axs2mlperf

This is a facilitator for the MLPerf Inference Benchmark Suite leveraging the axs2mlperf Docker container build and testing for currently facilitating ResNwt-50 reference model inference CPU benchmarks. See reference information at https://github.com/krai/axs2mlperf/blob/master/demo/README.md

To run this test with the Phoronix Test Suite, the basic command is: phoronix-test-suite benchmark axs2mlperf.

Project Site

mlcommons.org

Source Repository

github.com

Test Created

12 May 2023

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

1 Minute, 55 Seconds

Average Run Time

1 Hour, 52 Minutes, 44 Seconds

Test Dependencies

Python

Accolades

10k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page ViewsOpenBenchmarking.orgEventsaxs2mlperf Popularity Statisticspts/axs2mlperf2023.052023.062023.072023.082023.092023.102023.112023.122024.012024.022024.032024.042024.052024.062024.072024.082024.092024.102024.115001000150020002500
* 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 21 November 2024.

Revision History

pts/axs2mlperf-1.0.0   [View Source]   Fri, 12 May 2023 10:08:02 GMT
Initial commit for MLPerf benchmark.


Performance Metrics

Analyze Test Configuration:

axs2mlperf

Test: ResNet50 MLPerf Reference Model

OpenBenchmarking.org metrics for this test profile configuration based on 29 public results since 12 May 2023 with the latest data as of 19 May 2023.

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
Samples Per Second (Average)
Mid-Tier
75th
< 121
Median
50th
53
Low-Tier
25th
< 28
OpenBenchmarking.orgDistribution Of Public Results - Test: ResNet50 MLPerf Reference Model29 Results Range From 25 To 164 Samples Per Second253749617385971091211331451571693691215

Based on OpenBenchmarking.org data, the selected test / test configuration (axs2mlperf - Test: ResNet50 MLPerf Reference Model) has an average run-time of 1 hour, 43 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 BenchmarkTest: ResNet50 MLPerf Reference ModelRun-Time20406080100Min: 70 / Avg: 102.78 / Max: 122

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 2.3%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsTest: ResNet50 MLPerf Reference ModelDeviation48121620Min: 0 / Avg: 2.33 / Max: 14

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)