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 the Intel oneAPI initiative.

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

Project Site

github.com

Test Created

17 June 2020

Last Updated

20 December 2020

Test Maintainer

Michael Larabel

Test Type

Processor

Average Install Time

8 Minutes, 59 Seconds

Average Run Time

2 Minutes, 2 Seconds

Accolades

5k+ Downloads

Supported Platforms


Public Result UploadsReported Installs*Test Completions*OpenBenchmarking.orgEventsoneDNN Popularity Statisticspts/onednn2020.062020.072020.082020.092020.102020.112020.122021.014K8K12K16K20K
* 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 current as of Fri, 22 Jan 2021 18:34:28 GMT.
Deconvolution Batch shapes_3d11.4%Recurrent Neural Network Training15.9%IP Shapes 1D11.5%Convolution Batch Shapes Auto11.5%Matrix Multiply Batch Shapes Transformer11.3%Recurrent Neural Network Inference15.7%IP Shapes 3D11.5%Deconvolution Batch shapes_1d11.4%Harness Option PopularityOpenBenchmarking.org
bf16bf16bf1614.0%u8s8f3240.1%f3245.8%Data Type Option PopularityOpenBenchmarking.org

Revision History

pts/onednn-1.6.1   [View Source]   Sun, 20 Dec 2020 09:58:16 GMT
This test profile builds and works fine on macOS so enable it (MacOSX).

pts/onednn-1.6.0   [View Source]   Wed, 09 Dec 2020 13:47:31 GMT
Update against oneDNN 2.0 upstream.

pts/onednn-1.5.0   [View Source]   Wed, 17 Jun 2020 16:26:39 GMT
Initial commit of oneDNN test profile based on Intel oneDNN 1.5, forked from existing mkl-dnn test profile that is named from MKL-DNN before it was renamed to DNNL and then oneDNN. So create new test profile to match Intel naming convention.

Suites Using This Test

Multi-Core

Machine Learning

Intel oneAPI

CPU Massive

Server CPU Tests

Creator Workloads

HPC - High Performance Computing


Performance Metrics

Analyze Test Configuration:

oneDNN 1.5

Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU

OpenBenchmarking.org metrics for this test profile configuration based on 554 public results since 17 June 2020 with the latest data as of 8 December 2020.

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
# Matching Public Results
ms (Average)
98th
4
0.32 +/- 0.03
97th
3
0.38 +/- 0.02
94th
4
0.43 +/- 0.01
89th
4
0.51 +/- 0.01
89th
4
0.51 +/- 0.01
85th
8
0.53 +/- 0.02
79th
10
0.61 +/- 0.10
79th
6
0.62 +/- 0.11
Mid-Tier
75th
> 0.64
74th
5
0.67 +/- 0.09
59th
7
0.82 +/- 0.03
59th
5
0.83 +/- 0.01
56th
4
0.93 +/- 0.01
56th
10
0.94 +/- 0.16
54th
6
1.00 +/- 0.21
53rd
6
1.09 +/- 0.32
Median
50th
1.23
49th
5
1.34 +/- 0.26
46th
4
1.44 +/- 0.06
45th
19
1.45 +/- 0.09
41st
7
1.55 +/- 0.02
38th
4
1.80 +/- 0.04
36th
3
2.09 +/- 0.01
34th
11
2.17 +/- 0.45
33rd
5
2.32 +/- 0.03
32nd
6
2.43 +/- 0.07
30th
4
2.76 +/- 0.16
27th
9
3.62 +/- 0.68
26th
4
3.77 +/- 0.39
Low-Tier
25th
> 3.85
24th
7
3.97 +/- 0.06
21st
3
4.16 +/- 0.04
21st
6
4.28 +/- 0.35
20th
4
4.38 +/- 0.02
17th
4
4.78 +/- 0.11
16th
3
4.94 +/- 0.36
15th
3
5.08 +/- 0.06
15th
3
5.12 +/- 0.09
12th
4
5.42 +/- 0.36
12th
3
5.46 +/- 0.06
12th
3
5.69 +/- 0.02
11th
6
5.81 +/- 0.44
11th
3
5.92 +/- 0.06
10th
3
6.13 +/- 0.13
9th
3
6.26 +/- 0.04
6th
4
8.36 +/- 1.53
6th
3
8.73 +/- 0.01
4th
3
9.85 +/- 0.47
4th
3
10.70 +/- 0.41
2nd
3
16.26 +/- 0.73
OpenBenchmarking.orgDistribution Of Public Results - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU554 Results Range From 0 To 96 ms081624324048566472808896110220330440550

Based on OpenBenchmarking.org data, the selected test / test configuration (oneDNN 1.5 - Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU) has an average run-time of 2 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 BenchmarkHarness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPURun-Time246810Min: 1 / Avg: 1.25 / Max: 3

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 0.8%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsHarness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPUDeviation3691215Min: 0 / Avg: 0.81 / Max: 9

Recent Test Results

OpenBenchmarking.org Results Compare

3 Systems - 74 Benchmark Results

2 x AMD EPYC 7601 32-Core - Dell 02MJ3T - AMD 17h

Ubuntu 19.10 - 5.9.0-050900rc6daily20200922-generic - GNOME Shell 3.34.1

2 Systems - 129 Benchmark Results

AMD Ryzen 9 5950X 16-Core - ASUS ROG CROSSHAIR VIII HERO - AMD Starship

Ubuntu 20.10 - 5.11.0-051100rc2daily20210108-generic - GNOME Shell 3.38.1

3 Systems - 191 Benchmark Results

AMD Ryzen 3 2200G - ASUS PRIME B350M-E - AMD Raven

Ubuntu 20.10 - 5.8.0-38-generic - GNOME Shell 3.38.1

4 Systems - 104 Benchmark Results

AMD Ryzen 5 2400G - MSI B350M GAMING PRO - AMD Raven

Ubuntu 19.10 - 5.3.0-64-generic - GNOME Shell 3.34.1

4 Systems - 61 Benchmark Results

Intel Core i9-7960X - MSI X299 SLI PLUS - Intel Sky Lake-E DMI3 Registers

Ubuntu 20.04 - 5.4.0-58-generic - X Server 1.20.8

3 Systems - 113 Benchmark Results

Intel Core i9-7980XE - ASUS PRIME X299-A - Intel Sky Lake-E DMI3 Registers

Ubuntu 20.10 - 5.8.0-36-generic - GNOME Shell 3.38.1

3 Systems - 253 Benchmark Results

Intel Core i9-10885H - HP 8736 - Intel Comet Lake PCH

Ubuntu 20.04 - 5.6.0-1034-oem - GNOME Shell 3.36.4

4 Systems - 210 Benchmark Results

POWER9 - PowerNV T2P9D01 REV 1.01 - 64GB

Ubuntu 20.10 - 5.9.10-050910-generic - X Server

1 System - 219 Benchmark Results

AMD EPYC 7302P 16-Core - Supermicro H11SSL-i v2.00 - AMD Starship

Ubuntu 20.04 - 5.4.0-42-generic - GNOME Shell 3.36.4

2 Systems - 183 Benchmark Results

AMD FX-8370 Eight-Core - MSI 970 GAMING - AMD RD9x0

Ubuntu 20.10 - 5.8.0-33-generic - GNOME Shell 3.38.1

Most Popular Test Results

Find More Test Results

OpenBenchmarking.org Community User Comments

Post A Comment