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 toolkit.

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

Project Site

intel.com

Source Repository

github.com

Test Created

17 June 2020

Last Updated

15 October 2024

Test Maintainer

Michael Larabel

Test Type

Processor

Average Install Time

5 Minutes, 43 Seconds

Average Run Time

2 Minutes, 16 Seconds

Test Dependencies

C/C++ Compiler Toolchain + CMake

Accolades

90k+ Downloads

Supported Platforms

* 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 18 December 2024.

Revision History

pts/onednn-3.6.0 [View Source] Tue, 15 Oct 2024 14:53:25 GMT
Update against oneDNN 3.6 upstream.

pts/onednn-3.4.0 [View Source] Fri, 01 Mar 2024 13:02:43 GMT
Update against oneDNN 3.4 upstream.

pts/onednn-3.3.0 [View Source] Thu, 12 Oct 2023 11:14:07 GMT
Update against oneDNN 3.3 upstream.

pts/onednn-3.1.0 [View Source] Fri, 31 Mar 2023 18:14:37 GMT
Update against oneDNN 3.1 upstream.

pts/onednn-3.0.0 [View Source] Mon, 19 Dec 2022 21:07:39 GMT
Update against oneDNN 3.0 upstream.

pts/onednn-2.7.0 [View Source] Wed, 28 Sep 2022 13:00:44 GMT
Update against oneDNN 2.7 upstream.

pts/onednn-1.8.0 [View Source] Tue, 29 Mar 2022 19:55:25 GMT
Update against oneDNN 2.6 upstream.

pts/onednn-1.7.0 [View Source] Sat, 13 Mar 2021 07:49:33 GMT
Update against oneDNN 2.1.2 upstream.

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

HPC - High Performance Computing

Intel oneAPI

CPU Massive

Creator Workloads

Server CPU Tests

Performance Metrics

Analyze Test Configuration:

oneDNN 3.6

Harness: Convolution Batch Shapes Auto - Engine: CPU

OpenBenchmarking.org metrics for this test profile configuration based on 132 public results since 15 October 2024 with the latest data as of 20 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

ms (Average)

2 x AMD EPYC 9575F 64-Core

97th

0.3

Intel Xeon 6980P

93rd

0.3

2 x AMD EPYC 9755 128-Core

93rd

0.3

AMD EPYC 9655P 96-Core

87th

0.3

AMD Ryzen Threadripper 7980X 64-Cores

82nd

0.6

AMD Ryzen Threadripper 3990X 64-Core

79th

1.0 ^{+/- 0.1}

Mid-Tier

75th

> 4.1

AMD EPYC 4484PX 12-Core

75th

4.1

ARMv8 Neoverse-N1 128-Core

70th

4.6 ^{+/- 0.3}

AMD Ryzen 9 9950X 16-Core

63rd

5.3 ^{+/- 0.1}

Intel Core Ultra 9 285K

63rd

5.3 ^{+/- 0.4}

Intel Core Ultra 5 245K

60th

5.7

AMD Ryzen Threadripper 3970X 32-Core

55th

5.8 ^{+/- 0.3}

Intel Core i9-14900K

51st

6.2

Median

50th

6.2

AMD EPYC 4564P 16-Core

47th

6.7 ^{+/- 0.1}

Intel Core i7-1280P

43rd

8.3

AMD Ryzen AI 9 365

38th

8.4

AMD Ryzen AI 9 HX 370

37th

8.5

Intel Core i7-1185G7

37th

8.5

Intel Core Ultra 7 155H

32nd

8.7 ^{+/- 0.2}

Intel Core i9-10980XE

29th

9.9

Intel Core Ultra 7 256V

26th

11.7 ^{+/- 0.9}

Low-Tier

25th

> 13.1

AMD Ryzen 7 7840HS

23rd

13.3

AMD Ryzen 7 PRO 6850U

19th

14.0

Intel Core i7-1065G7

18th

14.1 ^{+/- 0.2}

Intel Core i5-12600K

15th

16.5

AMD Ryzen 7 7840U

13th

16.7 ^{+/- 0.1}

AMD Ryzen 9 3900XT 12-Core

10th

18.9 ^{+/- 0.1}

AMD Ryzen 7 PRO 5850U

8th

25.2

AMD Ryzen 9 5900HX

5th

33.0 ^{+/- 0.2}

ARMv8 Neoverse-V2 72-Core

2nd

33.5 ^{+/- 0.2}

Detailed Performance Overview

Based on OpenBenchmarking.org data, the selected test / test configuration (oneDNN 3.6 - Harness: Convolution Batch Shapes Auto - 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.

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

Advanced Vector Extensions (AVX)

Used by default on supported hardware.
Found on Intel processors since Sandy Bridge (2011).
Found on AMD processors since Bulldozer (2011).

VZEROUPPER VBROADCASTSS VINSERTF128 VPERMILPS VBROADCASTSD VEXTRACTF128 VPERMILPD VPERM2F128 VMASKMOVPS

Advanced Vector Extensions 2 (AVX2)

Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Excavator (2016).

VPBROADCASTQ VINSERTI128 VPBROADCASTD VPBLENDD VPSLLVD VEXTRACTI128 VPSRAVD VPERM2I128 VPGATHERQQ VGATHERQPS VPERMQ VPBROADCASTW VPSRLVQ VPBROADCASTB VPGATHERDQ VPGATHERQD VPSLLVQ VPMASKMOVQ VPERMD

FMA (FMA)

Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Bulldozer (2011).

VFMADD231SS VFMADD213SS VFMADD132SS VFMADD132SD VFMADD132PS VFMADD231PS VFMADD213PS VFNMADD132PS VFNMSUB231PS VFNMSUB132SS VFNMADD132SS VFNMSUB231SS VFNMADD231PS VFNMADD231SS VFNMADD213SS VFMADD231SD VFMSUB132SS VFMADD132PD VFMADD231PD VFMADD213PD VFMSUB231SS VFMSUB231SD

Advanced Vector Extensions 512 (AVX512)

Requires passing a supported compiler/build flag (verified with targets: cascadelake, sapphirerapids).

(ZMM REGISTER USE)

The test / benchmark does honor compiler flag changes.

Last automated analysis: 16 October 2024

This test profile binary relies on the shared libraries libdnnl.so.3, libm.so.6, libgomp.so.1, libc.so.6.

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 72-Core