99 -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: CPU CachecdEPYC 9124 2Pb200K400K600K800K1000KSE +/- 13475.92, N = 15786282.11776708.72774009.01763301.011. (CXX) g++ options: -O2 -std=gnu99 -lc

Test: IO_uring

EPYC 9124 2P: The test run did not produce a result.

b: The test run did not produce a result.

c: The test run did not produce a result.

d: The test run did not produce a result.

Test: Pthread

EPYC 9124 2P: The test quit with a non-zero exit status.

c: The test quit with a non-zero exit status.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.16.04Test: NUMAcdbEPYC 9124 2P510152025SE +/- 0.81, N = 1218.4518.3517.658.891. (CXX) g++ options: -O2 -std=gnu99 -lc

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

Clients Per Thread: 60 - Set To Get Ratio: 1:100

EPYC 9124 2P: The test run did not produce a result. E: Connection error: Connection reset by peer

b: The test run did not produce a result. E: Connection error: Connection reset by peer

c: The test run did not produce a result. E: Connection error: Connection reset by peer

d: The test run did not produce a result. E: Connection error: Connection reset by peer

Clients Per Thread: 60 - Set To Get Ratio: 1:10

EPYC 9124 2P: The test run did not produce a result. E: Connection error: Connection reset by peer

b: The test run did not produce a result. E: Connection error: Connection reset by peer

c: The test run did not produce a result. E: Connection error: Connection reset by peer

d: The test run did not produce a result. E: Connection error: Connection reset by peer

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 1.6.2Clients Per Thread: 50 - Set To Get Ratio: 1:10EPYC 9124 2Pbcd4M8M12M16M20MSE +/- 323344.22, N = 1516386566.1716274465.5916263996.1615253621.531. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 2 - Buffer Length: 256 - Filter Length: 57EPYC 9124 2Pdcb20M40M60M80M100MSE +/- 2143861.14, N = 151056500001056300001054500001020628671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

Build: allmodconfig

EPYC 9124 2P: The test quit with a non-zero exit status.

b: The test quit with a non-zero exit status.

c: The test quit with a non-zero exit status.

d: The test quit with a non-zero exit status.

188 Results Shown

Stress-NG:
  Crypto
  Semaphores
Dragonflydb
SVT-AV1
Dragonflydb:
  20 - 1:10
  20 - 1:100
NCNN:
  CPU - yolov4-tiny
  CPU - shufflenet-v2
  CPU - regnety_400m
Dragonflydb
Stress-NG
NCNN
SVT-AV1
Stress-NG
Liquid-DSP
NCNN:
  CPU-v2-v2 - mobilenet-v2
  CPU-v3-v3 - mobilenet-v3
  CPU - vision_transformer
  CPU - efficientnet-b0
  CPU - squeezenet_ssd
SPECFEM3D
Dragonflydb
Liquid-DSP
SVT-AV1
Stress-NG
NCNN
Stress-NG
Apache Cassandra
NCNN
Stress-NG
Liquid-DSP:
  8 - 256 - 512
  64 - 256 - 57
Remhos
Kripke
SPECFEM3D
nekRS
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Synchronous Single-Stream:
    ms/batch
    items/sec
  BERT-Large, NLP Question Answering, Sparse INT8 - Synchronous Single-Stream:
    items/sec
    ms/batch
Timed Linux Kernel Compilation
SPECFEM3D
Stress-NG
Liquid-DSP
Neural Magic DeepSparse:
  ResNet-50, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
SVT-AV1
Liquid-DSP
SVT-AV1:
  Preset 8 - Bosphorus 4K
  Preset 4 - Bosphorus 1080p
NCNN
SPECFEM3D
NCNN
VVenC:
  Bosphorus 1080p - Faster
  Bosphorus 4K - Fast
SVT-AV1
NCNN
Stress-NG
Liquid-DSP
NCNN
Stress-NG
Neural Magic DeepSparse:
  NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Synchronous Single-Stream:
    ms/batch
    items/sec
Liquid-DSP
Intel Open Image Denoise
Liquid-DSP
Embree
NCNN
Stress-NG:
  MMAP
  Mixed Scheduler
Liquid-DSP
Neural Magic DeepSparse:
  NLP Document Classification, oBERT base uncased on IMDB - Synchronous Single-Stream:
    ms/batch
    items/sec
Stress-NG
VVenC
Neural Magic DeepSparse
BRL-CAD
Neural Magic DeepSparse
Liquid-DSP
SPECFEM3D
Stress-NG
Embree:
  Pathtracer ISPC - Crown
  Pathtracer - Asian Dragon Obj
Stress-NG
OSPRay
Stress-NG
nekRS
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Synchronous Single-Stream:
    ms/batch
    items/sec
Stress-NG:
  Function Call
  SENDFILE
SVT-AV1
Neural Magic DeepSparse
Intel Open Image Denoise:
  RT.ldr_alb_nrm.3840x2160 - CPU-Only
  RT.hdr_alb_nrm.3840x2160 - CPU-Only
Blender
Liquid-DSP:
  64 - 256 - 512
  32 - 256 - 512
Neural Magic DeepSparse
Liquid-DSP
Neural Magic DeepSparse:
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Synchronous Single-Stream:
    items/sec
    ms/batch
OSPRay
Blender
Neural Magic DeepSparse
Embree
VVenC
Liquid-DSP
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream
  BERT-Large, NLP Question Answering - Asynchronous Multi-Stream
Liquid-DSP:
  64 - 256 - 32
  16 - 256 - 32
Neural Magic DeepSparse
Laghos
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream
  ResNet-50, Sparse INT8 - Asynchronous Multi-Stream
Blender
Embree
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream
  ResNet-50, Baseline - Asynchronous Multi-Stream
  ResNet-50, Baseline - Synchronous Single-Stream
  ResNet-50, Baseline - Synchronous Single-Stream
  BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream
  CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream
  ResNet-50, Baseline - Asynchronous Multi-Stream
  NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream
  NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream
Laghos
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Synchronous Single-Stream:
    items/sec
    ms/batch
  NLP Text Classification, DistilBERT mnli - Synchronous Single-Stream:
    items/sec
    ms/batch
Stress-NG
Neural Magic DeepSparse:
  BERT-Large, NLP Question Answering - Synchronous Single-Stream:
    ms/batch
    items/sec
Stress-NG
Neural Magic DeepSparse:
  NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Asynchronous Multi-Stream:
    items/sec
    ms/batch
OSPRay
Stress-NG
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream
Stress-NG
OSPRay
Neural Magic DeepSparse:
  CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream
  CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream
Blender
Neural Magic DeepSparse
Stress-NG:
  Mutex
  System V Message Passing
Liquid-DSP
Blender
OSPRay
Stress-NG
Neural Magic DeepSparse:
  CV Segmentation, 90% Pruned YOLACT Pruned - Synchronous Single-Stream:
    items/sec
    ms/batch
Stress-NG
Liquid-DSP
OSPRay
Neural Magic DeepSparse
Stress-NG
Neural Magic DeepSparse
Stress-NG
Neural Magic DeepSparse
Embree
Stress-NG:
  Poll
  Fused Multiply-Add
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO, Sparse INT8 - Synchronous Single-Stream:
    ms/batch
    items/sec
Stress-NG:
  Matrix Math
  Vector Math
Neural Magic DeepSparse:
  CV Detection, YOLOv5s COCO - Synchronous Single-Stream:
    items/sec
    ms/batch
Liquid-DSP
Neural Magic DeepSparse:
  NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream:
    ms/batch
    items/sec
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream:
    ms/batch
Stress-NG:
  Hash
  x86_64 RdRand
  Pthread
NCNN
Stress-NG:
  Socket Activity
  CPU Cache
  NUMA
Dragonflydb
Liquid-DSP