cascade lake summer 2022

2 x Intel Xeon Platinum 8280 testing with a GIGABYTE MD61-SC2-00 v01000100 (T15 BIOS) and llvmpipe on Ubuntu 21.04 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2206050-NE-CASCADELA50&grs&rdt.

Stress-NG

Test: CPU Cache

ONNX Runtime

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

ONNX Runtime

Model: fcn-resnet101-11 - Device: CPU - Executor: Standard

Apache HTTP Server

Concurrent Requests: 20

ONNX Runtime

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

perf-bench

Benchmark: Epoll Wait

Apache HTTP Server

Concurrent Requests: 200

Apache HTTP Server

Concurrent Requests: 1000

Renaissance

Test: Scala Dotty

AOM AV1

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

ONNX Runtime

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

AOM AV1

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

oneDNN

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

AOM AV1

Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K

Stress-NG

Test: Atomic

Renaissance

Test: Savina Reactors.IO

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 1080p

x264

Video Input: Bosphorus 1080p

AOM AV1

Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 1080p

SVT-VP9

Tuning: VMAF Optimized - Input: Bosphorus 4K

Renaissance

Test: In-Memory Database Shootout

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K

Stress-NG

Test: Futex

libavif avifenc

Encoder Speed: 6

libavif avifenc

Encoder Speed: 2

Renaissance

Test: Apache Spark ALS

SVT-VP9

Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p

Renaissance

Test: Random Forest

perf-bench

Benchmark: Memcpy 1MB

Renaissance

Test: Finagle HTTP Requests

AOM AV1

Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K

TensorFlow Lite

Model: Inception ResNet V2

SVT-VP9

Tuning: Visual Quality Optimized - Input: Bosphorus 4K

perf-bench

Benchmark: Sched Pipe

AOM AV1

Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p

TensorFlow Lite

Model: Mobilenet Quant

Stress-NG

Test: NUMA

oneDNN

Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU

perf-bench

Benchmark: Memset 1MB

Apache HTTP Server

Concurrent Requests: 100

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 1080p

AOM AV1

Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K

AOM AV1

Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 1080p

oneDNN

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

perf-bench

Benchmark: Futex Lock-Pi

Renaissance

Test: Genetic Algorithm Using Jenetics + Futures

AOM AV1

Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p

Renaissance

Test: ALS Movie Lens

Apache HTTP Server

Concurrent Requests: 500

SVT-HEVC

Tuning: 7 - Input: Bosphorus 4K

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 4K

Apache HTTP Server

Concurrent Requests: 1

oneDNN

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

nginx

Concurrent Requests: 500

Renaissance

Test: Akka Unbalanced Cobwebbed Tree

oneDNN

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU

SVT-HEVC

Tuning: 7 - Input: Bosphorus 1080p

SVT-AV1

Encoder Mode: Preset 10 - Input: Bosphorus 4K

TensorFlow Lite

Model: NASNet Mobile

SVT-AV1

Encoder Mode: Preset 10 - Input: Bosphorus 1080p

oneDNN

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

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 4K

Nettle

Test: aes256

Stress-NG

Test: Socket Activity

Facebook RocksDB

Test: Read While Writing

simdjson

Throughput Test: Kostya

Nettle

Test: chacha

Etcpak

Benchmark: Multi-Threaded - Configuration: DXT1

SVT-AV1

Encoder Mode: Preset 12 - Input: Bosphorus 1080p

TensorFlow Lite

Model: SqueezeNet

AOM AV1

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

Stress-NG

Test: Context Switching

Nettle

Test: sha512

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

nginx

Concurrent Requests: 20

libavif avifenc

Encoder Speed: 6, Lossless

Stress-NG

Test: Glibc C String Functions

TensorFlow Lite

Model: Inception V4

oneDNN

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

oneDNN

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

Renaissance

Test: Apache Spark PageRank

Renaissance

Test: Apache Spark Bayes

perf-bench

Benchmark: Syscall Basic

Facebook RocksDB

Test: Random Read

Glibc Benchmarks

Benchmark: log2

Etcpak

Benchmark: Multi-Threaded - Configuration: ETC2

oneDNN

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

AOM AV1

Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K

oneDNN

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

Etcpak

Benchmark: Single-Threaded - Configuration: DXT1

Facebook RocksDB

Test: Read Random Write Random

SVT-HEVC

Tuning: 10 - Input: Bosphorus 4K

ONNX Runtime

Model: yolov4 - Device: CPU - Executor: Standard

oneDNN

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

SVT-AV1

Encoder Mode: Preset 4 - Input: Bosphorus 1080p

Java JMH

Throughput

Stress-NG

Test: MMAP

Nettle

Test: poly1305-aes

oneDNN

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

nginx

Concurrent Requests: 100

OSPray

Benchmark: gravity_spheres_volume/dim_512/ao/real_time

simdjson

Throughput Test: LargeRandom

nginx

Concurrent Requests: 200

AOM AV1

Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K

Glibc Benchmarks

Benchmark: modf

OSPray

Benchmark: particle_volume/ao/real_time

OSPray

Benchmark: particle_volume/scivis/real_time

SVT-HEVC

Tuning: 10 - Input: Bosphorus 1080p

Glibc Benchmarks

Benchmark: cos

Glibc Benchmarks

Benchmark: sin

oneDNN

Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU

nginx

Concurrent Requests: 1

Glibc Benchmarks

Benchmark: pthread_once

TensorFlow Lite

Model: Mobilenet Float

OSPray

Benchmark: particle_volume/pathtracer/real_time

x264

Video Input: Bosphorus 4K

simdjson

Throughput Test: PartialTweets

libavif avifenc

Encoder Speed: 0

Glibc Benchmarks

Benchmark: sinh

OSPray

Benchmark: gravity_spheres_volume/dim_512/scivis/real_time

oneDNN

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

Glibc Benchmarks

Benchmark: sqrt

Timed MPlayer Compilation

Time To Compile

Glibc Benchmarks

Benchmark: exp

Glibc Benchmarks

Benchmark: ffs

simdjson

Throughput Test: DistinctUserID

Stress-NG

Test: Vector Math

oneDNN

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

Stress-NG

Test: CPU Stress

Stress-NG

Test: Forking

oneDNN

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

ONNX Runtime

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

OSPray

Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time

OSPray Studio

Camera: 2 - Resolution: 1080p - Samples Per Pixel: 16 - Renderer: Path Tracer

oneDNN

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

Stress-NG

Test: System V Message Passing

oneDNN

Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU

nginx

Concurrent Requests: 1000

Stress-NG

Test: Memory Copying

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

Stress-NG

Test: Malloc

oneDNN

Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU

libavif avifenc

Encoder Speed: 10, Lossless