simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others.

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

Project Site

simdjson.org

Source Repository

github.com

Test Created

13 December 2020

Last Updated

11 August 2024

Test Maintainer

Michael Larabel 

Test Type

Processor

Average Install Time

1 Minute, 53 Seconds

Average Run Time

3 Minutes, 29 Seconds

Test Dependencies

C/C++ Compiler Toolchain + CMake

Accolades

30k+ Downloads

Supported Platforms


Public Result Uploads *Reported Installs **Reported Test Completions **Test Profile Page Views ***OpenBenchmarking.orgEventssimdjson Popularity Statisticspts/simdjson2020.122021.022021.042021.062021.082021.102021.122022.022022.042022.062022.082022.102022.122023.022023.042023.062023.082023.102023.122024.022024.042024.062024.082024.102024.125K10K15K20K25K
* 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 22 December 2024.
Kostya19.9%LargeRandom20.1%DistinctUserID19.7%PartialTweets20.1%TopTweet20.1%Throughput Test Option PopularityOpenBenchmarking.org

Revision History

pts/simdjson-2.1.0   [View Source]   Sun, 11 Aug 2024 10:51:43 GMT
Update against simdjson 3.10 upstream.

pts/simdjson-2.0.1   [View Source]   Wed, 15 Jun 2022 18:18:53 GMT
Update against simdjson 2.0.4 upstream maintenance release to address GCC 12 build issues.

pts/simdjson-2.0.0   [View Source]   Wed, 25 May 2022 13:46:39 GMT
Update against simdjson 2.0 upstream.

pts/simdjson-1.3.0   [View Source]   Tue, 07 Sep 2021 19:19:27 GMT
Update against simdjson 1.0.0 upstream.

pts/simdjson-1.2.0   [View Source]   Fri, 12 Mar 2021 06:22:00 GMT
Update against simdjson 0.8.2 upstream.

pts/simdjson-1.1.1   [View Source]   Mon, 14 Dec 2020 13:25:30 GMT
Don't bother with the documents/s report for now.

pts/simdjson-1.1.0   [View Source]   Mon, 14 Dec 2020 11:39:20 GMT
Update against simdjson 0.7.1 that was just released with POWER/ARM fixes. Also report documents/s as well.

pts/simdjson-1.0.0   [View Source]   Sun, 13 Dec 2020 16:06:04 GMT
Initial commit of simdjson benchmark.

Suites Using This Test

Server

Programmer / Developer System Benchmarks


Performance Metrics

Analyze Test Configuration:

simdjson 3.10

Throughput Test: PartialTweets

OpenBenchmarking.org metrics for this test profile configuration based on 287 public results since 11 August 2024 with the latest data as of 23 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
GB/s (Average)
99th
22
13.11 +/- 0.60
95th
4
12.59 +/- 0.22
94th
4
12.41 +/- 0.02
85th
3
10.02 +/- 0.03
84th
7
9.94 +/- 0.36
81st
9
9.77 +/- 0.06
79th
5
9.69 +/- 0.11
Mid-Tier
75th
< 9.61
74th
8
9.48 +/- 0.49
70th
11
9.35 +/- 0.72
70th
7
9.34 +/- 0.93
70th
5
9.30 +/- 0.03
70th
4
9.30 +/- 0.73
62nd
8
8.61 +/- 0.07
60th
3
8.45 +/- 1.09
60th
4
8.37 +/- 0.01
57th
5
8.20 +/- 0.32
55th
4
8.14 +/- 0.09
52nd
6
7.55 +/- 0.12
51st
4
7.46 +/- 0.01
Median
50th
7.45
47th
10
7.14 +/- 0.64
47th
4
7.08 +/- 0.01
46th
3
7.03 +/- 0.03
43rd
7
6.78 +/- 0.29
39th
3
6.61 +/- 0.03
37th
10
6.57 +/- 0.05
35th
7
6.14 +/- 0.42
34th
3
6.08 +/- 0.07
29th
4
5.17 +/- 0.02
28th
3
4.98 +/- 0.08
27th
3
4.46 +/- 0.16
Low-Tier
25th
< 4.38
25th
4
4.36 +/- 0.04
21st
3
4.17 +/- 0.01
18th
3
4.04 +/- 0.01
18th
4
3.92 +/- 0.01
16th
4
3.67 +/- 0.01
10th
4
2.38 +/- 0.02
8th
6
2.05 +/- 0.07
Detailed Performance Overview
OpenBenchmarking.orgDistribution Of Public Results - Throughput Test: PartialTweets270 Results Range From 0 To 15 GB/s2468101214161820222420406080100

Based on OpenBenchmarking.org data, the selected test / test configuration (simdjson 3.10 - Throughput Test: PartialTweets) has an average run-time of 7 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 BenchmarkThroughput Test: PartialTweetsRun-Time510152025Min: 3 / Avg: 6.83 / Max: 19

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 0.2%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsThroughput Test: PartialTweetsDeviation246810Min: 0 / Avg: 0.18 / Max: 3

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.
 
MOVD PSHUFD MOVDQA MOVDQU SHUFPD COMISD CVTSI2SD MULSD XORPD DIVSD PUNPCKLQDQ ADDSD SUBSD CVTTSD2SI MOVAPD UCOMISD PUNPCKHQDQ PADDQ ANDPD MOVMSKPD MAXSD MOVUPD PSUBQ MOVHPD SQRTSD UNPCKLPD SUBPD MULPD UNPCKHPD DIVPD CMPNLTSD ANDNPD ORPD
SSSE3 (SSSE3)
Used by default on supported hardware.
 
PSHUFB PALIGNR PMADDUBSW
SSE 4.2 (SSE4_2)
Used by default on supported hardware.
Found on Intel processors since at least 2010.
Found on AMD processors since Bulldozer (2011).

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

 
VZEROUPPER VBROADCASTSS VINSERTF128 VEXTRACTF128 VPERMILPD VBROADCASTSD
Used by default on supported hardware.
Found on Intel processors since Haswell (2013).
Found on AMD processors since Excavator (2016).

 
VPBROADCASTB VINSERTI128 VEXTRACTI128 VPERM2I128 VPBROADCASTD VPERMQ VPERMD VPBROADCASTQ
Advanced Vector Extensions 512 (AVX512)
Used by default on supported hardware.
 
(ZMM REGISTER USE)
FMA (FMA)
Requires passing a supported compiler/build flag (verified with targets: skylake, tigerlake, cascadelake, sapphirerapids, alderlake, znver2, znver3).
Found on Intel processors since Haswell (2013).
Found on AMD processors since Bulldozer (2011).

 
VFMADD132SD VFMADD213SD VFMADD231SD VFNMADD213SD VFMSUB132SD
The test / benchmark does honor compiler flag changes.
Last automated analysis: 23 August 2024

This test profile binary relies on the shared libraries libm.so.6, 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
arm64
Apple M4 Max
ARMv8 64-bit
aarch64
ARMv8 Cortex-A72 4-Core, ARMv8 Cortex-A76 4-Core, ARMv8 Neoverse-N1 128-Core, ARMv8 Neoverse-V2, ARMv8 Neoverse-V2 72-Core, AmpereOne 192-Core, Apple M1