PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks.

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

Project Site

github.com

Test Created

10 January 2019

Last Updated

26 September 2019

Test Maintainer

Michael Larabel 

Test Type

Graphics

Average Install Time

15 Seconds

Average Run Time

8 Minutes, 22 Seconds

Test Dependencies

Python + OpenCL

Accolades

40k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventsPlaidML Popularity Statisticspts/plaidml2019.012019.032019.052019.072019.092019.112020.012020.032020.052020.072020.092020.112021.012021.032021.052021.072021.092021.112022.012022.032022.052022.072022.092022.112023.012023.032023.052023.072023.092023.112024.012024.032024.052024.072024.092K4K6K8K10K
* 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 updated weekly as of 4 October 2024.
No88.1%Yes11.9%FP16 Option PopularityOpenBenchmarking.org
Training6.1%Inference93.9%Mode Option PopularityOpenBenchmarking.org
Mobilenet20.9%VGG1917.3%Inception V37.4%DenseNet 2018.5%ResNet 5015.9%IMDB LSTM10.0%VGG1620.0%Network Option PopularityOpenBenchmarking.org
OpenCL52.9%CPU47.1%Device Option PopularityOpenBenchmarking.org

Revision History

pts/plaidml-1.0.4   [View Source]   Thu, 26 Sep 2019 14:22:09 GMT
Fixes for latest upstream PlaidML working around configuration files and library issues.

pts/plaidml-1.0.3   [View Source]   Sun, 27 Jan 2019 16:21:22 GMT
Set RequiresDisplay = FALSE

pts/plaidml-1.0.2   [View Source]   Fri, 11 Jan 2019 12:06:07 GMT
Always set --user for pip3 to avoid issues on some distros.

pts/plaidml-1.0.1   [View Source]   Thu, 10 Jan 2019 14:30:27 GMT
Add --train option which works in some configurations.

pts/plaidml-1.0.0   [View Source]   Thu, 10 Jan 2019 10:51:47 GMT
Initial commit of PlaidML deep learning framework benchmark, plaidbench.

Suites Using This Test

Machine Learning

HPC - High Performance Computing

CPU Massive

NVIDIA GPU Compute


Performance Metrics

Analyze Test Configuration:

PlaidML

FP16: No - Mode: Inference - Network: VGG16 - Device: CPU

OpenBenchmarking.org metrics for this test profile configuration based on 653 public results since 15 October 2019 with the latest data as of 20 August 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
FPS (Average)
93rd
4
36.8 +/- 1.0
93rd
9
36.4 +/- 0.8
91st
3
35.2 +/- 0.4
90th
9
34.7 +/- 0.9
87th
6
33.7 +/- 0.3
87th
4
33.4 +/- 1.2
87th
4
33.1 +/- 1.4
86th
8
32.9 +/- 0.2
84th
9
32.2 +/- 0.1
82nd
8
32.0 +/- 0.1
81st
3
31.7 +/- 0.4
81st
11
31.4 +/- 3.5
79th
8
30.2 +/- 0.1
76th
12
29.6 +/- 0.2
Mid-Tier
75th
< 29.5
73rd
6
28.7 +/- 0.9
73rd
9
28.3 +/- 2.0
70th
9
27.8 +/- 0.2
70th
16
27.6 +/- 1.4
69th
4
27.4 +/- 1.1
64th
5
25.3 +/- 0.3
60th
11
24.8 +/- 0.3
56th
8
23.8 +/- 0.9
55th
3
23.3 +/- 0.3
Median
50th
21.6
50th
10
21.6 +/- 0.4
49th
10
21.4 +/- 0.4
48th
31
21.3 +/- 0.4
47th
4
21.0 +/- 1.9
44th
8
20.2 +/- 0.1
42nd
5
19.7 +/- 1.2
37th
12
18.6 +/- 0.5
37th
13
18.5 +/- 0.5
37th
18
18.4 +/- 0.7
30th
6
16.4 +/- 0.5
30th
7
15.9 +/- 1.4
28th
5
15.8 +/- 0.1
27th
13
15.5 +/- 0.6
Low-Tier
25th
< 15.2
24th
8
14.9 +/- 0.2
23rd
3
14.7 +/- 0.4
19th
10
13.1 +/- 0.2
18th
6
12.9 +/- 0.6
14th
17
11.8 +/- 0.4
12th
9
11.2 +/- 0.3
10th
4
10.2 +/- 0.1
6th
4
7.4 +/- 1.0
3rd
3
2.1 +/- 0.1
OpenBenchmarking.orgDistribution Of Public Results - FP16: No - Mode: Inference - Network: VGG16 - Device: CPU653 Results Range From 0 To 63 FPS61218243036424854606672306090120150

Based on OpenBenchmarking.org data, the selected test / test configuration (PlaidML - FP16: No - Mode: Inference - Network: VGG16 - Device: CPU) 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 BenchmarkFP16: No - Mode: Inference - Network: VGG16 - Device: CPURun-Time1122334455Min: 2 / Avg: 8.24 / Max: 56

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 RunsFP16: No - Mode: Inference - Network: VGG16 - Device: CPUDeviation246810Min: 0 / Avg: 0.81 / Max: 3

Does It Scale Well With Increasing Cores?

Yes, based on the automated analysis of the collected public benchmark data, this test / test settings does 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 BasePlaidML CPU Core ScalingFP16: No - Mode: Inference - Network: VGG16 - Device: CPU6812162432641.48542.97084.45625.94167.427

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)

Recent Test Results

OpenBenchmarking.org Results Compare

2 Systems - 19 Benchmark Results

2 x AMD EPYC 9334 32-Core - Giga Computing MZ73-LM1-000 v01000100 - AMD Device 14a4

Ubuntu 22.04 - 6.8.0-45-generic - GNOME Shell 42.9

1 System - 334 Benchmark Results

Intel Core i3-12100 - ASRock H610M-HDV/M.2 R2.0 - Intel Device 7aa7

Ubuntu 20.04 - 5.15.0-89-generic - GNOME Shell 3.36.9

1 System - 116 Benchmark Results

AMD Ryzen 7 3700X 8-Core - ASUS TUF GAMING X570-PLUS - AMD Starship

RockyLinux 9.4 - 6.1.102-1.el9.elrepo.x86_64 - KDE Plasma 5.27.11

2 Systems - 91 Benchmark Results

Intel Xeon Gold 6226R - QEMU Standard PC - Intel 440FX 82441FX PMC

Ubuntu 24.04 - 6.8.0-39-generic - X Server

1 System - 83 Benchmark Results

2 x Intel Xeon Gold 6226R - (5.14 BIOS) - Intel Sky Lake-E DMI3 Registers

Ubuntu 24.04 - 6.8.0-38-generic - X Server

2 Systems - 130 Benchmark Results

1 System - 134 Benchmark Results

AMD Ryzen 7 7700X 8-Core - MSI PRO B650-P WIFI - AMD Device 14d8

Ubuntu 23.04 - 6.2.0-39-generic - GNOME Shell 44.3

3 Systems - 105 Benchmark Results

2 x Intel Xeon E5-2682 v4 - Supermicro SYS-7048GR-TR X10DRG-Q v1.10 - Intel Xeon E7 v4

Ubuntu 22.04 - 5.15.0-102-generic - X Server 1.21.1.3

Find More Test Results