AMD EPYC Genoa Memory Scaling Benchmarks by Michael Larabel for a future article. 12c: Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 22.10, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0 + Clang 15.0.2-1, File-System: ext4, Screen Resolution: 1920x1080 10c: Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1264GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 22.10, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0 + Clang 15.0.2-1, File-System: ext4, Screen Resolution: 1920x1080 8c: Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1008GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 22.10, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0 + Clang 15.0.2-1, File-System: ext4, Screen Resolution: 1920x1080 6c: Processor: 2 x AMD EPYC 9654 96-Core @ 2.40GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 800GB INTEL SSDPF21Q800GB, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 22.10, Kernel: 6.1.0-phx (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server 1.21.1.4, Vulkan: 1.3.224, Compiler: GCC 12.2.0 + Clang 15.0.2-1, File-System: ext4, Screen Resolution: 1920x1080 Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d Seconds < Lower Is Better 12c . 125.53 |======================= 10c . 146.29 |=========================== 8c .. 270.09 |================================================== 6c .. 348.88 |================================================================= OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time Seconds < Lower Is Better 12c . 109.54 |=============================== 10c . 117.94 |================================== 8c .. 166.15 |=============================================== 6c .. 227.90 |================================================================= RELION 3.1.1 Test: Basic - Device: CPU Seconds < Lower Is Better 12c . 128.10 |================================ 10c . 151.40 |====================================== 8c .. 221.34 |======================================================== 6c .. 258.50 |================================================================= WRF 4.2.2 Input: conus 2.5km Seconds < Lower Is Better 12c . 4070.19 |=================================== 10c . 4563.18 |======================================= 8c .. 6551.88 |======================================================== 6c .. 7432.66 |================================================================ NAS Parallel Benchmarks 3.4 Test / Class: MG.C Total Mop/s > Higher Is Better 12c . 209846.76 |============================================================== 10c . 177097.42 |==================================================== 8c .. 153458.78 |============================================= 6c .. 117733.57 |=================================== NAS Parallel Benchmarks 3.4 Test / Class: SP.C Total Mop/s > Higher Is Better 12c . 260471.50 |============================================================== 10c . 239496.01 |========================================================= 8c .. 208535.23 |================================================== 6c .. 167474.70 |======================================== Graph500 3.0 Scale: 26 sssp median_TEPS > Higher Is Better 12c . 565152000 |============================================================= 10c . 574018000 |============================================================== 8c .. 531854000 |========================================================= 6c .. 392496000 |========================================== Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 111.89 |================================================ 10c . 113.41 |================================================= 8c .. 123.86 |===================================================== 6c .. 150.92 |================================================================= Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 856.02 |================================================================= 10c . 844.43 |================================================================ 8c .. 773.07 |=========================================================== 6c .. 635.02 |================================================ Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 125.72 |================================================= 10c . 128.92 |================================================== 8c .. 135.62 |===================================================== 6c .. 166.43 |================================================================= Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 761.49 |================================================================= 10c . 742.80 |=============================================================== 8c .. 705.71 |============================================================ 6c .. 575.75 |================================================= nekRS 22.0 Input: TurboPipe Periodic FLOP/s > Higher Is Better 12c . 821462000000 |=========================================================== 10c . 786258000000 |======================================================== 8c .. 740247000000 |===================================================== 6c .. 659554333333 |=============================================== OpenVINO 2022.3 Model: Machine Translation EN To DE FP16 - Device: CPU ms < Lower Is Better 12c . 49.98 |======================================================== 10c . 51.29 |========================================================== 8c .. 54.80 |============================================================== 6c .. 58.67 |================================================================== OpenVINO 2022.3 Model: Machine Translation EN To DE FP16 - Device: CPU FPS > Higher Is Better 12c . 959.16 |================================================================= 10c . 934.71 |=============================================================== 8c .. 875.39 |=========================================================== 6c .. 817.27 |======================================================= TensorFlow 2.10 Device: CPU - Batch Size: 256 - Model: ResNet-50 images/sec > Higher Is Better 12c . 109.13 |================================================================= 10c . 105.91 |=============================================================== 8c .. 105.01 |=============================================================== 6c .. 95.67 |========================================================= GPAW 22.1 Input: Carbon Nanotube Seconds < Lower Is Better 12c . 23.15 |========================================================== 10c . 23.37 |=========================================================== 8c .. 24.60 |============================================================== 6c .. 26.31 |================================================================== NAS Parallel Benchmarks 3.4 Test / Class: CG.C Total Mop/s > Higher Is Better 12c . 80225.01 |============================================================== 10c . 81179.00 |=============================================================== 8c .. 79784.15 |============================================================== 6c .. 71662.28 |======================================================== 7-Zip Compression 22.01 Test: Compression Rating MIPS > Higher Is Better 12c . 923176 |================================================================= 10c . 893433 |=============================================================== 8c .. 879430 |============================================================== 6c .. 824926 |========================================================== OpenVKL 1.3.1 Benchmark: vklBenchmark ISPC Items / Sec > Higher Is Better 12c . 1325 |=================================================================== 10c . 1317 |=================================================================== 8c .. 1325 |=================================================================== 6c .. 1212 |============================================================= oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better 12c . 2275.86 |=========================================================== 10c . 2325.71 |============================================================ 8c .. 2371.78 |============================================================= 6c .. 2471.57 |================================================================ NAS Parallel Benchmarks 3.4 Test / Class: LU.C Total Mop/s > Higher Is Better 12c . 489164.65 |============================================================== 10c . 489995.20 |============================================================== 8c .. 466769.54 |=========================================================== 6c .. 454360.62 |========================================================= Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 Images / Sec > Higher Is Better 12c . 1.65 |=================================================================== 10c . 1.63 |================================================================== 8c .. 1.64 |=================================================================== 6c .. 1.54 |=============================================================== Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 Images / Sec > Higher Is Better 12c . 3.52 |=================================================================== 10c . 3.44 |================================================================= 8c .. 3.47 |================================================================== 6c .. 3.29 |=============================================================== Rodinia 3.1 Test: OpenMP Streamcluster Seconds < Lower Is Better 12c . 6.001 |============================================================== 10c . 6.285 |================================================================= 8c .. 6.018 |============================================================== 6c .. 6.409 |================================================================== oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 12c . 2344.29 |============================================================= 10c . 2438.00 |=============================================================== 8c .. 2375.45 |============================================================= 6c .. 2479.62 |================================================================ Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Frames Per Second > Higher Is Better 12c . 73.44 |================================================================ 10c . 75.35 |================================================================== 8c .. 73.04 |================================================================ 6c .. 71.41 |=============================================================== Apache Cassandra 4.0 Test: Writes Op/s > Higher Is Better 12c . 251793 |================================================================= 10c . 243603 |=============================================================== 8c .. 240854 |============================================================== 6c .. 246882 |================================================================ GROMACS 2022.1 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day > Higher Is Better 12c . 18.71 |================================================================== 10c . 18.68 |================================================================== 8c .. 18.68 |================================================================== 6c .. 17.94 |=============================================================== oneDNN 3.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 12c . 0.446930 |============================================================ 10c . 0.463454 |=============================================================== 8c .. 0.465796 |=============================================================== 6c .. 0.465059 |=============================================================== OpenVINO 2022.3 Model: Person Detection FP32 - Device: CPU FPS > Higher Is Better 12c . 42.95 |================================================================== 10c . 43.18 |================================================================== 8c .. 42.22 |================================================================= 6c .. 41.44 |=============================================================== OpenVINO 2022.3 Model: Person Detection FP32 - Device: CPU ms < Lower Is Better 12c . 1110.68 |============================================================== 10c . 1104.59 |============================================================= 8c .. 1129.01 |=============================================================== 6c .. 1150.54 |================================================================ Xmrig 6.18.1 Variant: Monero - Hash Count: 1M H/s > Higher Is Better 12c . 104604.6 |=============================================================== 10c . 102599.6 |============================================================== 8c .. 101953.5 |============================================================= 6c .. 100446.2 |============================================================ OpenVINO 2022.3 Model: Person Detection FP16 - Device: CPU FPS > Higher Is Better 12c . 42.98 |================================================================== 10c . 42.94 |================================================================== 8c .. 42.59 |================================================================= 6c .. 41.33 |=============================================================== OpenVINO 2022.3 Model: Person Detection FP16 - Device: CPU ms < Lower Is Better 12c . 1109.45 |============================================================== 10c . 1110.44 |============================================================== 8c .. 1119.79 |============================================================== 6c .. 1153.70 |================================================================ Timed GDB GNU Debugger Compilation 10.2 Time To Compile Seconds < Lower Is Better 12c . 41.71 |================================================================ 10c . 42.41 |================================================================= 8c .. 42.41 |================================================================= 6c .. 43.25 |================================================================== Timed Gem5 Compilation 21.2 Time To Compile Seconds < Lower Is Better 12c . 139.24 |================================================================= 10c . 134.37 |=============================================================== 8c .. 136.79 |================================================================ 6c .. 134.70 |=============================================================== CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 512 ops/s > Higher Is Better 12c . 47621.9 |============================================================== 10c . 49102.7 |================================================================ 8c .. 47596.6 |============================================================== 6c .. 47428.0 |============================================================== Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second > Higher Is Better 12c . 213.75 |=============================================================== 10c . 214.31 |=============================================================== 8c .. 217.41 |================================================================ 6c .. 221.29 |================================================================= CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 512 ops/s > Higher Is Better 12c . 35970.0 |================================================================ 10c . 35993.1 |================================================================ 8c .. 34832.9 |============================================================== 6c .. 35742.3 |================================================================ OpenVINO 2022.3 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU FPS > Higher Is Better 12c . 119606.21 |============================================================ 10c . 122938.23 |============================================================== 8c .. 123571.68 |============================================================== 6c .. 121027.25 |============================================================= Timed Linux Kernel Compilation 6.1 Build: defconfig Seconds < Lower Is Better 12c . 25.50 |================================================================== 10c . 25.41 |================================================================== 8c .. 25.53 |================================================================== 6c .. 24.75 |================================================================ OpenVINO 2022.3 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU FPS > Higher Is Better 12c . 147769.26 |============================================================ 10c . 147717.32 |============================================================ 8c .. 152292.39 |============================================================== 6c .. 151213.17 |============================================================== Rodinia 3.1 Test: OpenMP CFD Solver Seconds < Lower Is Better 12c . 6.050 |================================================================= 10c . 6.074 |================================================================= 8c .. 5.970 |================================================================ 6c .. 6.152 |================================================================== Blender 3.4 Blend File: BMW27 - Compute: CPU-Only Seconds < Lower Is Better 12c . 8.58 |=================================================================== 10c . 8.42 |================================================================== 8c .. 8.34 |================================================================= 6c .. 8.33 |================================================================= CockroachDB 22.2 Workload: KV, 10% Reads - Concurrency: 1024 ops/s > Higher Is Better 12c . 36846.9 |================================================================ 10c . 35776.8 |============================================================== 8c .. 36685.7 |================================================================ 6c .. 36329.6 |=============================================================== Embree 3.13 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better 12c . 182.45 |=============================================================== 10c . 184.73 |================================================================ 8c .. 185.49 |================================================================ 6c .. 187.61 |================================================================= Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Frames Per Second > Higher Is Better 12c . 77.83 |================================================================== 10c . 77.30 |================================================================== 8c .. 76.84 |================================================================= 6c .. 75.86 |================================================================ OSPRay 2.10 Benchmark: particle_volume/scivis/real_time Items Per Second > Higher Is Better 12c . 42.80 |================================================================ 10c . 43.00 |================================================================= 8c .. 43.84 |================================================================== 6c .. 43.24 |================================================================= Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 84.25 |================================================================== 10c . 84.27 |================================================================== 8c .. 84.15 |================================================================== 6c .. 82.26 |================================================================ Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 84.35 |================================================================== 10c . 84.48 |================================================================== 8c .. 84.21 |================================================================== 6c .. 82.49 |================================================================ CockroachDB 22.2 Workload: KV, 60% Reads - Concurrency: 512 ops/s > Higher Is Better 12c . 52330.1 |================================================================ 10c . 51748.8 |=============================================================== 8c .. 52515.2 |================================================================ 6c .. 51275.1 |============================================================== libavif avifenc 0.11 Encoder Speed: 10, Lossless Seconds < Lower Is Better 12c . 4.241 |================================================================= 10c . 4.337 |================================================================== 8c .. 4.252 |================================================================= 6c .. 4.250 |================================================================= OSPRay 2.10 Benchmark: particle_volume/ao/real_time Items Per Second > Higher Is Better 12c . 43.71 |================================================================== 10c . 43.03 |================================================================= 8c .. 43.97 |================================================================== 6c .. 43.36 |================================================================= nginx 1.23.2 Connections: 500 Requests Per Second > Higher Is Better 12c . 201032.06 |============================================================== 10c . 198858.66 |============================================================= 8c .. 197081.98 |============================================================= 6c .. 196805.30 |============================================================= DaCapo Benchmark 9.12-MR1 Java Test: H2 msec < Lower Is Better 12c . 4802 |=================================================================== 10c . 4832 |=================================================================== 8c .. 4731 |================================================================== 6c .. 4830 |=================================================================== CockroachDB 22.2 Workload: KV, 50% Reads - Concurrency: 1024 ops/s > Higher Is Better 12c . 47465.5 |=============================================================== 10c . 48449.0 |================================================================ 8c .. 47498.1 |=============================================================== 6c .. 47593.9 |=============================================================== libavif avifenc 0.11 Encoder Speed: 6 Seconds < Lower Is Better 12c . 2.459 |================================================================== 10c . 2.411 |================================================================= 8c .. 2.420 |================================================================= 6c .. 2.435 |================================================================= Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second > Higher Is Better 12c . 62.56 |================================================================== 10c . 62.23 |================================================================== 8c .. 61.81 |================================================================= 6c .. 61.40 |================================================================= 7-Zip Compression 22.01 Test: Decompression Rating MIPS > Higher Is Better 12c . 1181435 |================================================================ 10c . 1171627 |=============================================================== 8c .. 1159901 |=============================================================== 6c .. 1177484 |================================================================ OpenVINO 2022.3 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU ms < Lower Is Better 12c . 0.55 |=================================================================== 10c . 0.55 |=================================================================== 8c .. 0.55 |=================================================================== 6c .. 0.54 |================================================================== Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 1964.27 |================================================================ 10c . 1965.56 |================================================================ 8c .. 1954.12 |================================================================ 6c .. 1930.33 |=============================================================== Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 48.77 |================================================================= 10c . 48.74 |================================================================= 8c .. 49.00 |================================================================= 6c .. 49.63 |================================================================== Timed LLVM Compilation 13.0 Build System: Ninja Seconds < Lower Is Better 12c . 75.66 |================================================================= 10c . 75.44 |================================================================= 8c .. 75.73 |================================================================= 6c .. 76.75 |================================================================== OpenVINO 2022.3 Model: Vehicle Detection FP16 - Device: CPU ms < Lower Is Better 12c . 6.48 |================================================================== 10c . 6.45 |================================================================== 8c .. 6.49 |================================================================== 6c .. 6.56 |=================================================================== OpenVINO 2022.3 Model: Vehicle Detection FP16 - Device: CPU FPS > Higher Is Better 12c . 7394.65 |================================================================ 10c . 7425.10 |================================================================ 8c .. 7389.00 |================================================================ 6c .. 7306.47 |=============================================================== Timed Node.js Compilation 18.8 Time To Compile Seconds < Lower Is Better 12c . 101.47 |================================================================ 10c . 101.94 |================================================================ 8c .. 101.15 |================================================================ 6c .. 102.78 |================================================================= ONNX Runtime 1.11 Model: fcn-resnet101-11 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better 12c . 254 |=================================================================== 10c . 255 |=================================================================== 8c .. 257 |==================================================================== 6c .. 253 |=================================================================== OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time Items Per Second > Higher Is Better 12c . 53.77 |================================================================= 10c . 54.41 |================================================================== 8c .. 54.51 |================================================================== 6c .. 54.61 |================================================================== simdjson 2.0 Throughput Test: TopTweet GB/s > Higher Is Better 12c . 6.59 |=================================================================== 10c . 6.49 |================================================================== 8c .. 6.57 |=================================================================== 6c .. 6.55 |=================================================================== DaCapo Benchmark 9.12-MR1 Java Test: Jython msec < Lower Is Better 12c . 3380 |=================================================================== 10c . 3329 |================================================================== 8c .. 3369 |=================================================================== 6c .. 3345 |================================================================== Blender 3.4 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better 12c . 81.03 |================================================================== 10c . 80.37 |================================================================= 8c .. 80.18 |================================================================= 6c .. 79.93 |================================================================= Timed Linux Kernel Compilation 6.1 Build: allmodconfig Seconds < Lower Is Better 12c . 147.15 |================================================================= 10c . 145.41 |================================================================ 8c .. 147.38 |================================================================= 6c .. 145.77 |================================================================ Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 1133.28 |=============================================================== 10c . 1133.18 |=============================================================== 8c .. 1136.85 |=============================================================== 6c .. 1148.50 |================================================================ libavif avifenc 0.11 Encoder Speed: 0 Seconds < Lower Is Better 12c . 63.25 |================================================================= 10c . 63.25 |================================================================= 8c .. 62.96 |================================================================= 6c .. 63.80 |================================================================== Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 1133.48 |=============================================================== 10c . 1135.18 |=============================================================== 8c .. 1137.51 |=============================================================== 6c .. 1148.33 |================================================================ CockroachDB 22.2 Workload: KV, 60% Reads - Concurrency: 1024 ops/s > Higher Is Better 12c . 52573.3 |================================================================ 10c . 51959.5 |=============================================================== 8c .. 52559.0 |================================================================ 6c .. 52626.4 |================================================================ Timed Apache Compilation 2.4.41 Time To Compile Seconds < Lower Is Better 12c . 20.46 |================================================================= 10c . 20.48 |================================================================= 8c .. 20.59 |================================================================== 6c .. 20.72 |================================================================== NWChem 7.0.2 Input: C240 Buckyball Seconds < Lower Is Better 12c . 1537.1 |================================================================= 10c . 1531.0 |================================================================= 8c .. 1519.6 |================================================================ 6c .. 1517.9 |================================================================ CockroachDB 22.2 Workload: MoVR - Concurrency: 512 ops/s > Higher Is Better 12c . 948.5 |================================================================= 10c . 949.6 |================================================================= 8c .. 960.3 |================================================================== 6c .. 954.7 |================================================================== Timed Godot Game Engine Compilation 3.2.3 Time To Compile Seconds < Lower Is Better 12c . 34.03 |================================================================== 10c . 33.62 |================================================================= 8c .. 33.91 |================================================================== 6c .. 33.67 |================================================================= OpenVINO 2022.3 Model: Vehicle Detection FP16-INT8 - Device: CPU FPS > Higher Is Better 12c . 11018.37 |============================================================== 10c . 11066.16 |=============================================================== 8c .. 11108.16 |=============================================================== 6c .. 11150.32 |=============================================================== OpenRadioss 2022.10.13 Model: Bird Strike on Windshield Seconds < Lower Is Better 12c . 216.88 |================================================================ 10c . 218.22 |================================================================= 8c .. 219.45 |================================================================= 6c .. 219.10 |================================================================= OpenVINO 2022.3 Model: Vehicle Detection FP16-INT8 - Device: CPU ms < Lower Is Better 12c . 4.35 |=================================================================== 10c . 4.33 |=================================================================== 8c .. 4.31 |================================================================== 6c .. 4.30 |================================================================== Blender 3.4 Blend File: Classroom - Compute: CPU-Only Seconds < Lower Is Better 12c . 20.92 |================================================================== 10c . 20.76 |================================================================= 8c .. 20.68 |================================================================= 6c .. 20.71 |================================================================= libavif avifenc 0.11 Encoder Speed: 6, Lossless Seconds < Lower Is Better 12c . 5.287 |================================================================= 10c . 5.286 |================================================================= 8c .. 5.270 |================================================================= 6c .. 5.330 |================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 615.45 |================================================================= 10c . 611.29 |================================================================= 8c .. 614.61 |================================================================= 6c .. 608.53 |================================================================ Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 155.48 |================================================================ 10c . 156.54 |================================================================= 8c .. 155.82 |================================================================ 6c .. 157.22 |================================================================= OpenVINO 2022.3 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU ms < Lower Is Better 12c . 0.97 |================================================================== 10c . 0.98 |=================================================================== 8c .. 0.98 |=================================================================== 6c .. 0.97 |================================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better 12c . 1195.91 |================================================================ 10c . 1201.14 |================================================================ 8c .. 1201.98 |================================================================ 6c .. 1190.53 |=============================================================== Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better 12c . 80.08 |================================================================== 10c . 79.71 |================================================================= 8c .. 79.69 |================================================================= 6c .. 80.44 |================================================================== OpenRadioss 2022.10.13 Model: INIVOL and Fluid Structure Interaction Drop Container Seconds < Lower Is Better 12c . 81.57 |================================================================== 10c . 81.15 |================================================================== 8c .. 81.09 |================================================================== 6c .. 80.81 |================================================================= OpenVINO 2022.3 Model: Weld Porosity Detection FP16 - Device: CPU FPS > Higher Is Better 12c . 9867.41 |=============================================================== 10c . 9900.47 |================================================================ 8c .. 9931.49 |================================================================ 6c .. 9959.38 |================================================================ Xmrig 6.18.1 Variant: Wownero - Hash Count: 1M H/s > Higher Is Better 12c . 126465.6 |=============================================================== 10c . 127226.6 |=============================================================== 8c .. 127081.2 |=============================================================== 6c .. 126057.7 |============================================================== OpenVINO 2022.3 Model: Face Detection FP16 - Device: CPU FPS > Higher Is Better 12c . 101.74 |================================================================= 10c . 102.01 |================================================================= 8c .. 101.26 |================================================================= 6c .. 101.08 |================================================================ OpenVINO 2022.3 Model: Face Detection FP16 - Device: CPU ms < Lower Is Better 12c . 470.98 |================================================================= 10c . 469.43 |================================================================ 8c .. 472.84 |================================================================= 6c .. 473.69 |================================================================= ACES DGEMM 1.0 Sustained Floating-Point Rate GFLOP/s > Higher Is Better 12c . 70.41 |================================================================= 10c . 70.61 |================================================================== 8c .. 71.01 |================================================================== 6c .. 70.90 |================================================================== OpenRadioss 2022.10.13 Model: Bumper Beam Seconds < Lower Is Better 12c . 79.86 |================================================================== 10c . 79.70 |================================================================== 8c .. 79.20 |================================================================= 6c .. 79.62 |================================================================== OpenVINO 2022.3 Model: Weld Porosity Detection FP16 - Device: CPU ms < Lower Is Better 12c . 4.85 |=================================================================== 10c . 4.83 |=================================================================== 8c .. 4.82 |=================================================================== 6c .. 4.81 |================================================================== OpenVINO 2022.3 Model: Person Vehicle Bike Detection FP16 - Device: CPU FPS > Higher Is Better 12c . 9038.47 |=============================================================== 10c . 9063.84 |================================================================ 8c .. 9113.11 |================================================================ 6c .. 9081.73 |================================================================ ASTC Encoder 4.0 Preset: Exhaustive MT/s > Higher Is Better 12c . 11.73 |================================================================= 10c . 11.76 |================================================================== 8c .. 11.81 |================================================================== 6c .. 11.82 |================================================================== Stargate Digital Audio Workstation 22.11.5 Sample Rate: 192000 - Buffer Size: 1024 Render Ratio > Higher Is Better 12c . 2.829061 |=============================================================== 10c . 2.806190 |============================================================== 8c .. 2.811555 |=============================================================== 6c .. 2.824814 |=============================================================== OSPRay 2.10 Benchmark: particle_volume/pathtracer/real_time Items Per Second > Higher Is Better 12c . 229.27 |================================================================= 10c . 230.28 |================================================================= 8c .. 228.58 |================================================================ 6c .. 230.44 |================================================================= simdjson 2.0 Throughput Test: LargeRandom GB/s > Higher Is Better 12c . 1.25 |=================================================================== 10c . 1.25 |=================================================================== 8c .. 1.25 |=================================================================== 6c .. 1.24 |================================================================== OpenVINO 2022.3 Model: Person Vehicle Bike Detection FP16 - Device: CPU ms < Lower Is Better 12c . 5.30 |=================================================================== 10c . 5.28 |=================================================================== 8c .. 5.26 |================================================================== 6c .. 5.28 |=================================================================== OpenVINO 2022.3 Model: Weld Porosity Detection FP16-INT8 - Device: CPU FPS > Higher Is Better 12c . 19171.51 |=============================================================== 10c . 19254.08 |=============================================================== 8c .. 19278.93 |=============================================================== 6c .. 19314.04 |=============================================================== CockroachDB 22.2 Workload: MoVR - Concurrency: 1024 ops/s > Higher Is Better 12c . 953.8 |================================================================== 10c . 949.5 |================================================================== 8c .. 946.9 |================================================================== 6c .. 952.7 |================================================================== simdjson 2.0 Throughput Test: PartialTweets GB/s > Higher Is Better 12c . 5.65 |=================================================================== 10c . 5.67 |=================================================================== 8c .. 5.66 |=================================================================== 6c .. 5.69 |=================================================================== OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/scivis/real_time Items Per Second > Higher Is Better 12c . 43.13 |================================================================== 10c . 43.33 |================================================================== 8c .. 43.43 |================================================================== 6c .. 43.29 |================================================================== Timed MPlayer Compilation 1.5 Time To Compile Seconds < Lower Is Better 12c . 7.777 |================================================================== 10c . 7.755 |================================================================== 8c .. 7.808 |================================================================== 6c .. 7.773 |================================================================== OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/ao/real_time Items Per Second > Higher Is Better 12c . 43.98 |================================================================== 10c . 44.00 |================================================================== 8c .. 44.23 |================================================================== 6c .. 44.27 |================================================================== libavif avifenc 0.11 Encoder Speed: 2 Seconds < Lower Is Better 12c . 34.85 |================================================================== 10c . 34.91 |================================================================== 8c .. 34.69 |================================================================== 6c .. 34.87 |================================================================== OpenVINO 2022.3 Model: Weld Porosity Detection FP16-INT8 - Device: CPU ms < Lower Is Better 12c . 9.95 |=================================================================== 10c . 9.91 |=================================================================== 8c .. 9.90 |=================================================================== 6c .. 9.89 |=================================================================== miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 GFInst/s > Higher Is Better 12c . 8640.31 |================================================================ 10c . 8666.98 |================================================================ 8c .. 8615.97 |================================================================ 6c .. 8651.92 |================================================================ miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Billion Interactions/s > Higher Is Better 12c . 345.61 |================================================================= 10c . 346.68 |================================================================= 8c .. 344.64 |================================================================= 6c .. 346.08 |================================================================= Build2 0.13 Time To Compile Seconds < Lower Is Better 12c . 49.92 |================================================================== 10c . 49.80 |================================================================== 8c .. 49.87 |================================================================== 6c .. 50.08 |================================================================== ASTC Encoder 4.0 Preset: Thorough MT/s > Higher Is Better 12c . 106.57 |================================================================= 10c . 106.85 |================================================================= 8c .. 107.11 |================================================================= 6c .. 106.51 |================================================================= OpenVINO 2022.3 Model: Face Detection FP16-INT8 - Device: CPU ms < Lower Is Better 12c . 250.34 |================================================================= 10c . 249.12 |================================================================= 8c .. 249.26 |================================================================= 6c .. 250.49 |================================================================= OpenVINO 2022.3 Model: Face Detection FP16-INT8 - Device: CPU FPS > Higher Is Better 12c . 191.43 |================================================================= 10c . 192.30 |================================================================= 8c .. 192.25 |================================================================= 6c .. 191.29 |================================================================= Timed Mesa Compilation 21.0 Time To Compile Seconds < Lower Is Better 12c . 20.12 |================================================================== 10c . 20.21 |================================================================== 8c .. 20.11 |================================================================== 6c .. 20.16 |================================================================== NAMD 2.14 ATPase Simulation - 327,506 Atoms days/ns < Lower Is Better 12c . 0.12783 |================================================================ 10c . 0.12759 |================================================================ 8c .. 0.12768 |================================================================ 6c .. 0.12820 |================================================================ simdjson 2.0 Throughput Test: DistinctUserID GB/s > Higher Is Better 12c . 6.86 |=================================================================== 10c . 6.84 |=================================================================== 8c .. 6.86 |=================================================================== 6c .. 6.83 |=================================================================== Stargate Digital Audio Workstation 22.11.5 Sample Rate: 96000 - Buffer Size: 1024 Render Ratio > Higher Is Better 12c . 4.345890 |=============================================================== 10c . 4.354556 |=============================================================== 8c .. 4.351402 |=============================================================== 6c .. 4.364767 |=============================================================== Timed PHP Compilation 8.1.9 Time To Compile Seconds < Lower Is Better 12c . 44.52 |================================================================== 10c . 44.61 |================================================================== 8c .. 44.58 |================================================================== 6c .. 44.70 |================================================================== Liquid-DSP 2021.01.31 Threads: 256 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better 12c . 10347000000 |============================================================ 10c . 10340000000 |============================================================ 8c .. 10337666667 |============================================================ 6c .. 10340333333 |============================================================ Liquid-DSP 2021.01.31 Threads: 384 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better 12c . 10347000000 |============================================================ 10c . 10352666667 |============================================================ 8c .. 10349666667 |============================================================ 6c .. 10349000000 |============================================================ simdjson 2.0 Throughput Test: Kostya GB/s > Higher Is Better 12c . 4.11 |=================================================================== 10c . 4.11 |=================================================================== 8c .. 4.11 |=================================================================== 6c .. 4.11 |=================================================================== CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 1024 ops/s > Higher Is Better 12c . 64661.8 |================================================================ 10c . 62029.8 |============================================================= 8c .. 58195.5 |========================================================== 6c .. 60137.3 |============================================================ CockroachDB 22.2 Workload: KV, 95% Reads - Concurrency: 512 ops/s > Higher Is Better 12c . 64467.6 |================================================================ 10c . 60769.7 |============================================================ 8c .. 64111.9 |================================================================ 6c .. 62666.5 |============================================================== oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 12c . 1968.70 |============================================================= 10c . 2030.72 |=============================================================== 8c .. 1982.15 |============================================================= 6c .. 2072.57 |================================================================ oneDNN 3.0 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better 12c . 3.95471 |=============================================================== 10c . 4.00938 |================================================================ 8c .. 3.99305 |================================================================ 6c .. 3.96488 |=============================================================== SVT-AV1 1.4 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better 12c . 251.77 |================================================================= 10c . 241.37 |============================================================== 8c .. 227.90 |=========================================================== 6c .. 221.16 |========================================================= LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU M samples/sec > Higher Is Better 12c . 28.82 |================================================================== 10c . 28.19 |================================================================ 8c .. 29.04 |================================================================== 6c .. 28.90 |================================================================== LuxCoreRender 2.6 Scene: Danish Mood - Acceleration: CPU M samples/sec > Higher Is Better 12c . 9.69 |=================================================================== 10c . 9.62 |=================================================================== 8c .. 9.56 |================================================================== 6c .. 9.49 |================================================================== NAS Parallel Benchmarks 3.4 Test / Class: IS.D Total Mop/s > Higher Is Better 12c . 8491.01 |================================================================ 10c . 7124.92 |====================================================== 8c .. 6675.71 |================================================== 6c .. 5690.01 |=========================================== High Performance Conjugate Gradient 3.1 GFLOP/s > Higher Is Better 12c . 86.81 |================================================================== 10c . 48.29 |===================================== 8c .. 45.00 |================================== 6c .. 36.54 |============================