Xeon Cascade Lake R Intel FSGSBASE

Intel FSGSBASE benchmarking by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/2006246-NE-XEONGOLD517&obr_sgm=y&obr_swl=y&grs&sro.

PostgreSQL pgbench

Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Write

Redis

Test: LPOP

Flexible IO Tester

Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory

Flexible IO Tester

Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory

Stress-NG

Test: Context Switching

MariaDB

Clients: 512

CP2K Molecular Dynamics

Fayalite-FIST Data

Redis

Test: GET

Apache HBase

Test: Increment - Clients: 1

Apache HBase

Test: Increment - Clients: 1

Stress-NG

Test: Atomic

MariaDB

Clients: 256

Java Gradle Build

Gradle Build: Reactor

PlaidML

FP16: No - Mode: Inference - Network: VGG19 - Device: CPU

MariaDB

Clients: 128

LevelDB

Benchmark: Sequential Fill

MariaDB

Clients: 64

AOM AV1

Encoder Mode: Speed 6 Two-Pass

KeyDB

LevelDB

Benchmark: Sequential Fill

PlaidML

FP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPU

LevelDB

Benchmark: Random Fill

PlaidML

FP16: No - Mode: Inference - Network: NASNer Large - Device: CPU

LevelDB

Benchmark: Random Delete

SVT-AV1

Encoder Mode: Enc Mode 4 - Input: 1080p

Timed GDB GNU Debugger Compilation

Time To Compile

LevelDB

Benchmark: Random Fill

Numenta Anomaly Benchmark

Detector: Earthgecko Skyline

pmbench

Concurrent Worker Threads: 72 - Read-Write Ratio: 100% Writes

Timed Apache Compilation

Time To Compile

dav1d

Video Input: Summer Nature 4K

LevelDB

Benchmark: Random Read

PlaidML

FP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPU

LevelDB

Benchmark: Overwrite

VP9 libvpx Encoding

Speed: Speed 5

dav1d

Video Input: Summer Nature 1080p

PlaidML

FP16: No - Mode: Inference - Network: Inception V3 - Device: CPU

Facebook RocksDB

Test: Sequential Fill

Mlpack Benchmark

Benchmark: scikit_svm

LevelDB

Benchmark: Overwrite

Numenta Anomaly Benchmark

Detector: EXPoSE

Himeno Benchmark

Poisson Pressure Solver

ctx_clock

Context Switch Time

Redis

Test: SADD

Stress-NG

Test: SENDFILE

Facebook RocksDB

Test: Read While Writing

AOM AV1

Encoder Mode: Speed 6 Realtime

Stress-NG

Test: CPU Stress

SVT-AV1

Encoder Mode: Enc Mode 8 - Input: 1080p

LevelDB

Benchmark: Hot Read

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: CPU

Numenta Anomaly Benchmark

Detector: Relative Entropy

PlaidML

FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU

Timed Linux Kernel Compilation

Time To Compile

Mlpack Benchmark

Benchmark: scikit_ica

LevelDB

Benchmark: Fill Sync

Timed LLVM Compilation

Time To Compile

Facebook RocksDB

Test: Random Read

AOM AV1

Encoder Mode: Speed 4 Two-Pass

Redis

Test: SET

Apache HBase

Test: Random Read - Clients: 1

AOM AV1

Encoder Mode: Speed 8 Realtime

Numenta Anomaly Benchmark

Detector: Bayesian Changepoint

ebizzy

dav1d

Video Input: Chimera 1080p

Apache HBase

Test: Random Read - Clients: 1

LevelDB

Benchmark: Seek Random

VP9 libvpx Encoding

Speed: Speed 0

dav1d

Video Input: Chimera 1080p 10-bit

oneDNN

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

GROMACS

Water Benchmark

Facebook RocksDB

Test: Random Fill

PlaidML

FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU

oneDNN

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

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Only

oneDNN

Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU

Apache Siege

Concurrent Users: 10

Mlpack Benchmark

Benchmark: scikit_qda

PostgreSQL pgbench

Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only

Timed HMMer Search

Pfam Database Search

NAMD

ATPase Simulation - 327,506 Atoms

QMCPACK

oneDNN

Harness: IP Batch All - Data Type: bf16bf16bf16 - Engine: CPU

Apache Siege

Concurrent Users: 50

oneDNN

Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16 - Engine: CPU

oneDNN

Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16 - Engine: CPU

LevelDB

Benchmark: Fill Sync

SVT-AV1

Encoder Mode: Enc Mode 0 - Input: 1080p

AOM AV1

Encoder Mode: Speed 0 Two-Pass

pmbench

Concurrent Worker Threads: 1 - Read-Write Ratio: 80% Reads 20% Writes

PostMark

Disk Transaction Performance

Mlpack Benchmark

Benchmark: scikit_linearridgeregression

Apache Siege

Concurrent Users: 200

Apache HBase

Test: Async Random Read - Clients: 1

Apache HBase

Test: Async Random Read - Clients: 1

Apache HBase

Test: Sequential Read - Clients: 1

Apache HBase

Test: Sequential Read - Clients: 1

Facebook RocksDB

Test: Random Fill Sync

Apache Cassandra

Test: Writes

Memtier_benchmark

Protocol: Redis

YafaRay

Total Time For Sample Scene

Node.js Express HTTP Load Test

Renaissance

Test: Savina Reactors.IO

Renaissance

Test: Apache Spark ALS

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

Timed MAFFT Alignment

Multiple Sequence Alignment

pmbench

Concurrent Worker Threads: 72 - Read-Write Ratio: 100% Reads

BlogBench

Test: Write

Flexible IO Tester

Type: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory

Flexible IO Tester

Type: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory

Geometric Mean Of All Test Results

Result Composite - Xeon Cascade Lake R Intel FSGSBASE

Number Of First Place Finishes

Wins - 111 Tests

Number Of Last Place Finishes

Losses - 111 Tests

Phoronix Test Suite v10.8.4