onnx 1.11 Threadripper

AMD Ryzen Threadripper 3990X 64-Core testing with a Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS) and AMD Radeon RX 5700 8GB on Pop 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 2203274-PTS-ONNX111T03
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

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
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
Performance Per
Dollar
Date
Run
  Test
  Duration
A
March 26 2022
  48 Minutes
B
March 27 2022
  53 Minutes
C
March 27 2022
  53 Minutes
D
March 27 2022
  49 Minutes
Invert Hiding All Results Option
  51 Minutes

Only show results where is faster than
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):


onnx 1.11 ThreadripperOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads)Gigabyte TRX40 AORUS PRO WIFI (F4p BIOS)AMD Starship/Matisse128GBSamsung SSD 970 EVO Plus 500GBAMD Radeon RX 5700 8GB (1750/875MHz)AMD Navi 10 HDMI AudioDELL P2415QIntel I211 + Intel Wi-Fi 6 AX200Pop 21.105.17.0-rc1-sched-core-phx (x86_64)GNOME Shell 40.5X Server4.6 Mesa 21.2.2 (LLVM 12.0.1)1.2.182GCC 11.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionOnnx 1.11 Threadripper BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x8301039- 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

ABCDResult OverviewPhoronix Test Suite100%102%104%105%107%perf-benchONNX RuntimeONNX Runtimeperf-benchONNX Runtimeperf-benchONNX Runtimeperf-benchperf-benchONNX Runtimeperf-benchperf-benchONNX RuntimeSched PipeGPT-2 - CPUsuper-resolution-10 - CPUEpoll Waityolov4 - CPUFutex Hashbertsquad-12 - CPUSyscall BasicMemcpy 1MBArcFace ResNet-100 - CPUMemset 1MBFutex Lock-Pifcn-resnet101-11 - CPU

onnx 1.11 Threadripperperf-bench: Epoll Waitperf-bench: Futex Hashperf-bench: Memcpy 1MBperf-bench: Memset 1MBperf-bench: Sched Pipeperf-bench: Futex Lock-Piperf-bench: Syscall Basiconnx: GPT-2 - CPUonnx: yolov4 - CPUonnx: bertsquad-12 - CPUonnx: fcn-resnet101-11 - CPUonnx: ArcFace ResNet-100 - CPUonnx: super-resolution-10 - CPUABCD5044355328512.07873769.6375442554879316605197412038043581109636345166357839011.95903469.5552292467459316727281385036842681109236625158348336912.08627769.4945422564739316659267386837043181108636925213354056311.97361769.496049264487931687814238693694318110923787OpenBenchmarking.org

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: Epoll WaitABCD11002200330044005500SE +/- 60.84, N = 4SE +/- 56.49, N = 3SE +/- 24.06, N = 3SE +/- 4.63, N = 350445166515852131. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitABCD9001800270036004500Min: 4941 / Avg: 5044 / Max: 5218Min: 5054 / Avg: 5166.33 / Max: 5233Min: 5131 / Avg: 5158 / Max: 5206Min: 5205 / Avg: 5213.33 / Max: 52211. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashABCD800K1600K2400K3200K4000KSE +/- 9635.16, N = 3SE +/- 21250.47, N = 3SE +/- 10646.17, N = 3SE +/- 7910.60, N = 335532853578390348336935405631. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashABCD600K1200K1800K2400K3000KMin: 3543286 / Avg: 3553285.33 / Max: 3572551Min: 3549444 / Avg: 3578390.33 / Max: 3619814Min: 3462821 / Avg: 3483369 / Max: 3498476Min: 3530495 / Avg: 3540562.67 / Max: 35561661. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBABCD3691215SE +/- 0.04, N = 3SE +/- 0.13, N = 15SE +/- 0.08, N = 13SE +/- 0.13, N = 312.0811.9612.0911.971. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBABCD48121620Min: 12 / Avg: 12.08 / Max: 12.13Min: 11.54 / Avg: 11.96 / Max: 13.47Min: 11.64 / Avg: 12.09 / Max: 12.63Min: 11.72 / Avg: 11.97 / Max: 12.151. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBABCD1530456075SE +/- 0.56, N = 9SE +/- 0.56, N = 3SE +/- 0.81, N = 15SE +/- 0.65, N = 1569.6469.5669.4969.501. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBABCD1326395265Min: 67.23 / Avg: 69.64 / Max: 73.46Min: 68.86 / Avg: 69.56 / Max: 70.66Min: 65.92 / Avg: 69.49 / Max: 77.14Min: 64.23 / Avg: 69.5 / Max: 73.681. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeABCD60K120K180K240K300KSE +/- 13951.36, N = 12SE +/- 10178.25, N = 15SE +/- 11466.52, N = 15SE +/- 12339.61, N = 152554872467452564732644871. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeABCD50K100K150K200K250KMin: 214227 / Avg: 255487 / Max: 360500Min: 211392 / Avg: 246745.13 / Max: 314048Min: 214593 / Avg: 256472.73 / Max: 325588Min: 212115 / Avg: 264486.93 / Max: 3385691. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiABCD20406080100SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3939393931. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiABCD20406080100Min: 93 / Avg: 93 / Max: 93Min: 92 / Avg: 92.67 / Max: 93Min: 93 / Avg: 93 / Max: 93Min: 92 / Avg: 92.67 / Max: 931. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicABCD4M8M12M16M20MSE +/- 92684.37, N = 3SE +/- 124978.04, N = 3SE +/- 167024.68, N = 6SE +/- 14405.48, N = 3166051971672728116659267168781421. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicABCD3M6M9M12M15MMin: 16486336 / Avg: 16605196.67 / Max: 16787815Min: 16477963 / Avg: 16727281.33 / Max: 16867394Min: 15829229 / Avg: 16659267.33 / Max: 16862126Min: 16851411 / Avg: 16878141.67 / Max: 169008161. (CC) gcc options: -pthread -shared -lunwind-x86_64 -lunwind -llzma -Xlinker -O6 -ggdb3 -funwind-tables -std=gnu99 -lnuma

ONNX Runtime

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPUABCD9001800270036004500SE +/- 14.70, N = 3SE +/- 34.56, N = 3SE +/- 10.15, N = 3SE +/- 1.15, N = 341203850386838691. (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: GPT-2 - Device: CPUABCD7001400210028003500Min: 4099.5 / Avg: 4120 / Max: 4148.5Min: 3781.5 / Avg: 3849.83 / Max: 3893Min: 3851.5 / Avg: 3868 / Max: 3886.5Min: 3867 / Avg: 3869 / Max: 38711. (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: yolov4 - Device: CPUABCD80160240320400SE +/- 1.26, N = 3SE +/- 0.87, N = 3SE +/- 0.44, N = 3SE +/- 0.76, N = 33803683703691. (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: yolov4 - Device: CPUABCD70140210280350Min: 378.5 / Avg: 380 / Max: 382.5Min: 366.5 / Avg: 368 / Max: 369.5Min: 369 / Avg: 369.67 / Max: 370.5Min: 367.5 / Avg: 368.5 / Max: 3701. (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: bertsquad-12 - Device: CPUABCD90180270360450SE +/- 1.32, N = 3SE +/- 0.73, N = 3SE +/- 0.93, N = 3SE +/- 0.93, N = 34354264314311. (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: bertsquad-12 - Device: CPUABCD80160240320400Min: 432.5 / Avg: 434.5 / Max: 437Min: 424.5 / Avg: 425.67 / Max: 427Min: 429 / Avg: 430.83 / Max: 432Min: 429 / Avg: 430.83 / Max: 4321. (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: fcn-resnet101-11 - Device: CPUABCD20406080100SE +/- 0.29, N = 3SE +/- 0.17, N = 3SE +/- 0.17, N = 3SE +/- 0.17, N = 3818181811. (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: fcn-resnet101-11 - Device: CPUABCD1530456075Min: 80.5 / Avg: 81 / Max: 81.5Min: 80.5 / Avg: 80.67 / Max: 81Min: 80.5 / Avg: 80.83 / Max: 81Min: 81 / Avg: 81.17 / Max: 81.51. (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: ArcFace ResNet-100 - Device: CPUABCD2004006008001000SE +/- 7.58, N = 3SE +/- 3.77, N = 3SE +/- 14.26, N = 3SE +/- 2.68, N = 310961092108610921. (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: ArcFace ResNet-100 - Device: CPUABCD2004006008001000Min: 1081 / Avg: 1095.83 / Max: 1106Min: 1084.5 / Avg: 1092 / Max: 1096.5Min: 1071 / Avg: 1086 / Max: 1114.5Min: 1088.5 / Avg: 1092.33 / Max: 1097.51. (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: CPUABCD8001600240032004000SE +/- 24.28, N = 3SE +/- 7.29, N = 3SE +/- 17.99, N = 3SE +/- 9.97, N = 336343662369237871. (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: CPUABCD7001400210028003500Min: 3585 / Avg: 3633.5 / Max: 3660Min: 3648 / Avg: 3661.5 / Max: 3673Min: 3658 / Avg: 3692.17 / Max: 3719Min: 3769 / Avg: 3786.67 / Max: 3803.51. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt