Xeon Max Linux 6.9 Kernel Benchmarks

2 x Intel Xeon Max 9480 benchmarks for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2404035-NE-XEONMAXLI15&grs.

Hackbench

Count: 32 - Type: Process

Stress-NG

Test: Pthread

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU

FFmpeg

Encoder: libx265 - Scenario: Platform

TensorFlow

Device: CPU - Batch Size: 1 - Model: AlexNet

ONNX Runtime

Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Standard

ONNX Runtime

Model: T5 Encoder - Device: CPU - Executor: Parallel

Stress-NG

Test: SENDFILE

ONNX Runtime

Model: CaffeNet 12-int8 - Device: CPU - Executor: Standard

TensorFlow

Device: CPU - Batch Size: 64 - Model: AlexNet

TensorFlow

Device: CPU - Batch Size: 64 - Model: GoogLeNet

OpenVINO

Model: Vehicle Detection FP16 - Device: CPU

OpenVINO

Model: Vehicle Detection FP16 - Device: CPU

Stress-NG

Test: System V Message Passing

WebP2 Image Encode

Encode Settings: Quality 75, Compression Effort 7

Stress-NG

Test: Context Switching

OpenVINO

Model: Person Detection FP32 - Device: CPU

OpenVINO

Model: Person Detection FP32 - Device: CPU

ONNX Runtime

Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel

Intel Open Image Denoise

Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only

Stress-NG

Test: MMAP

OpenVINO

Model: Weld Porosity Detection FP16-INT8 - Device: CPU

JPEG-XL libjxl

Input: JPEG - Quality: 90

ONNX Runtime

Model: super-resolution-10 - Device: CPU - Executor: Standard

OpenVINO

Model: Weld Porosity Detection FP16-INT8 - Device: CPU

Stress-NG

Test: Socket Activity

ONNX Runtime

Model: GPT-2 - Device: CPU - Executor: Parallel

OpenSSL

Algorithm: SHA256

OSPRay

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

Stress-NG

Test: IO_uring

OpenVINO

Model: Person Detection FP16 - Device: CPU

Llama.cpp

Model: llama-2-13b.Q4_0.gguf

OpenVINO

Model: Person Detection FP16 - Device: CPU

Blender

Blend File: Fishy Cat - Compute: CPU-Only

OpenVINO

Model: Road Segmentation ADAS FP16 - Device: CPU

OpenVINO

Model: Road Segmentation ADAS FP16 - Device: CPU

OSPRay Studio

Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

Stress-NG

Test: Semaphores

OSPRay

Benchmark: gravity_spheres_volume/dim_512/scivis/real_time

WebP2 Image Encode

Encode Settings: Quality 95, Compression Effort 7

OSPRay

Benchmark: particle_volume/ao/real_time

TensorFlow

Device: CPU - Batch Size: 16 - Model: ResNet-50

OpenSSL

Algorithm: RSA4096

JPEG-XL libjxl

Input: PNG - Quality: 90

Stress-NG

Test: Crypto

Llama.cpp

Model: llama-2-70b-chat.Q5_0.gguf

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

VVenC

Video Input: Bosphorus 1080p - Video Preset: Faster

Hackbench

Count: 16 - Type: Thread

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

Llamafile

Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU

OSPRay Studio

Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

RocksDB

Test: Read Random Write Random

OSPRay Studio

Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

Stress-NG

Test: Zlib

Stress-NG

Test: Memory Copying

Intel Open Image Denoise

Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only

oneDNN

Harness: Convolution Batch Shapes Auto - Engine: CPU

BRL-CAD

VGR Performance Metric

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU

Blender

Blend File: Junkshop - Compute: CPU-Only

OSPRay Studio

Camera: 3 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU

Timed Linux Kernel Compilation

Build: defconfig

Xmrig

Variant: Monero - Hash Count: 1M

OpenVINO

Model: Handwritten English Recognition FP16-INT8 - Device: CPU

TensorFlow

Device: CPU - Batch Size: 64 - Model: ResNet-50

OpenVINO

Model: Handwritten English Recognition FP16-INT8 - Device: CPU

Intel Open Image Denoise

Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only

Hackbench

Count: 16 - Type: Process

OSPRay

Benchmark: gravity_spheres_volume/dim_512/ao/real_time

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 4K

Llamafile

Test: mistral-7b-instruct-v0.2.Q8_0 - Acceleration: CPU

JPEG-XL libjxl

Input: PNG - Quality: 80

Primesieve

Length: 1e12

oneDNN

Harness: Deconvolution Batch shapes_1d - Engine: CPU

ONNX Runtime

Model: bertsquad-12 - Device: CPU - Executor: Parallel

OpenVINO

Model: Face Detection FP16 - Device: CPU

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection FP16 - Device: CPU

ONNX Runtime

Model: yolov4 - Device: CPU - Executor: Parallel

TensorFlow

Device: CPU - Batch Size: 256 - Model: AlexNet

OSPRay Studio

Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU

Stress-NG

Test: Mutex

OSPRay Studio

Camera: 1 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU

Stress-NG

Test: Poll

VVenC

Video Input: Bosphorus 1080p - Video Preset: Fast

Quicksilver

Input: CORAL2 P1

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 1080p

Chaos Group V-RAY

Mode: CPU

Timed Linux Kernel Compilation

Build: allmodconfig

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

LZ4 Compression

Compression Level: 9 - Decompression Speed

Stress-NG

Test: Vector Shuffle

TensorFlow

Device: CPU - Batch Size: 32 - Model: GoogLeNet

FFmpeg

Encoder: libx265 - Scenario: Video On Demand

TensorFlow

Device: CPU - Batch Size: 32 - Model: ResNet-50

TensorFlow

Device: CPU - Batch Size: 16 - Model: AlexNet

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

Google Draco

Model: Lion

OpenVINO

Model: Noise Suppression Poconet-Like FP16 - Device: CPU

Speedb

Test: Random Read

TensorFlow

Device: CPU - Batch Size: 32 - Model: AlexNet

JPEG-XL Decoding libjxl

CPU Threads: All

WebP2 Image Encode

Encode Settings: Quality 100, Compression Effort 5

OSPRay Studio

Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

Speedb

Test: Random Fill

ONNX Runtime

Model: Faster R-CNN R-50-FPN-int8 - Device: CPU - Executor: Parallel

Stress-NG

Test: NUMA

RocksDB

Test: Overwrite

ONNX Runtime

Model: ResNet50 v1-12-int8 - Device: CPU - Executor: Parallel

OpenSSL

Algorithm: SHA512

Blender

Blend File: BMW27 - Compute: CPU-Only

oneDNN

Harness: IP Shapes 3D - Engine: CPU

CacheBench

Test: Write

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

Speedb

Test: Random Fill Sync

Speedb

Test: Sequential Fill

LZ4 Compression

Compression Level: 1 - Decompression Speed

Stress-NG

Test: Hash

OSPRay

Benchmark: particle_volume/scivis/real_time

Stress-NG

Test: CPU Stress

VVenC

Video Input: Bosphorus 4K - Video Preset: Fast

oneDNN

Harness: Recurrent Neural Network Inference - Engine: CPU

Xmrig

Variant: KawPow - Hash Count: 1M

OpenVINO

Model: Person Re-Identification Retail FP16 - Device: CPU

OpenVINO

Model: Person Re-Identification Retail FP16 - Device: CPU

TensorFlow

Device: CPU - Batch Size: 256 - Model: GoogLeNet

TensorFlow

Device: CPU - Batch Size: 256 - Model: ResNet-50

ONNX Runtime

Model: super-resolution-10 - Device: CPU - Executor: Parallel

Stress-NG

Test: Wide Vector Math

Stress-NG

Test: Mixed Scheduler

Speedb

Test: Read Random Write Random

OpenVINO

Model: Handwritten English Recognition FP16 - Device: CPU

OpenVINO

Model: Handwritten English Recognition FP16 - Device: CPU

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

RocksDB

Test: Update Random

LZ4 Compression

Compression Level: 3 - Decompression Speed

OpenVINO

Model: Noise Suppression Poconet-Like FP16 - Device: CPU

OSPRay Studio

Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

Stress-NG

Test: Glibc Qsort Data Sorting

Speedb

Test: Update Random

OpenVINO

Model: Face Detection Retail FP16-INT8 - Device: CPU