Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba.

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

Project Site

mnn.zone

Source Repository

github.com

Test Created

17 September 2020

Last Updated

18 June 2021

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

1 Minute, 31 Seconds

Average Run Time

8 Minutes, 24 Seconds

Test Dependencies

CMake + C/C++ Compiler Toolchain

Accolades

10k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventsMobile Neural Network Popularity Statisticspts/mnn2020.092020.102020.112020.122021.012021.022021.032021.042021.052021.062021.072021.082021.092021.1010002000300040005000
* 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.
*** Test profile page view reporting began March 2021.
Data current as of 27 October 2021.
MobileNetV2_22414.3%inception-v314.2%mobilenet-v1-1.014.3%SqueezeNetV1.014.2%mobilenetV314.2%squeezenetv1.114.4%resnet-v2-5014.3%Model Option PopularityOpenBenchmarking.org

Revision History

pts/mnn-1.3.0   [View Source]   Fri, 18 Jun 2021 06:27:34 GMT
Update against new upstream MNN 1.2.0 release.

pts/mnn-1.2.0   [View Source]   Fri, 12 Mar 2021 07:05:09 GMT
Update against upstream MNN 1.1.3.

pts/mnn-1.1.1   [View Source]   Tue, 12 Jan 2021 16:25:37 GMT
Test builds fine on macOS.

pts/mnn-1.1.0   [View Source]   Wed, 06 Jan 2021 12:46:43 GMT
Update against MNN 1.1.1 upstream.

pts/mnn-1.0.1   [View Source]   Thu, 17 Sep 2020 20:25:29 GMT
Add min/max reporting to result parser.

pts/mnn-1.0.0   [View Source]   Thu, 17 Sep 2020 18:57:32 GMT
Initial commit of Alibaba MNN deep learning framework benchmark.

Suites Using This Test

Machine Learning

HPC - High Performance Computing


Performance Metrics

Analyze Test Configuration:

Mobile Neural Network 1.2

Model: squeezenetv1.1

OpenBenchmarking.org metrics for this test profile configuration based on 323 public results since 18 June 2021 with the latest data as of 24 October 2021.

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
ms (Average)
99th
9
2.32 +/- 0.03
97th
3
2.40 +/- 0.03
95th
8
2.75 +/- 0.10
94th
3
2.88 +/- 0.01
93rd
3
2.91 +/- 0.01
92nd
3
2.99 +/- 0.03
90th
3
3.11 +/- 0.06
90th
3
3.12 +/- 0.01
89th
3
3.26 +/- 0.05
87th
6
3.59 +/- 0.15
76th
3
3.69 +/- 0.03
Mid-Tier
75th
> 3.72
73rd
15
3.82 +/- 0.16
68th
3
4.07 +/- 0.08
66th
3
4.12 +/- 0.04
64th
4
4.22 +/- 0.25
64th
16
4.23 +/- 0.15
62nd
3
4.24 +/- 0.01
60th
4
4.28 +/- 0.04
58th
3
4.32 +/- 0.02
55th
16
4.40 +/- 0.26
55th
3
4.40 +/- 0.05
55th
8
4.42 +/- 0.03
Median
50th
4.47
50th
8
4.53 +/- 0.22
50th
4
4.54 +/- 0.05
46th
3
4.59 +/- 0.03
44th
6
4.62 +/- 0.20
43rd
3
4.65 +/- 0.03
42nd
4
4.67 +/- 0.15
41st
3
4.70 +/- 0.04
40th
8
4.74 +/- 0.21
38th
5
4.85 +/- 0.16
33rd
3
5.22 +/- 0.16
31st
6
5.33 +/- 0.05
29th
11
5.35 +/- 0.20
29th
3
5.36 +/- 0.02
Low-Tier
25th
> 5.63
25th
6
5.63 +/- 0.20
21st
3
5.87 +/- 0.11
21st
3
5.88 +/- 0.03
20th
3
5.99 +/- 0.29
20th
5
6.02 +/- 0.41
18th
3
6.51 +/- 0.20
17th
3
6.52 +/- 0.06
16th
4
6.61 +/- 0.05
14th
3
6.79 +/- 0.02
13th
3
7.11 +/- 0.23
13th
3
7.22 +/- 0.19
11th
4
8.16 +/- 0.44
9th
3
9.19 +/- 0.07
6th
3
10.36 +/- 0.38
6th
4
10.60 +/- 0.83
5th
3
11.20 +/- 0.71
4th
3
11.47 +/- 0.10
3rd
3
12.18 +/- 0.05
2nd
4
17.31 +/- 1.89
OpenBenchmarking.orgDistribution Of Public Results - Model: squeezenetv1.1318 Results Range From 2 To 27 ms258111417202326293235384080120160200

Based on OpenBenchmarking.org data, the selected test / test configuration (Mobile Neural Network 1.2 - Model: squeezenetv1.1) has an average run-time of 8 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: squeezenetv1.1Run-Time714212835Min: 1 / Avg: 7.46 / Max: 35

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 1.2%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsModel: squeezenetv1.1Deviation3691215Min: 0 / Avg: 1.22 / Max: 8

Does It Scale Well With Increasing Cores?

No, based on the automated analysis of the collected public benchmark data, this test / test settings does not generally scale well with increasing CPU core counts. Data based on publicly available results for this test / test settings, separated by vendor, result divided by the reference CPU clock speed, grouped by matching physical CPU core count, and normalized against the smallest core count tested from each vendor for each CPU having a sufficient number of test samples and statistically significant data.

IntelAMDOpenBenchmarking.orgRelative Core Scaling To BaseMobile Neural Network CPU Core ScalingModel: squeezenetv1.1468162432640.43960.87921.31881.75842.198

Notable Instruction Set Usage

Notable instruction set extensions supported by this test, based on an automatic analysis by the Phoronix Test Suite / OpenBenchmarking.org analytics engine.

Instruction Set
Support
Instructions Detected
SSE2 (SSE2)
Used by default on supported hardware.
 
CVTSS2SD MOVAPD MOVDQU PUNPCKLQDQ MOVDQA CVTSI2SD ADDSD MULSD CVTSD2SS MOVD ANDPD COMISD CVTTSD2SI CMPNLESD SUBSD CVTDQ2PS PSHUFD PMULUDQ PSRLDQ CVTDQ2PD CMPLTPD ADDPD CVTPS2PD UNPCKLPD MOVUPD MAXPD UNPCKHPD MAXSD DIVSD CVTTPS2DQ ORPD XORPD PSHUFLW DIVPD CVTPD2PS SQRTSD PADDQ CVTPS2DQ
SSE3 (SSE3)
Used by default on supported hardware.
 
MOVSLDUP MOVSHDUP HADDPS
SSSE3 (SSSE3)
Used by default on supported hardware.
 
PSHUFB PHADDD
Used by default on supported hardware.
Found on Intel processors since Sandy Bridge (2011).
Found on AMD processors since Bulldozer (2011).

 
VBROADCASTF128 VINSERTF128 VEXTRACTF128 VZEROALL VBROADCASTSS VZEROUPPER VPERM2F128 VPERMILPS VMASKMOVPS VBROADCASTSD
Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Excavator (2016).

 
VEXTRACTI128 VPBROADCASTB VINSERTI128 VPERM2I128 VPBROADCASTQ VPGATHERQD VPERMQ VPERMD VPBROADCASTD VPMASKMOVD VPBROADCASTW
FMA (FMA)
Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Bulldozer (2011).

 
VFNMADD132PS VFMADD132PS VFMADD231PS VFMADD231SS VFMADD213PS VFMADD132SS VFNMADD231SS VFMADD231SD VFMSUB132SS VFNMADD132SS VFMSUB231SS VFMSUB132SD VFMSUB231SD VFMADD213SS VFMADD132SD VFNMADD213PS VFNMADD213SS VFNMADD132SD VFNMADD231PS VFMSUB132PS VFMSUB231PS
The test / benchmark does honor compiler flag changes.
Last automated analysis: 10 May 2021

This test profile binary relies on the shared libraries libMNN.so, libpthread.so.0, libc.so.6, libm.so.6, libmvec.so.1.

Recent Test Results

OpenBenchmarking.org Results Compare

1 System - 1022 Benchmark Results

Intel Core i7-1165G7 - Framework FRANBMCP06 - Intel Tiger Lake-LP

Ubuntu 21.10 - 5.13.0-20-generic - GNOME Shell 40.5

1 System - 290 Benchmark Results

AMD Ryzen 9 5900X 12-Core - ASUS ROG CROSSHAIR VIII HERO - AMD Starship

Ubuntu 21.10 - 5.13.0-20-generic - GNOME Shell 40.5

8 Systems - 364 Benchmark Results

2 x AMD EPYC 7742 64-Core - Supermicro H11DSi-NT v2.00 - AMD Starship

Ubuntu 21.10 - 5.13.0-19-generic - GNOME Shell 40.5

1 System - 495 Benchmark Results

AMD EPYC 75F3 32-Core - ASRockRack ROME2D16-2T - AMD Starship

Ubuntu 21.10 - 5.13.0-19-generic - GNOME Shell 40.5

3 Systems - 326 Benchmark Results

AMD EPYC 7742 64-Core - Supermicro H11DSi-NT v2.00 - AMD Starship

Ubuntu 21.10 - 5.13.0-19-generic - GNOME Shell 40.5

4 Systems - 318 Benchmark Results

AMD EPYC 7F52 16-Core - Supermicro H11DSi-NT v2.00 - AMD Starship

Ubuntu 21.10 - 5.13.0-19-generic - GNOME Shell 40.5

3 Systems - 210 Benchmark Results

AMD EPYC 74F3 24-Core - ASRockRack ROME2D16-2T - AMD Starship

Ubuntu 21.04 - 5.11.0-37-generic - X Server

3 Systems - 185 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 - 149 Benchmark Results

AMD Ryzen 5 3600XT 6-Core - MSI X470 GAMING M7 AC - AMD Starship

Ubuntu 20.10 - 5.8.0-55-generic - GNOME Shell 3.38.1

1 System - 149 Benchmark Results

AMD Ryzen 9 3900XT 12-Core - MSI MEG X570 GODLIKE - AMD Starship

Ubuntu 19.10 - 5.3.0-18-generic - GNOME Shell 3.34.1

3 Systems - 126 Benchmark Results

AMD Ryzen 3 3300X 4-Core - MSI B350M GAMING PRO - AMD Starship

Ubuntu 20.04 - 5.9.0-rc5-14sep-patch - GNOME Shell 3.36.4

8 Systems - 111 Benchmark Results

2 x Intel Xeon Platinum 8380 - Intel M50CYP2SB2U - Intel Device 0998

Fedora Linux 35 - 5.14.7-300.fc35.x86_64 - GCC 11.2.1 20210728

7 Systems - 110 Benchmark Results

2 x Intel Xeon Platinum 8380 - Intel M50CYP2SB2U - Intel Device 0998

Fedora Linux 35 - 5.14.7-300.fc35.x86_64 - GCC 11.2.1 20210728

6 Systems - 186 Benchmark Results

2 x Intel Xeon Platinum 8380 - Intel M50CYP2SB2U - Intel Device 0998

openSUSE Tumbleweed 20210929 - 5.14.6-1-default - KDE Plasma

Most Popular Test Results

OpenBenchmarking.org Results Compare

2 Systems - 354 Benchmark Results

AMD Ryzen Threadripper 3990X 64-Core - Gigabyte TRX40 AORUS PRO WIFI - AMD Starship

Pop 21.04 - 5.11.0-7620-generic - GNOME Shell 3.38.4

4 Systems - 78 Benchmark Results

AMD Ryzen 9 5900HX - ASUS G513QY v1.0 - AMD Renoir Root Complex

Ubuntu 21.04 - 5.13.2 - GNOME Shell 3.38.4

2 Systems - 75 Benchmark Results

AMD Ryzen 9 5900HX - ASUS G513QY v1.0 - AMD Renoir Root Complex

Ubuntu 21.04 - 5.13.2 - GNOME Shell 3.38.4

3 Systems - 78 Benchmark Results

AMD Ryzen 9 5900HX - ASUS G513QY v1.0 - AMD Renoir Root Complex

Ubuntu 21.04 - 5.13.2 - GNOME Shell 3.38.4

3 Systems - 34 Benchmark Results

Intel Xeon E-2288G - Compulab SBC-ATCFL v1.2 - Intel Cannon Lake PCH

Ubuntu 20.10 - 5.8.0-41-generic - GNOME Shell 3.38.2

3 Systems - 41 Benchmark Results

AMD Ryzen 5 4500U - LENOVO LNVNB161216 - AMD Renoir Root Complex

Ubuntu 21.04 - 5.11.0-17-generic - GNOME Shell 3.38.4

3 Systems - 89 Benchmark Results

Intel Core i9-10900K - Gigabyte Z490 AORUS MASTER - Intel Comet Lake PCH

Ubuntu 20.04 - 5.9.0-050900daily20201012-generic - GNOME Shell 3.36.4

2 Systems - 8 Benchmark Results

Intel Core i9-8950HK - MSI MS-17A3 - Intel Cannon Lake PCH

Ubuntu 20.04 - 5.8.0-63-generic - GNOME Shell 3.36.9

3 Systems - 100 Benchmark Results

Intel Core i7-3770K - ECS Z77H2-A2X v1.0 - Intel Xeon E3-1200 v2

Ubuntu 20.04 - 5.8.0-55-generic - GNOME Shell 3.36.9

3 Systems - 26 Benchmark Results

Intel Core i7-1165G7 - Dell 0GG9PT - Intel Tiger Lake-LP

Ubuntu 21.04 - 5.11.0-18-generic - GNOME Shell 3.38.4

3 Systems - 45 Benchmark Results

AMD Ryzen Threadripper 3970X 32-Core - ASUS ROG ZENITH II EXTREME - AMD Starship

Ubuntu 20.10 - 5.11.0-rc6-phx - GNOME Shell 3.38.1

3 Systems - 40 Benchmark Results

Intel Core i7-10700T - Logic Supply RXM-181 - Intel Comet Lake PCH

Ubuntu 21.04 - 5.11.0-17-generic - GNOME Shell 3.38.4

6 Systems - 102 Benchmark Results

AMD EPYC 7543 32-Core - TYAN S8036GM2NE-LE - AMD Starship

Debian 11 - 5.10.0-7-amd64 - GNOME Shell 3.38.4

11 Systems - 56 Benchmark Results

AMD EPYC 7543 32-Core - TYAN S8036GM2NE-LE - AMD Starship

Arch Linux - 5.12.12-arch1-1 - GCC 11.1.0

Find More Test Results