Google Cloud c3 Sapphire Rapids vs. AMD Milan

Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2303280-NE-2303286NE24
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

Limit displaying results to tests within:

Timed Code Compilation 2 Tests
C/C++ Compiler Tests 8 Tests
Compression Tests 2 Tests
CPU Massive 14 Tests
Creator Workloads 9 Tests
Cryptography 2 Tests
Database Test Suite 4 Tests
Fortran Tests 2 Tests
Game Development 4 Tests
HPC - High Performance Computing 12 Tests
Common Kernel Benchmarks 2 Tests
Machine Learning 5 Tests
Molecular Dynamics 4 Tests
MPI Benchmarks 2 Tests
Multi-Core 16 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 5 Tests
OpenMPI Tests 6 Tests
Programmer / Developer System Benchmarks 3 Tests
Python Tests 2 Tests
Renderers 2 Tests
Scientific Computing 4 Tests
Server 6 Tests
Server CPU Tests 9 Tests
Common Workstation Benchmarks 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
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

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
c3-highcpu-8 SPR
March 21 2023
  12 Hours, 6 Minutes
c2-standard-8 CLX
March 21 2023
  13 Hours
n2-standard-8 CLX
March 22 2023
  16 Hours, 1 Minute
n2-highcpu-8 CLX
March 23 2023
  14 Hours, 13 Minutes
t2d-standard-8 AMD
March 27 2023
  11 Hours, 20 Minutes
c2d-highcpu-8 AMD
March 28 2023
  11 Hours, 59 Minutes
Invert Hiding All Results Option
  13 Hours, 7 Minutes

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):


Google Cloud c3 Sapphire Rapids vs. AMD Milan Suite 1.0.0 System Test suite extracted from Google Cloud c3 Sapphire Rapids vs. AMD Milan. pts/onednn-3.0.0 --conv --batch=inputs/conv/shapes_auto --cfg=bf16bf16bf16 --engine=cpu Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU pts/onednn-3.0.0 --matmul --batch=inputs/matmul/shapes_transformer --cfg=bf16bf16bf16 --engine=cpu Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU pts/openssl-3.1.0 sha256 Algorithm: SHA256 pts/onednn-3.0.0 --deconv --batch=inputs/deconv/shapes_1d --cfg=bf16bf16bf16 --engine=cpu Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU pts/openssl-3.1.0 -evp aes-256-gcm Algorithm: AES-256-GCM pts/openssl-3.1.0 -evp aes-128-gcm Algorithm: AES-128-GCM pts/openradioss-1.0.0 RUBBER_SEAL_IMPDISP_GEOM_0000.rad RUBBER_SEAL_IMPDISP_GEOM_0001.rad Model: Rubber O-Ring Seal Installation pts/openradioss-1.0.0 Bumper_Beam_AP_meshed_0000.rad Bumper_Beam_AP_meshed_0001.rad Model: Bumper Beam pts/openssl-3.1.0 sha512 Algorithm: SHA512 pts/ospray-studio-1.1.0 --cameras 3 3 --resolution 3840 2160 --spp 1 --renderer pathtracer Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer pts/openradioss-1.0.0 Cell_Phone_Drop_0000.rad Cell_Phone_Drop_0001.rad Model: Cell Phone Drop Test pts/deepsparse-1.3.2 zoo:nlp/document_classification/obert-base/pytorch/huggingface/imdb/base-none --scenario async Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream pts/deepsparse-1.3.2 zoo:nlp/token_classification/bert-base/pytorch/huggingface/conll2003/base-none --scenario async Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream pts/minibude-1.0.0 --deck ../data/bm1 --iterations 500 Implementation: OpenMP - Input Deck: BM1 pts/brl-cad-1.4.0 VGR Performance Metric pts/pgbench-1.13.0 -s 100 -c 1000 -S Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency pts/pgbench-1.13.0 -s 100 -c 1000 -S Scaling Factor: 100 - Clients: 1000 - Mode: Read Only pts/uvg266-1.0.0 -i Bosphorus_3840x2160.y4m --preset superfast Video Input: Bosphorus 4K - Video Preset: Super Fast pts/uvg266-1.0.0 -i Bosphorus_3840x2160.y4m --preset veryfast Video Input: Bosphorus 4K - Video Preset: Very Fast pts/uvg266-1.0.0 -i Bosphorus_1920x1080_120fps_420_8bit_YUV.y4m --preset superfast Video Input: Bosphorus 1080p - Video Preset: Super Fast pts/uvg266-1.0.0 -i Bosphorus_1920x1080_120fps_420_8bit_YUV.y4m --preset veryfast Video Input: Bosphorus 1080p - Video Preset: Very Fast pts/uvg266-1.0.0 -i Bosphorus_3840x2160.y4m --preset ultrafast Video Input: Bosphorus 4K - Video Preset: Ultra Fast pts/pgbench-1.13.0 -s 100 -c 800 -S Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency pts/pgbench-1.13.0 -s 100 -c 800 -S Scaling Factor: 100 - Clients: 800 - Mode: Read Only pts/specfem3d-1.0.0 homogeneous_halfspace Model: Homogeneous Halfspace pts/openradioss-1.0.0 BIRD_WINDSHIELD_v1_0000.rad BIRD_WINDSHIELD_v1_0001.rad Model: Bird Strike on Windshield pts/gromacs-1.8.0 mpi-build water-cut1.0_GMX50_bare/1536 Implementation: MPI CPU - Input: water_GMX50_bare pts/opencv-1.3.0 objdetect Test: Object Detection pts/uvg266-1.0.0 -i Bosphorus_1920x1080_120fps_420_8bit_YUV.y4m --preset ultrafast Video Input: Bosphorus 1080p - Video Preset: Ultra Fast pts/specfem3d-1.0.0 tomographic_model Model: Tomographic Model pts/deepsparse-1.3.2 zoo:nlp/text_classification/bert-base/pytorch/huggingface/sst2/base-none --scenario async Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream pts/memcached-1.1.0 --ratio=1:100 Set To Get Ratio: 1:100 pts/openssl-3.1.0 rsa4096 Algorithm: RSA4096 pts/specfem3d-1.0.0 Mount_StHelens Model: Mount St. Helens pts/onednn-3.0.0 --ip --batch=inputs/ip/shapes_1d --cfg=bf16bf16bf16 --engine=cpu Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU pts/onednn-3.0.0 --deconv --batch=inputs/deconv/shapes_3d --cfg=bf16bf16bf16 --engine=cpu Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU pts/deepsparse-1.3.2 zoo:nlp/text_classification/distilbert-none/pytorch/huggingface/mnli/base-none --scenario async Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream pts/compress-zstd-1.6.0 -b19 Compression Level: 19 - Decompression Speed pts/memcached-1.1.0 --ratio=1:10 Set To Get Ratio: 1:10 pts/openfoam-1.2.0 incompressible/simpleFoam/drivaerFastback/ -m S Input: drivaerFastback, Small Mesh Size - Execution Time pts/openfoam-1.2.0 incompressible/simpleFoam/drivaerFastback/ -m S Input: drivaerFastback, Small Mesh Size - Mesh Time pts/specfem3d-1.0.0 waterlayered_halfspace Model: Water-layered Halfspace pts/build-ffmpeg-6.0.0 Time To Compile pts/incompact3d-2.0.2 input_129_nodes.i3d Input: input.i3d 129 Cells Per Direction pts/deepsparse-1.3.2 zoo:cv/classification/resnet_v1-50/pytorch/sparseml/imagenet/base-none --scenario async Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream pts/oidn-1.4.0 -r RTLightmap.hdr.4096x4096 Run: RTLightmap.hdr.4096x4096 pts/oidn-1.4.0 -r RT.hdr_alb_nrm.3840x2160 Run: RT.hdr_alb_nrm.3840x2160 pts/specfem3d-1.0.0 layered_halfspace Model: Layered Halfspace pts/tensorflow-2.0.0 --device cpu --batch_size=64 --model=resnet50 Device: CPU - Batch Size: 64 - Model: ResNet-50 pts/openssl-3.1.0 -evp chacha20-poly1305 Algorithm: ChaCha20-Poly1305 pts/compress-zstd-1.6.0 -b19 --long Compression Level: 19, Long Mode - Decompression Speed pts/deepsparse-1.3.2 zoo:cv/segmentation/yolact-darknet53/pytorch/dbolya/coco/pruned90-none --scenario async Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream pts/john-the-ripper-1.8.0 --format=bcrypt Test: Blowfish pts/john-the-ripper-1.8.0 --format=bcrypt Test: bcrypt pts/nekrs-1.0.0 turbPipePeriodic turbPipe.par Input: TurboPipe Periodic pts/build-linux-kernel-1.15.0 defconfig Build: defconfig pts/nginx-3.0.0 -c 500 Connections: 500 pts/deepsparse-1.3.2 zoo:nlp/question_answering/bert-base/pytorch/huggingface/squad/12layer_pruned90-none --scenario async Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream pts/nginx-3.0.0 -c 1000 Connections: 1000 pts/john-the-ripper-1.8.0 --format=md5crypt Test: MD5 pts/tensorflow-2.0.0 --device cpu --batch_size=32 --model=resnet50 Device: CPU - Batch Size: 32 - Model: ResNet-50 pts/draco-1.6.0 -i church.ply Model: Church Facade pts/compress-7zip-1.10.0 Test: Compression Rating pts/nginx-3.0.0 -c 4000 Connections: 4000 pts/ospray-studio-1.1.0 --cameras 1 1 --resolution 3840 2160 --spp 1 --renderer pathtracer Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer pts/embree-1.4.0 pathtracer_ispc -c crown/crown.ecs Binary: Pathtracer ISPC - Model: Crown pts/cockroach-1.0.2 kv --ramp 10s --read-percent 95 --concurrency 128 Workload: KV, 95% Reads - Concurrency: 128 pts/nginx-3.0.0 -c 200 Connections: 200 pts/lczero-1.6.0 -b blas Backend: BLAS pts/lczero-1.6.0 -b eigen Backend: Eigen pts/tensorflow-2.0.0 --device cpu --batch_size=16 --model=resnet50 Device: CPU - Batch Size: 16 - Model: ResNet-50 pts/namd-1.2.1 ATPase Simulation - 327,506 Atoms pts/onednn-3.0.0 --rnn --batch=inputs/rnn/perf_rnn_training --cfg=bf16bf16bf16 --engine=cpu Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU pts/nginx-3.0.0 -c 100 Connections: 100 pts/onednn-3.0.0 --ip --batch=inputs/ip/shapes_3d --cfg=bf16bf16bf16 --engine=cpu Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU pts/embree-1.4.0 pathtracer_ispc -c asian_dragon/asian_dragon.ecs Binary: Pathtracer ISPC - Model: Asian Dragon pts/onednn-3.0.0 --rnn --batch=inputs/rnn/perf_rnn_inference_lb --cfg=bf16bf16bf16 --engine=cpu Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU pts/openvkl-1.3.0 vklBenchmark --benchmark_filter=ispc Benchmark: vklBenchmark ISPC pts/opencv-1.3.0 imgproc Test: Image Processing pts/mysqlslap-1.4.0 --concurrency=4096 Clients: 4096 pts/cockroach-1.0.2 kv --ramp 10s --read-percent 50 --concurrency 128 Workload: KV, 50% Reads - Concurrency: 128 pts/openssl-3.1.0 -evp chacha20 Algorithm: ChaCha20 pts/mysqlslap-1.4.0 --concurrency=2048 Clients: 2048 pts/opencv-1.3.0 stitching Test: Stitching pts/draco-1.6.0 -i lion.ply Model: Lion pts/opencv-1.3.0 gapi Test: Graph API pts/compress-zstd-1.6.0 -b19 Compression Level: 19 - Compression Speed pts/compress-zstd-1.6.0 -b19 --long Compression Level: 19, Long Mode - Compression Speed pts/blender-3.4.0 -b ../bmw27_gpu.blend -o output.test -x 1 -F JPEG -f 1 -- --cycles-device CPU Blend File: BMW27 - Compute: CPU-Only pts/opencv-1.3.0 core Test: Core