EPYC 7763 LLVM Clang Compiler Tests

AMD EPYC 7763 64-Core testing with a Supermicro H12SSL-i v1.01 (2.0 BIOS) and ASPEED on Ubuntu 20.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2104140-IB-EPYC7763L31&sor&grs.

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU

ViennaCL

Test: CPU BLAS - dCOPY

ViennaCL

Test: CPU BLAS - dAXPY

Etcpak

Configuration: DXT1

ViennaCL

Test: CPU BLAS - dGEMM-NN

ViennaCL

Test: CPU BLAS - dGEMM-TN

dav1d

Video Input: Chimera 1080p 10-bit

GraphicsMagick

Operation: Resizing

Botan

Test: ChaCha20Poly1305 - Decrypt

C-Ray

Total Time - 4K, 16 Rays Per Pixel

Botan

Test: ChaCha20Poly1305

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU

LibRaw

Post-Processing Benchmark

FinanceBench

Benchmark: Bonds OpenMP

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU

ViennaCL

Test: CPU BLAS - dGEMM-NT

ViennaCL

Test: CPU BLAS - dDOT

SVT-AV1

Encoder Mode: Enc Mode 0 - Input: 1080p

toyBrot Fractal Generator

Implementation: C++ Threads

Etcpak

Configuration: ETC1

toyBrot Fractal Generator

Implementation: TBB

toyBrot Fractal Generator

Implementation: OpenMP

toyBrot Fractal Generator

Implementation: C++ Tasks

ViennaCL

Test: CPU BLAS - dGEMM-TT

SciMark

Computational Test: Sparse Matrix Multiply

Botan

Test: Blowfish

GraphicsMagick

Operation: Sharpen

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU

GraphicsMagick

Operation: HWB Color Space

oneDNN

Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU

FinanceBench

Benchmark: Repo OpenMP

SVT-AV1

Encoder Mode: Enc Mode 4 - Input: 1080p

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU

ViennaCL

Test: CPU BLAS - dGEMV-T

SVT-AV1

Encoder Mode: Enc Mode 8 - Input: 1080p

Coremark

CoreMark Size 666 - Iterations Per Second

ASTC Encoder

Preset: Medium

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

FFTW

Build: Float + SSE - Size: 1D FFT Size 2048

Liquid-DSP

Threads: 128 - Buffer Length: 256 - Filter Length: 57

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU

oneDNN

Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU

SciMark

Computational Test: Jacobi Successive Over-Relaxation

GraphicsMagick

Operation: Noise-Gaussian

ONNX Runtime

Model: shufflenet-v2-10 - Device: OpenMP CPU

Botan

Test: Blowfish - Decrypt

Etcpak

Configuration: ETC2

Botan

Test: AES-256

ASTC Encoder

Preset: Thorough

FLAC Audio Encoding

WAV To FLAC

Botan

Test: AES-256 - Decrypt

LAME MP3 Encoding

WAV To MP3

TSCP

AI Chess Performance

GraphicsMagick

Operation: Enhanced

Ngspice

Circuit: C2670

simdjson

Throughput Test: PartialTweets

QuantLib

simdjson

Throughput Test: DistinctUserID

simdjson

Throughput Test: LargeRandom

ONNX Runtime

Model: yolov4 - Device: OpenMP CPU

libavif avifenc

Encoder Speed: 6, Lossless

FFTW

Build: Float + SSE - Size: 1D FFT Size 4096

oneDNN

Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU

FFTW

Build: Stock - Size: 1D FFT Size 32

Botan

Test: Twofish

FFTW

Build: Float + SSE - Size: 1D FFT Size 32

oneDNN

Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU

WebP Image Encode

Encode Settings: Quality 100, Highest Compression

ONNX Runtime

Model: fcn-resnet101-11 - Device: OpenMP CPU

GraphicsMagick

Operation: Swirl

Liquid-DSP

Threads: 1 - Buffer Length: 256 - Filter Length: 57

Botan

Test: Twofish - Decrypt

oneDNN

Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU

FFTW

Build: Stock - Size: 1D FFT Size 4096

FFTW

Build: Stock - Size: 2D FFT Size 1024

SecureMark

Benchmark: SecureMark-TLS

AOM AV1

Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p

WebP2 Image Encode

Encode Settings: Quality 100, Compression Effort 5

FFTW

Build: Stock - Size: 1D FFT Size 1024

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average Latency

PostgreSQL pgbench

Scaling Factor: 100 - Clients: 100 - Mode: Read Write

FFTW

Build: Stock - Size: 1D FFT Size 2048

libavif avifenc

Encoder Speed: 2

Liquid-DSP

Threads: 32 - Buffer Length: 256 - Filter Length: 57

FFTW

Build: Float + SSE - Size: 2D FFT Size 4096

SciMark

Computational Test: Fast Fourier Transform

AOM AV1

Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p

libavif avifenc

Encoder Speed: 6

oneDNN

Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU

oneDNN

Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU

libavif avifenc

Encoder Speed: 0

libavif avifenc

Encoder Speed: 10

AOM AV1

Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p

WebP2 Image Encode

Encode Settings: Quality 100, Lossless Compression

WebP2 Image Encode

Encode Settings: Quality 95, Compression Effort 7

oneDNN

Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU

WebP2 Image Encode

Encode Settings: Quality 75, Compression Effort 7

LZ4 Compression

Compression Level: 9 - Compression Speed

FFTW

Build: Stock - Size: 2D FFT Size 2048

Timed MrBayes Analysis

Primate Phylogeny Analysis