new tests may aarch64

Ampere ARMv8 Neoverse-N1 testing with a WIWYNN Mt.Jade (2.03.20210719 SCP: BIOS) and ASPEED on Ubuntu 21.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2205067-NE-NEWTESTSM45
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

C/C++ Compiler Tests 2 Tests
CPU Massive 4 Tests
Creator Workloads 3 Tests
Go Language Tests 3 Tests
HPC - High Performance Computing 2 Tests
Imaging 2 Tests
Common Kernel Benchmarks 3 Tests
Machine Learning 2 Tests
Multi-Core 2 Tests
Server 3 Tests
Server CPU Tests 2 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
A
May 06 2022
  1 Hour, 48 Minutes
B
May 06 2022
  1 Hour, 48 Minutes
Invert Hiding All Results Option
  1 Hour, 48 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


new tests may aarch64OpenBenchmarking.orgPhoronix Test SuiteAmpere ARMv8 Neoverse-N1 @ 3.00GHz (256 Cores)WIWYNN Mt.Jade (2.03.20210719 SCPAmpere Computing LLC Altra PCI Root Complex A512GB3841GB Micron_9300_MTFDHAL3T8TDP + 960GB SAMSUNG MZ1LB960HAJQ-00007ASPEEDMellanox MT28908 + Intel I210Ubuntu 21.105.13.0-27-generic (aarch64)X ServerGCC 11.2.0ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDisplay ServerCompilerFile-SystemScreen ResolutionNew Tests May Aarch64 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - Scaling Governor: cppc_cpufreq performance (Boost: Disabled)- OpenJDK Runtime Environment (build 11.0.14+9-Ubuntu-0ubuntu2.22.10)- Python 3.9.7- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

A vs. B ComparisonPhoronix Test SuiteBaseline+20.3%+20.3%+40.6%+40.6%+60.9%+60.9%81.2%29.1%27%26.2%23.3%16.3%16.2%12%8.7%8.2%8.2%6.2%5.2%4.8%2.9%2.2%C.B.S.A - f32 - CPUM.M.B.S.T - u8s8f32 - CPU54.5%R.N.N.I - u8s8f32 - CPUR.N.N.T - u8s8f32 - CPUGPT-2 - CPU - StandardC.B.S.A - u8s8f32 - CPU24.6%R.N.N.T - bf16bf16bf16 - CPUSENDFILE22.6%1ArcFace ResNet-100 - CPU - StandardR.N.N.I - bf16bf16bf16 - CPU12.5%200D.B.s - f32 - CPU11.9%M.M.B.S.T - f32 - CPU10.2%1000R.N.N.I - f32 - CPU20Memcpy 1MB7.2%CPU Cache6.6%Futex6.4%Sched PipeEpoll WaitD.B.s - u8s8f32 - CPU4.8%IP Shapes 3D - u8s8f32 - CPUyolov4 - CPU - Standard4.6%MMAP4.3%Forking3.7%Malloc3.5%1003.4%10, LosslessThroughput2.8%fcn-resnet101-11 - CPU - Standard2.7%R.N.N.T - f32 - CPU2.4%D.B.s - f32 - CPU2.4%Futex Lock-Pibertsquad-12 - CPU - Standard2.2%oneDNNoneDNNoneDNNoneDNNONNX RuntimeoneDNNoneDNNStress-NGApache HTTP ServerONNX RuntimeoneDNNApache HTTP ServeroneDNNoneDNNApache HTTP ServeroneDNNApache HTTP Serverperf-benchStress-NGStress-NGperf-benchperf-benchoneDNNoneDNNONNX RuntimeStress-NGStress-NGStress-NGApache HTTP Serverlibavif avifencJava JMHONNX RuntimeoneDNNoneDNNperf-benchONNX RuntimeAB

new tests may aarch64onednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUonnx: GPT-2 - CPU - Standardonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUstress-ng: SENDFILEapache: 1onnx: ArcFace ResNet-100 - CPU - Standardonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUapache: 200onednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUapache: 1000onednn: Recurrent Neural Network Inference - f32 - CPUapache: 20perf-bench: Memcpy 1MBstress-ng: CPU Cachestress-ng: Futexperf-bench: Sched Pipeperf-bench: Epoll Waitonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUonnx: yolov4 - CPU - Standardstress-ng: MMAPstress-ng: Forkingstress-ng: Mallocapache: 100avifenc: 10, Losslessjava-jmh: Throughputonnx: fcn-resnet101-11 - CPU - Standardonednn: Recurrent Neural Network Training - f32 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUperf-bench: Futex Lock-Pionnx: bertsquad-12 - CPU - Standardapache: 500onnx: GPT-2 - CPU - Parallelstress-ng: Memory Copyingonednn: IP Shapes 1D - f32 - CPUnginx: 1stress-ng: IO_uringonnx: bertsquad-12 - CPU - Parallelonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonnx: yolov4 - CPU - Parallelnginx: 500stress-ng: Context Switchingonnx: ArcFace ResNet-100 - CPU - Parallelonnx: super-resolution-10 - CPU - Standardavifenc: 2stress-ng: System V Message Passingonednn: IP Shapes 1D - u8s8f32 - CPUstress-ng: Cryptoavifenc: 6webp2: Defaultnginx: 200nginx: 1000onednn: IP Shapes 3D - f32 - CPUavifenc: 6, Losslessnginx: 100stress-ng: MEMFDstress-ng: NUMAstress-ng: Socket Activitystress-ng: Glibc C String Functionsperf-bench: Syscall Basicnginx: 20webp2: Quality 95, Compression Effort 7avifenc: 0stress-ng: Glibc Qsort Data Sortingwebp2: Quality 75, Compression Effort 7webp2: Quality 100, Lossless Compressionstress-ng: Semaphoresonnx: super-resolution-10 - CPU - Parallelwebp2: Quality 100, Compression Effort 5stress-ng: Matrix Mathstress-ng: Vector Mathstress-ng: CPU Stressperf-bench: Futex Hashperf-bench: Memset 1MBonnx: fcn-resnet101-11 - CPU - Parallelstress-ng: Atomiconednn: IP Shapes 1D - bf16bf16bf16 - CPUAB20.8501109.23413539.1180834235130.806172231979249.096546.582291135245403.0356.444934.424144329.512883.716782.5815.734928419.89328915.8200009106489.0685166.8962032051.8714393.7114773687.9337378.826.131493572978516.113816357.149.76984551651287.2615969674.4944.231741813.431937964.9330336.194421463406.631806812.552394242163.4372426671.24106.976333734.794.6413.44461794.3765624.2240.8157.87968830.21422.2632.3718596.3419212257.31782796183090.65318.767252.0642477.96161.268625.64718356640.2335585.1741198793723548.6665472.6632695844.7462840111.5082168.78710490.5142395345163.01413968.61614754.327614.7626612769.650848.2163.162637.935248169.541190418160.3114.680252393.71309072.4212491111993.3418159.3041941967.4313877.64110879866.1936149.795.958480270249798.563716750.350.95784650552228.9416259837.1943.734641347.931917001.5830035.85321663986.7631522925.812414276164.6752444337.47106.213331354.534.613.42362156.7765932.6440.99957.84768559.661427.3132.4718652.5519155740.91785090982864.74318.017252.6372482.09161.48624.94318374889.535555.1781198297.22723262.765456.332699744.742487401OpenBenchmarking.org

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUAB51015202520.8511.51MIN: 4.72MIN: 4.621. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPUAB4080120160200109.23168.79MIN: 103.21MIN: 160.751. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUAB3K6K9K12K15K13539.110490.5MIN: 11569.9MIN: 9172.451. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUAB4K8K12K16K20K1808314239MIN: 14969.5MIN: 13174.41. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: StandardAB11002200330044005500423553451. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUAB4080120160200130.81163.01MIN: 126.78MIN: 155.371. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUAB4K8K12K16K20K17223.013968.6MIN: 14988MIN: 12588.71. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: SENDFILEAB400K800K1200K1600K2000K1979249.091614754.321. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 1AB160032004800640080006546.587614.761. (CC) gcc options: -shared -fPIC -O2

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: StandardAB601201802403002292661. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUAB3K6K9K12K15K11352.012769.6MIN: 10383.3MIN: 10807.21. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 200AB11K22K33K44K55K45403.0350848.211. (CC) gcc options: -shared -fPIC -O2

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUAB142842567056.4463.16MIN: 42.46MIN: 47.631. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUAB91827364534.4237.94MIN: 30.23MIN: 33.641. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 1000AB10K20K30K40K50K44329.5048169.541. (CC) gcc options: -shared -fPIC -O2

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUAB3K6K9K12K15K12883.711904.0MIN: 10111.2MIN: 104601. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 20AB4K8K12K16K20K16782.5818160.311. (CC) gcc options: -shared -fPIC -O2

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBAB4812162015.7314.681. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU CacheAB90180270360450419.89393.711. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: FutexAB70K140K210K280K350K328915.8309072.41. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeAB50K100K150K200K250K2000092124911. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitAB2004006008001000106411191. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUAB2040608010089.0793.34MIN: 61.45MIN: 59.721. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUAB4080120160200166.90159.30MIN: 163.69MIN: 156.651. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: StandardAB40801201602002031941. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: MMAPAB4008001200160020002051.871967.431. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: ForkingAB3K6K9K12K15K14393.7013877.641. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: MallocAB20M40M60M80M100M114773687.93110879866.191. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 100AB8K16K24K32K40K37378.8236149.791. (CC) gcc options: -shared -fPIC -O2

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 10, LosslessAB2468106.1315.9581. (CXX) g++ options: -O3 -fPIC -lm

Java JMH

This very basic test profile runs the stock benchmark of the Java JMH benchmark via Maven. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/s, More Is BetterJava JMHThroughputAB110000M220000M330000M440000M550000M493572978516.11480270249798.56

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: StandardAB91827364538371. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUAB4K8K12K16K20K16357.116750.3MIN: 13765MIN: 13295.31. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUAB112233445549.7750.96MIN: 32.58MIN: 35.781. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiAB102030405045461. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: StandardAB1102203304405505165051. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 500AB11K22K33K44K55K51287.2652228.941. (CC) gcc options: -shared -fPIC -O2

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: ParallelAB30060090012001500159616251. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Memory CopyingAB2K4K6K8K10K9674.499837.191. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUAB102030405044.2343.73MIN: 35.62MIN: 36.571. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1AB9K18K27K36K45K41813.4341347.931. (CC) gcc options: -lcrypt -lz -O3 -march=native

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: IO_uringAB400K800K1200K1600K2000K1937964.931917001.581. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: ParallelAB701402102803503033001. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUAB81624324036.1935.85MIN: 34.9MIN: 33.871. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: yolov4 - Device: CPU - Executor: ParallelAB501001502002502142161. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 500AB14K28K42K56K70K63406.6063986.761. (CC) gcc options: -lcrypt -lz -O3 -march=native

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Context SwitchingAB7M14M21M28M35M31806812.5531522925.811. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: ParallelAB501001502002502392411. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: StandardAB9001800270036004500424242761. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 2AB4080120160200163.44164.681. (CXX) g++ options: -O3 -fPIC -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: System V Message PassingAB500K1000K1500K2000K2500K2426671.242444337.471. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUAB20406080100106.98106.21MIN: 87.35MIN: 85.761. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CryptoAB70K140K210K280K350K333734.79331354.531. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6AB1.04422.08843.13264.17685.2214.6414.6101. (CXX) g++ options: -O3 -fPIC -lm

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: DefaultAB0.77491.54982.32473.09963.87453.4443.4231. (CXX) g++ options: -fno-rtti -O3 -ldl

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 200AB13K26K39K52K65K61794.3762156.771. (CC) gcc options: -lcrypt -lz -O3 -march=native

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1000AB14K28K42K56K70K65624.2265932.641. (CC) gcc options: -lcrypt -lz -O3 -march=native

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.6Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUAB91827364540.8241.00MIN: 38.09MIN: 36.511. (CXX) g++ options: -O3 -march=native -fopenmp -mcpu=native -fPIC -std=c++11 -pie -ldl -lpthread

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6, LosslessAB2468107.8797.8471. (CXX) g++ options: -O3 -fPIC -lm

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 100AB15K30K45K60K75K68830.2068559.661. (CC) gcc options: -lcrypt -lz -O3 -march=native

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: MEMFDAB300600900120015001422.261427.311. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: NUMAAB81624324032.3732.471. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Socket ActivityAB4K8K12K16K20K18596.3418652.551. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Glibc C String FunctionsAB4M8M12M16M20M19212257.3119155740.911. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicAB2M4M6M8M10M782796178509091. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 20AB20K40K60K80K100K83090.6582864.741. (CC) gcc options: -lcrypt -lz -O3 -march=native

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 95, Compression Effort 7AB70140210280350318.77318.021. (CXX) g++ options: -fno-rtti -O3 -ldl

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 0AB60120180240300252.06252.641. (CXX) g++ options: -O3 -fPIC -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Glibc Qsort Data SortingAB50010001500200025002477.962482.091. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 75, Compression Effort 7AB4080120160200161.27161.481. (CXX) g++ options: -fno-rtti -O3 -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 100, Lossless CompressionAB140280420560700625.65624.941. (CXX) g++ options: -fno-rtti -O3 -ldl

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: SemaphoresAB4M8M12M16M20M18356640.2318374889.501. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: ParallelAB8001600240032004000355835551. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20220422Encode Settings: Quality 100, Compression Effort 5AB1.16512.33023.49534.66045.82555.1745.1781. (CXX) g++ options: -fno-rtti -O3 -ldl

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Matrix MathAB300K600K900K1200K1500K1198793.001198297.221. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Vector MathAB150K300K450K600K750K723548.66723262.701. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU StressAB14K28K42K56K70K65472.6665456.301. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashAB70K140K210K280K350K3269583269971. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBAB102030405044.7544.741. (CC) gcc options: -pthread -shared -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: ParallelAB91827364540401. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: AtomicAB0.2250.450.6750.91.125111. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000

A: The test quit with a non-zero exit status. E: unable to connect to "http://localhost:8086": Get "http://localhost:8086/ping": dial tcp 127.0.0.1:8086: connect: connection refused

B: The test quit with a non-zero exit status. E: unable to connect to "http://localhost:8086": Get "http://localhost:8086/ping": dial tcp 127.0.0.1:8086: connect: connection refused

Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000

A: The test quit with a non-zero exit status. E: unable to connect to "http://localhost:8086": Get "http://localhost:8086/ping": dial tcp 127.0.0.1:8086: connect: connection refused

B: The test quit with a non-zero exit status. E: unable to connect to "http://localhost:8086": Get "http://localhost:8086/ping": dial tcp 127.0.0.1:8086: connect: connection refused

Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000

A: The test quit with a non-zero exit status. E: unable to connect to "http://localhost:8086": Get "http://localhost:8086/ping": dial tcp 127.0.0.1:8086: connect: connection refused

B: The test quit with a non-zero exit status. E: unable to connect to "http://localhost:8086": Get "http://localhost:8086/ping": dial tcp 127.0.0.1:8086: connect: connection refused

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.

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

A: The test run did not produce a result.

B: The test run did not produce a result.

Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU

A: The test run did not produce a result.

B: The test run did not produce a result.

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

A: The test run did not produce a result.

B: The test run did not produce a result.

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

A: The test run did not produce a result.

B: The test run did not produce a result.

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

A: The test run did not produce a result.

B: The test run did not produce a result.

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

A: The test run did not produce a result.

B: The test run did not produce a result.

81 Results Shown

oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPU
  Recurrent Neural Network Inference - u8s8f32 - CPU
  Recurrent Neural Network Training - u8s8f32 - CPU
ONNX Runtime
oneDNN:
  Convolution Batch Shapes Auto - u8s8f32 - CPU
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
Stress-NG
Apache HTTP Server
ONNX Runtime
oneDNN
Apache HTTP Server
oneDNN:
  Deconvolution Batch shapes_1d - f32 - CPU
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
Apache HTTP Server
oneDNN
Apache HTTP Server
perf-bench
Stress-NG:
  CPU Cache
  Futex
perf-bench:
  Sched Pipe
  Epoll Wait
oneDNN:
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
  IP Shapes 3D - u8s8f32 - CPU
ONNX Runtime
Stress-NG:
  MMAP
  Forking
  Malloc
Apache HTTP Server
libavif avifenc
Java JMH
ONNX Runtime
oneDNN:
  Recurrent Neural Network Training - f32 - CPU
  Deconvolution Batch shapes_3d - f32 - CPU
perf-bench
ONNX Runtime
Apache HTTP Server
ONNX Runtime
Stress-NG
oneDNN
nginx
Stress-NG
ONNX Runtime
oneDNN
ONNX Runtime
nginx
Stress-NG
ONNX Runtime:
  ArcFace ResNet-100 - CPU - Parallel
  super-resolution-10 - CPU - Standard
libavif avifenc
Stress-NG
oneDNN
Stress-NG
libavif avifenc
WebP2 Image Encode
nginx:
  200
  1000
oneDNN
libavif avifenc
nginx
Stress-NG:
  MEMFD
  NUMA
  Socket Activity
  Glibc C String Functions
perf-bench
nginx
WebP2 Image Encode
libavif avifenc
Stress-NG
WebP2 Image Encode:
  Quality 75, Compression Effort 7
  Quality 100, Lossless Compression
Stress-NG
ONNX Runtime
WebP2 Image Encode
Stress-NG:
  Matrix Math
  Vector Math
  CPU Stress
perf-bench:
  Futex Hash
  Memset 1MB
ONNX Runtime
Stress-NG