XNNPACK

XNNPACK is a Google library for high efficiency floating-point neural network inference operators across mobile / server / web use. XNNPACK is used by machine learning frameworks like TensorFlow, PyTorch, ONNX Runtime, MediaPipe, and others. This test profile uses XNNPACK with its bench-models benchmark and testing all available CPU threads.

To run this test with the Phoronix Test Suite, the basic command is: phoronix-test-suite benchmark xnnpack.

Project Site

github.com

Source Repository

github.com

Test Created

11 August 2024

Last Updated

15 October 2024

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

6 Minutes, 8 Seconds

Average Run Time

11 Minutes, 43 Seconds

Test Dependencies

CMake + C/C++ Compiler Toolchain

Accolades

5k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page ViewsOpenBenchmarking.orgEventsXNNPACK Popularity Statisticspts/xnnpack2024.082024.092024.102024.112024.127001400210028003500
* Uploading of benchmark result data to OpenBenchmarking.org is always optional (opt-in) via the Phoronix Test Suite for users wishing to share their results publicly.
** Data based on those opting to upload their test results to OpenBenchmarking.org and users enabling the opt-in anonymous statistics reporting while running benchmarks from an Internet-connected platform.
Data updated weekly as of 30 December 2024.
QS8MobileNetV211.1%FP32MobileNetV3Small11.1%FP16MobileNetV211.1%FP16MobileNetV3Large11.0%FP32MobileNetV3Large11.0%FP16MobileNetV111.0%FP16MobileNetV3Small11.1%FP32MobileNetV211.1%FP32MobileNetV111.1%Model Option PopularityOpenBenchmarking.org

Revision History

pts/xnnpack-1.1.0   [View Source]   Tue, 15 Oct 2024 15:30:38 GMT
Update against XNNPACK upstream, switch to new benchmark.

pts/xnnpack-1.0.0   [View Source]   Sun, 11 Aug 2024 10:34:29 GMT
Add XNNPACK benchmark.

Suites Using This Test

Machine Learning

HPC - High Performance Computing


Performance Metrics

Analyze Test Configuration:

XNNPACK b7b048

Model: QS8MobileNetV2

OpenBenchmarking.org metrics for this test profile configuration based on 211 public results since 15 October 2024 with the latest data as of 30 December 2024.

Below is an overview of the generalized performance for components where there is sufficient statistically significant data based upon user-uploaded results. It is important to keep in mind particularly in the Linux/open-source space there can be vastly different OS configurations, with this overview intended to offer just general guidance as to the performance expectations.

Component
Percentile Rank
# Compatible Public Results
us (Average)
95th
3
547 +/- 30
92nd
5
624 +/- 3
78th
4
781 +/- 6
Mid-Tier
75th
> 801
72nd
7
836 +/- 29
68th
3
932 +/- 131
65th
4
984 +/- 53
63rd
3
1097 +/- 12
60th
4
1145 +/- 18
57th
3
1239 +/- 7
54th
4
1409 +/- 5
Median
50th
1435
50th
3
1451 +/- 37
46th
4
1866 +/- 26
45th
3
1939 +/- 3
39th
3
2071 +/- 285
38th
4
2092 +/- 39
37th
11
2117 +/- 126
32nd
3
2480 +/- 23
Low-Tier
25th
> 5032
14th
21
10163 +/- 686
11th
3
16276 +/- 298
Detailed Performance Overview
OpenBenchmarking.orgDistribution Of Public Results - Model: QS8MobileNetV2202 Results Range From 302 To 480430 us30299051950829111387144831757920675237712686729963321059351155381251411347441443471539501635531731561827591923622019652115682211712307742403772499802595832691862787892883922979953075983172013268043364073460103556133652163748193844223940254036284132314228344324374420404516434612464708494804524080120160200

Based on OpenBenchmarking.org data, the selected test / test configuration (XNNPACK b7b048 - Model: QS8MobileNetV2) has an average run-time of 16 minutes. By default this test profile is set to run at least 3 times but may increase if the standard deviation exceeds pre-defined defaults or other calculations deem additional runs necessary for greater statistical accuracy of the result.

OpenBenchmarking.orgMinutesTime Required To Complete BenchmarkModel: QS8MobileNetV2Run-Time918273645Min: 9 / Avg: 15.4 / Max: 45

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 0.5%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsModel: QS8MobileNetV2Deviation48121620Min: 0 / Avg: 0.47 / Max: 13

Tested CPU Architectures

This benchmark has been successfully tested on the below mentioned architectures. The CPU architectures listed is where successful OpenBenchmarking.org result uploads occurred, namely for helping to determine if a given test is compatible with various alternative CPU architectures.

CPU Architecture
Kernel Identifier
Verified On
Intel / AMD x86 64-bit
x86_64
(Many Processors)
ARMv8 64-bit
aarch64
ARMv8 Neoverse-N1 128-Core, ARMv8 Neoverse-V2, ARMv8 Neoverse-V2 72-Core