Comet Lake Mitigations
Tests for a future article.
HTML result view exported from: https://openbenchmarking.org/result/2008083-NE-COMETLAKE71&grs&sro&export=pdf.
Intel Open Image Denoise
Scene: Memorial
Flexible IO Tester
Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory
Facebook RocksDB
Test: Read While Writing
OpenVKL
Benchmark: vklBenchmark
LevelDB
Benchmark: Sequential Fill
BRL-CAD
VGR Performance Metric
OpenSSL
RSA 4096-bit Performance
LevelDB
Benchmark: Random Fill
LevelDB
Benchmark: Random Delete
Stress-NG
Test: Glibc C String Functions
Stress-NG
Test: Context Switching
Stress-NG
Test: Forking
Stress-NG
Test: CPU Stress
NAS Parallel Benchmarks
Test / Class: BT.C
Rodinia
Test: OpenMP Leukocyte
Stress-NG
Test: Malloc
Stress-NG
Test: Matrix Math
Stress-NG
Test: Vector Math
7-Zip Compression
Compress Speed Test
OSBench
Test: Launch Programs
Stress-NG
Test: Glibc Qsort Data Sorting
Ethr
Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 8
x265
H.265 1080p Video Encoding
Timed Linux Kernel Compilation
Time To Compile
dav1d
Video Input: Chimera 1080p
LevelDB
Benchmark: Fill Sync
Rodinia
Test: OpenMP Streamcluster
Build2
Time To Compile
XZ Compression
Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9
Timed MPlayer Compilation
Time To Compile
Stress-NG
Test: RdRand
OCRMyPDF
Processing 60 Page PDF Document
YafaRay
Total Time For Sample Scene
NAS Parallel Benchmarks
Test / Class: EP.D
Basis Universal
Settings: UASTC Level 3
dav1d
Video Input: Chimera 1080p 10-bit
DaCapo Benchmark
Java Test: Tradesoap
Go Benchmarks
Test: http
Numenta Anomaly Benchmark
Detector: Relative Entropy
NeatBench
Acceleration: CPU
Numenta Anomaly Benchmark
Detector: Bayesian Changepoint
Darktable
Test: Masskrug - Acceleration: CPU-only
Timed GDB GNU Debugger Compilation
Time To Compile
Ethr
Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 8
Stress-NG
Test: Crypto
Tesseract OCR
Time To OCR 7 Images
Mlpack Benchmark
Benchmark: scikit_linearridgeregression
Darktable
Test: Server Room - Acceleration: CPU-only
Darktable
Test: Server Rack - Acceleration: CPU-only
Rodinia
Test: OpenMP CFD Solver
Ethr
Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 8
OSBench
Test: Create Processes
RawTherapee
Total Benchmark Time
Hugin
Panorama Photo Assistant + Stitching Time
Darktable
Test: Boat - Acceleration: CPU-only
Basis Universal
Settings: ETC1S
Flexible IO Tester
Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory
PlaidML
FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU
Flexible IO Tester
Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory
PlaidML
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPU
Sockperf
Test: Throughput
Stress-NG
Test: CPU Cache
NAS Parallel Benchmarks
Test / Class: LU.C
Flexible IO Tester
Type: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Sequential 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: 4KB - Disk Target: Default Test Directory
Flexible IO Tester
Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test Directory
Scikit-Learn
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
GIMP
Test: resize
NAS Parallel Benchmarks
Test / Class: FT.C
Zstd Compression
Compression Level: 19
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenMP - Kernel: Euclidean Cluster
PlaidML
FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU
Mlpack Benchmark
Benchmark: scikit_ica
glibc bench
Benchmark: exp
Rodinia
Test: OpenMP HotSpot3D
Flexible IO Tester
Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - 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: Memory Copying
Compile Bench
Test: Compile
SQLite Speedtest
Timed Time - Size 1,000
glibc bench
Benchmark: asinh
FS-Mark
Test: 1000 Files, 1MB Size
DaCapo Benchmark
Java Test: Jython
GIMP
Test: auto-levels
Core-Latency
Average Latency Between CPU Cores
Java Gradle Build
Gradle Build: Reactor
glibc bench
Benchmark: sinh
glibc bench
Benchmark: cos
GEGL
Operation: Antialias
glibc bench
Benchmark: sin
LevelDB
Benchmark: Fill Sync
GIMP
Test: unsharp-mask
libjpeg-turbo tjbench
Test: Decompression Throughput
GEGL
Operation: Rotate 90 Degrees
t-test1
Threads: 2
Git
Time To Complete Common Git Commands
Redis
Test: LPUSH
PyPerformance
Benchmark: python_startup
G'MIC
Test: Plotting Isosurface Of A 3D Volume, 1000 Times
Java SciMark
Computational Test: Jacobi Successive Over-Relaxation
PyPerformance
Benchmark: pathlib
LibreOffice
Test: 20 Documents To PDF
Numpy Benchmark
GEGL
Operation: Cartoon
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenMP - Kernel: Points2Image
t-test1
Threads: 1
Compile Bench
Test: Read Compiled Tree
GEGL
Operation: Scale
Open Porous Media
OPM Benchmark: Flow MPI Norne - Threads: 1
PyPerformance
Benchmark: 2to3
OSBench
Test: Create Threads
FS-Mark
Test: 1000 Files, 1MB Size, No Sync/FSync
Zstd Compression
Compression Level: 3
Perl Benchmarks
Test: Pod2html
Open Porous Media
OPM Benchmark: Flow MPI Norne - Threads: 2
PHP Micro Benchmarks
Test: Zend bench
GEGL
Operation: Tile Glass
Bork File Encrypter
File Encryption Time
PyPerformance
Benchmark: pickle_pure_python
PyPerformance
Benchmark: nbody
LevelDB
Benchmark: Sequential Fill
Montage Astronomical Image Mosaic Engine
Mosaic of M17, K band, 1.5 deg x 1.5 deg
PyPerformance
Benchmark: float
GIMP
Test: rotate
Redis
Test: SADD
Mlpack Benchmark
Benchmark: scikit_svm
glibc bench
Benchmark: sqrt
PHPBench
PHP Benchmark Suite
glibc bench
Benchmark: pthread_once
glibc bench
Benchmark: ffsll
glibc bench
Benchmark: ffs
Compile Bench
Test: Initial Create
PyBench
Total For Average Test Times
PyPerformance
Benchmark: raytrace
PyPerformance
Benchmark: django_template
PyPerformance
Benchmark: json_loads
Java SciMark
Computational Test: Monte Carlo
glibc bench
Benchmark: modf
PHP Micro Benchmarks
Test: Zend micro_bench
GEGL
Operation: Color Enhance
Optcarrot
Optimized Benchmark
PyPerformance
Benchmark: chaos
SQLite
Threads / Copies: 8
GEGL
Operation: Reflect
glibc bench
Benchmark: atanh
OSBench
Test: Create Files
librsvg
Operation: SVG Files To PNG
Inkscape
Operation: SVG Files To PNG
Redis
Test: SET
GEGL
Operation: Wavelet Blur
Java SciMark
Computational Test: Fast Fourier Transform
glibc bench
Benchmark: tanh
PyPerformance
Benchmark: regex_compile
glibc bench
Benchmark: sincos
PyPerformance
Benchmark: go
Go Benchmarks
Test: build
Ethr
Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 8
Ethr
Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1
glibc bench
Benchmark: log2
Coremark
CoreMark Size 666 - Iterations Per Second
High Performance Conjugate Gradient
PyPerformance
Benchmark: crypto_pyaes
CloverLeaf
Lagrangian-Eulerian Hydrodynamics
Numenta Anomaly Benchmark
Detector: EXPoSE
Java SciMark
Computational Test: Composite
G'MIC
Test: 3D Elevated Function In Random Colors, 100 Times
Redis
Test: GET
LevelDB
Benchmark: Random Fill
GEGL
Operation: Crop
Java SciMark
Computational Test: Sparse Matrix Multiply
GNU Octave Benchmark
BLAKE2
FS-Mark
Test: 5000 Files, 1MB Size, 4 Threads
OSBench
Test: Memory Allocations
Java SciMark
Computational Test: Dense LU Matrix Factorization
ctx_clock
Context Switch Time
Ethr
Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1
Darmstadt Automotive Parallel Heterogeneous Suite
Backend: OpenMP - Kernel: NDT Mapping
NAS Parallel Benchmarks
Test / Class: SP.B
Ethr
Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1
NAS Parallel Benchmarks
Test / Class: MG.C
G'MIC
Test: 2D Function Plotting, 1000 Times
Stress-NG
Test: Atomic
Ethr
Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 8
Sockperf
Test: Latency Under Load
Sockperf
Test: Latency Ping Pong
FS-Mark
Test: 4000 Files, 32 Sub Dirs, 1MB Size
SQLite
Threads / Copies: 1
Phoronix Test Suite v10.8.5