Xeon Cascade Lake Refresh vs. AMD EPYC Rome Linux Benchmarks
Benchmarks for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2004100-NI-XEONEPYC951&sor&grs.
MKL-DNN DNNL
Harness: Convolution Batch conv_alexnet - Data Type: f32
MKL-DNN DNNL
Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32
MKL-DNN DNNL
Harness: Deconvolution Batch deconv_3d - Data Type: f32
OSPray
Demo: Magnetic Reconnection - Renderer: SciVis
MKL-DNN DNNL
Harness: Convolution Batch conv_all - Data Type: bf16bf16bf16
NAS Parallel Benchmarks
Test / Class: EP.D
OSPray
Demo: NASA Streamlines - Renderer: Path Tracer
NAS Parallel Benchmarks
Test / Class: EP.C
Blender
Blend File: Classroom - Compute: CPU-Only
John The Ripper
Test: MD5
OSPray
Demo: XFrog Forest - Renderer: SciVis
OSPray
Demo: XFrog Forest - Renderer: Path Tracer
OSPray
Demo: NASA Streamlines - Renderer: SciVis
LuxCoreRender
Scene: Rainbow Colors and Prism
MKL-DNN DNNL
Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32
LuxCoreRender
Scene: DLSC
Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
OSPray
Demo: San Miguel - Renderer: SciVis
Embree
Binary: Pathtracer ISPC - Model: Crown
OSPray
Demo: San Miguel - Renderer: Path Tracer
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon Obj
MKL-DNN DNNL
Harness: Deconvolution Batch deconv_1d - Data Type: f32
Stockfish
Total Time
NAMD
ATPase Simulation - 327,506 Atoms
Blender
Blend File: BMW27 - Compute: CPU-Only
John The Ripper
Test: Blowfish
LAMMPS Molecular Dynamics Simulator
Model: Rhodopsin Protein
Embree
Binary: Pathtracer - Model: Crown
Intel Open Image Denoise
Scene: Memorial
Blender
Blend File: Fishy Cat - Compute: CPU-Only
asmFish
1024 Hash Memory, 26 Depth
Chaos Group V-RAY
Mode: CPU
Embree
Binary: Pathtracer - Model: Asian Dragon
Embree
Binary: Pathtracer - Model: Asian Dragon Obj
Blender
Blend File: Barbershop - Compute: CPU-Only
Tungsten Renderer
Scene: Hair
Basis Universal
Settings: UASTC Level 3
NAS Parallel Benchmarks
Test / Class: LU.C
MKL-DNN DNNL
Harness: IP Batch 1D - Data Type: f32
PostgreSQL pgbench
Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
Intel MPI Benchmarks
Test: IMB-P2P PingPong
NAS Parallel Benchmarks
Test / Class: BT.C
High Performance Conjugate Gradient
SVT-AV1
Encoder Mode: Enc Mode 8 - Input: 1080p
dav1d
Video Input: Summer Nature 1080p
Timed Linux Kernel Compilation
Time To Compile
dav1d
Video Input: Chimera 1080p 10-bit
Timed LLVM Compilation
Time To Compile
Timed MrBayes Analysis
Primate Phylogeny Analysis
dav1d
Video Input: Summer Nature 4K
Appleseed
Scene: Emily
SVT-AV1
Encoder Mode: Enc Mode 4 - Input: 1080p
OpenVKL
Benchmark: vklBenchmark
ArrayFire
Test: Conjugate Gradient CPU
ArrayFire
Test: BLAS CPU
DeepSpeech
Acceleration: CPU
Mlpack Benchmark
Benchmark: scikit_linearridgeregression
Build2
Time To Compile
Mlpack Benchmark
Benchmark: scikit_ica
Nettle
Test: chacha
Nettle
Test: poly1305-aes
Mlpack Benchmark
Benchmark: scikit_svm
Nettle
Test: aes256
Radiance Benchmark
Test: SMP Parallel
PyPerformance
Benchmark: django_template
Himeno Benchmark
Poisson Pressure Solver
Scikit-Learn
PyPerformance
Benchmark: float
PHPBench
PHP Benchmark Suite
Tungsten Renderer
Scene: Water Caustic
PyPerformance
Benchmark: regex_compile
OpenVKL
Benchmark: vklBenchmarkStructuredVolume
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon
Embree
Binary: Pathtracer ISPC - Model: Crown
Embree
Binary: Pathtracer - Model: Crown
Embree
Binary: Pathtracer - Model: Asian Dragon Obj
Numpy Benchmark
Nettle
Test: sha512
Embree
Binary: Pathtracer - Model: Asian Dragon
PyBench
Total For Average Test Times
Embree
Binary: Pathtracer ISPC - Model: Asian Dragon Obj
Intel Open Image Denoise
Scene: Memorial
OpenVKL
Benchmark: vklBenchmarkVdbVolume
Intel Open Image Denoise
Performance / Cost - Scene: Memorial
Embree
Performance / Cost - Binary: Pathtracer ISPC - Model: Crown
Embree
Performance / Cost - Binary: Pathtracer ISPC - Model: Asian Dragon Obj
Embree
Performance / Cost - Binary: Pathtracer ISPC - Model: Asian Dragon
Embree
Performance / Cost - Binary: Pathtracer - Model: Crown
Embree
Performance / Cost - Binary: Pathtracer - Model: Asian Dragon Obj
Embree
Performance / Cost - Binary: Pathtracer - Model: Asian Dragon
OpenVKL
Performance / Cost - Benchmark: vklBenchmarkStructuredVolume
OpenVKL
Performance / Cost - Benchmark: vklBenchmarkVdbVolume
OpenVKL
Performance / Cost - Benchmark: vklBenchmark
Scikit-Learn
Performance / Cost -
Mlpack Benchmark
Performance / Cost - Benchmark: scikit_linearridgeregression
Mlpack Benchmark
Performance / Cost - Benchmark: scikit_svm
Mlpack Benchmark
Performance / Cost - Benchmark: scikit_ica
PHPBench
Performance / Cost - PHP Benchmark Suite
Appleseed
Performance / Cost - Scene: Emily
PyPerformance
Performance / Cost - Benchmark: django_template
PyPerformance
Performance / Cost - Benchmark: regex_compile
PyPerformance
Performance / Cost - Benchmark: float
PyBench
Performance / Cost - Total For Average Test Times
Blender
Performance / Cost - Blend File: Pabellon Barcelona - Compute: CPU-Only
Blender
Performance / Cost - Blend File: Barbershop - Compute: CPU-Only
Blender
Performance / Cost - Blend File: Fishy Cat - Compute: CPU-Only
Blender
Performance / Cost - Blend File: Classroom - Compute: CPU-Only
Blender
Performance / Cost - Blend File: BMW27 - Compute: CPU-Only
Chaos Group V-RAY
Performance / Cost - Mode: CPU
Basis Universal
Performance / Cost - Settings: UASTC Level 3
PostgreSQL pgbench
Performance / Cost - Scaling: Buffer Test - Test: Normal Load - Mode: Read Only
Intel MPI Benchmarks
Performance / Cost - Test: IMB-P2P PingPong
Radiance Benchmark
Performance / Cost - Test: SMP Parallel
DeepSpeech
Performance / Cost - Acceleration: CPU
Numpy Benchmark
Performance / Cost -
Tungsten Renderer
Performance / Cost - Scene: Water Caustic
Tungsten Renderer
Performance / Cost - Scene: Hair
Build2
Performance / Cost - Time To Compile
Timed LLVM Compilation
Performance / Cost - Time To Compile
Timed Linux Kernel Compilation
Performance / Cost - Time To Compile
asmFish
Performance / Cost - 1024 Hash Memory, 26 Depth
Stockfish
Performance / Cost - Total Time
Himeno Benchmark
Performance / Cost - Poisson Pressure Solver
LuxCoreRender
Performance / Cost - Scene: Rainbow Colors and Prism
LuxCoreRender
Performance / Cost - Scene: DLSC
Intel Open Image Denoise
Performance / Cost - Scene: Memorial
SVT-AV1
Performance / Cost - Encoder Mode: Enc Mode 8 - Input: 1080p
SVT-AV1
Performance / Cost - Encoder Mode: Enc Mode 4 - Input: 1080p
Embree
Performance / Cost - Binary: Pathtracer ISPC - Model: Asian Dragon Obj
Embree
Performance / Cost - Binary: Pathtracer ISPC - Model: Asian Dragon
Embree
Performance / Cost - Binary: Pathtracer - Model: Asian Dragon Obj
Embree
Performance / Cost - Binary: Pathtracer - Model: Asian Dragon
Embree
Performance / Cost - Binary: Pathtracer ISPC - Model: Crown
Embree
Performance / Cost - Binary: Pathtracer - Model: Crown
OSPray
Performance / Cost - Demo: NASA Streamlines - Renderer: Path Tracer
OSPray
Performance / Cost - Demo: Magnetic Reconnection - Renderer: SciVis
OSPray
Performance / Cost - Demo: XFrog Forest - Renderer: Path Tracer
OSPray
Performance / Cost - Demo: NASA Streamlines - Renderer: SciVis
OSPray
Performance / Cost - Demo: San Miguel - Renderer: Path Tracer
OSPray
Performance / Cost - Demo: XFrog Forest - Renderer: SciVis
OSPray
Performance / Cost - Demo: San Miguel - Renderer: SciVis
dav1d
Performance / Cost - Video Input: Chimera 1080p 10-bit
dav1d
Performance / Cost - Video Input: Summer Nature 1080p
dav1d
Performance / Cost - Video Input: Summer Nature 4K
MKL-DNN DNNL
Performance / Cost - Harness: Convolution Batch conv_googlenet_v3 - Data Type: f32
MKL-DNN DNNL
Performance / Cost - Harness: Convolution Batch conv_all - Data Type: bf16bf16bf16
MKL-DNN DNNL
Performance / Cost - Harness: Deconvolution Batch deconv_3d - Data Type: u8s8f32
MKL-DNN DNNL
Performance / Cost - Harness: Convolution Batch conv_alexnet - Data Type: f32
MKL-DNN DNNL
Performance / Cost - Harness: Deconvolution Batch deconv_3d - Data Type: f32
MKL-DNN DNNL
Performance / Cost - Harness: Deconvolution Batch deconv_1d - Data Type: f32
MKL-DNN DNNL
Performance / Cost - Harness: IP Batch 1D - Data Type: f32
John The Ripper
Performance / Cost - Test: MD5
John The Ripper
Performance / Cost - Test: Blowfish
ArrayFire
Performance / Cost - Test: Conjugate Gradient CPU
ArrayFire
Performance / Cost - Test: BLAS CPU
Nettle
Performance / Cost - Test: poly1305-aes
Nettle
Performance / Cost - Test: sha512
Nettle
Performance / Cost - Test: chacha
Nettle
Performance / Cost - Test: aes256
LAMMPS Molecular Dynamics Simulator
Performance / Cost - Model: Rhodopsin Protein
Timed MrBayes Analysis
Performance / Cost - Primate Phylogeny Analysis
NAMD
Performance / Cost - ATPase Simulation - 327,506 Atoms
NAS Parallel Benchmarks
Performance / Cost - Test / Class: LU.C
NAS Parallel Benchmarks
Performance / Cost - Test / Class: EP.D
NAS Parallel Benchmarks
Performance / Cost - Test / Class: EP.C
NAS Parallel Benchmarks
Performance / Cost - Test / Class: BT.C
High Performance Conjugate Gradient
Performance / Cost -
Phoronix Test Suite v10.8.5