PostgreSQL pgbench

This is a simple benchmark of PostgreSQL using pgbench.

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

Project Site

postgresql.org

Test Created

6 December 2010

Last Updated

24 September 2020

Test Maintainer

Michael Larabel 

Test Type

System

Average Install Time

1 Minute, 22 Seconds

Average Run Time

5 Minutes, 37 Seconds

Test Dependencies

C/C++ Compiler Toolchain + BC

Accolades

250k+ Downloads + 10k+ Public Benchmark Results

Supported Platforms


Public Result Uploads *Reported Test Completions **Reported Installs **Test Profile Page Views ***OpenBenchmarking.orgEventsPostgreSQL pgbench Popularity Statisticspts/pgbench2010.122011.042011.082011.122012.042012.082012.122013.042013.082013.122014.042014.082014.122015.042015.082015.122016.042016.082016.122017.042017.082017.122018.042018.082018.122019.042019.082019.122020.042020.082020.122021.0440K80K120K160K200K
* 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 15 June 2021.
10091.7%18.3%Scaling Factor Option PopularityOpenBenchmarking.org
15.5%2505.1%1005.6%5083.8%Clients Option PopularityOpenBenchmarking.org
Read Write89.1%Read Only10.9%Mode Option PopularityOpenBenchmarking.org

Revision History

pts/pgbench-1.10.1   [View Source]   Thu, 24 Sep 2020 14:41:34 GMT
Settings tweak for the 1.10.0 changes.

pts/pgbench-1.10.0   [View Source]   Thu, 24 Sep 2020 12:11:09 GMT
Update against PostgreSQL 13.0 upstream, restructure test option handling.

pts/pgbench-1.9.1   [View Source]   Fri, 04 Oct 2019 06:53:32 GMT
Fix heavy contentiob by going off physical CPU cores.

pts/pgbench-1.9.0   [View Source]   Thu, 03 Oct 2019 09:57:40 GMT
Update against PostgreSQL 12.0. NOTE: still waiting on official Windows postgresql 12 release binaries.

pts/pgbench-1.8.7   [View Source]   Fri, 29 Mar 2019 03:20:12 GMT
Update download URLs.

pts/pgbench-1.8.6   [View Source]   Fri, 07 Dec 2018 12:47:49 GMT
Enable BSD build support.

pts/pgbench-1.8.5   [View Source]   Fri, 07 Sep 2018 09:43:34 GMT
Add bc as an external dependency.

pts/pgbench-1.8.4   [View Source]   Sun, 25 Mar 2018 13:10:40 GMT
Fix the bundled patch allowing PostgreSQL to run as root.

pts/pgbench-1.8.3   [View Source]   Thu, 22 Mar 2018 18:07:18 GMT
Adjust run count as was bumped lower during debugging.

pts/pgbench-1.8.2   [View Source]   Thu, 22 Mar 2018 18:06:30 GMT
Create db directory on Windows otherwise the test seems to sometimes fail, also add note about MSVCR120.dll needing to be on system for PostgreSQL to work.

pts/pgbench-1.8.1   [View Source]   Thu, 01 Mar 2018 11:46:17 GMT
Tidy up the Windows support, appears to be good to go.

pts/pgbench-1.8.0   [View Source]   Thu, 01 Mar 2018 11:26:11 GMT
Update against upstream PostgreSQL 10.3, add initial Windows bits.

pts/pgbench-1.7.0   [View Source]   Sat, 14 Oct 2017 15:23:03 GMT
Updating against postgresql 10.0

pts/pgbench-1.6.0   [View Source]   Fri, 19 May 2017 11:57:32 GMT
Update against upstream PostgreSQL 9.6.3.

pts/pgbench-1.5.2   [View Source]   Thu, 08 Oct 2015 16:05:08 GMT
Update download links, advertise BSD compatibility.

pts/pgbench-1.5.1   [View Source]   Tue, 09 Jun 2015 21:16:09 GMT
Hack up pgbench to let it run as root

pts/pgbench-1.5.0   [View Source]   Tue, 09 Jun 2015 20:05:58 GMT
Major update/overhaul of pgbench PostgreSQL benchmark.

pts/pgbench-1.4.0   [View Source]   Wed, 11 Apr 2012 10:15:04 GMT
Move back to older PostgreSQL 8.4.11 upstream server version to work around PostgreSQL 9.x issue right now dealing with the plpg

pts/pgbench-1.3.0   [View Source]   Sat, 27 Aug 2011 16:23:30 GMT
Update PostgreSQL to version 9.0.4.

pts/pgbench-1.2.0   [View Source]   Mon, 06 Dec 2010 15:16:36 GMT
Initial import into OpenBenchmarking.org

Suites Using This Test

C/C++ Compiler Tests

Database Test Suite

Multi-Core

Common Kernel Benchmarks

Server

CPU Massive


Performance Metrics

Analyze Test Configuration:

PostgreSQL pgbench 13.0

Scaling Factor: 100 - Clients: 50 - Mode: Read Write

OpenBenchmarking.org metrics for this test profile configuration based on 32,514 public results since 24 September 2020 with the latest data as of 12 June 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
TPS (Average)
100th
4
48574 +/- 532
100th
7
39382 +/- 4666
100th
52
28204 +/- 2103
100th
4
24252 +/- 291
100th
5
24230 +/- 3181
100th
12
24038 +/- 1673
100th
3
22835 +/- 187
100th
3
19892 +/- 556
100th
11
19780 +/- 2169
100th
3
16698 +/- 2248
100th
3
13434 +/- 670
100th
3
13196 +/- 51
100th
12
11460 +/- 41
100th
5
8861 +/- 327
100th
20
7894 +/- 954
100th
3
7721 +/- 680
100th
3
6291 +/- 347
100th
3
4088 +/- 67
99th
4
3524 +/- 190
99th
4
3502 +/- 148
99th
3
3487 +/- 138
96th
3
1552 +/- 25
Mid-Tier
75th
< 1405
2nd
5
658 +/- 8
OpenBenchmarking.orgDistribution Of Public Results - Scaling Factor: 100 - Clients: 50 - Mode: Read Write32514 Results Range From 44 To 58428 TPS44121223803548471658847052822093881055611724128921406015228163961756418732199002106822236234042457225740269082807629244304123158032748339163508436252374203858839756409244209243260444284559646764479324910050268514365260453772549405610857276584446K12K18K24K30K

Based on OpenBenchmarking.org data, the selected test / test configuration (PostgreSQL pgbench 13.0 - Scaling Factor: 100 - Clients: 50 - Mode: Read Write) has an average run-time of 16 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 BenchmarkScaling Factor: 100 - Clients: 50 - Mode: Read WriteRun-Time612182430Min: 1 / Avg: 15.1 / Max: 29

Based on public OpenBenchmarking.org results, the selected test / test configuration has an average standard deviation of 6.4%.

OpenBenchmarking.orgPercent, Fewer Is BetterAverage Deviation Between RunsScaling Factor: 100 - Clients: 50 - Mode: Read WriteDeviation714212835Min: 0 / Avg: 6.38 / Max: 35

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 BasePostgreSQL pgbench CPU Core ScalingScaling Factor: 100 - Clients: 50 - Mode: Read Write48162432481.833.665.497.329.15

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.
 
COMISD DIVSD MULSD CVTTSD2SI CVTSI2SD ADDSD SUBSD UCOMISD MOVAPD SQRTSD ANDPD ANDNPD ORPD MOVDQU XORPD MAXSD MINSD MOVDQA
Requires passing a supported compiler/build flag (verified with targets: sandybridge, skylake, znver2, znver3).
Found on Intel processors since Sandy Bridge (2011).
Found on AMD processors since Bulldozer (2011).

 
VZEROUPPER VINSERTF128 VEXTRACTF128 VBROADCASTSS
Requires passing a supported compiler/build flag (verified with targets: skylake, znver2, znver3).
Found on Intel processors since Haswell (2013).
Found on AMD processors since Excavator (2016).

 
VPBROADCASTQ VEXTRACTI128 VINSERTI128 VPBROADCASTD
FMA (FMA)
Requires passing a supported compiler/build flag (verified with targets: skylake, znver2, znver3).
Found on Intel processors since Haswell (2013).
Found on AMD processors since Bulldozer (2011).

 
VFNMADD231SD VFMADD213SD VFMSUB132SD VFNMADD213SD VFNMADD132SD VFMADD132SD
The test / benchmark does honor compiler flag changes.
Last automated analysis: 10 May 2021

This test profile binary relies on the shared libraries libpq.so.5, libpthread.so.0, libm.so.6, libc.so.6.

Recent Test Results

OpenBenchmarking.org Results Compare

8 Systems - 41 Benchmark Results

AMD Ryzen 7 5800X 8-Core - ASRock X570 Pro4 - AMD Starship

Ubuntu 20.10 - 5.8.0-29-generic - GNOME Shell 3.38.1

7 Systems - 37 Benchmark Results

AMD Ryzen 9 5900X 12-Core - Gigabyte X570 AORUS MASTER - 4 x 16384 MB 3600MHz F4-3600C14-16GVK

Microsoft Windows 10 Pro Build 19042 - 10.0 - 461.09

12 Systems - 44 Benchmark Results

AMD Ryzen 9 3900X 12-Core - Gigabyte X570 AORUS MASTER - AMD Starship

Fedora 34 - 5.12.5-300.fc34.x86_64 - GNOME Shell 40.1

17 Systems - 43 Benchmark Results

ARMVv8 Cortex-A53 - Libre Board ALL-H3-CC H5 - 2048MB

Raspbian 9.4 - 4.18.0-rc3-gfc36def997cf-dirty - X Server 1.19.2

17 Systems - 43 Benchmark Results

ARMv7 rev 5 - Allwinner sun8i Family Libre Board ALL-H3-CC H3 - 493MB

Ubuntu 18.04 - 4.14.65-sunxi - GCC 7.3.0

8 Systems - 195 Benchmark Results

AMD EPYC - Hetzner vServer v20171111 - 1 x 8000 MB RAM QEMU

Debian 11 - 5.12.8-051208-generic - GCC 10.2.1 20210110

7 Systems - 195 Benchmark Results

AMD EPYC - Hetzner vServer v20171111 - 1 x 8000 MB RAM QEMU

Debian 11 - 5.12.8-051208-generic - GCC 10.2.1 20210110

6 Systems - 195 Benchmark Results

AMD EPYC - Hetzner vServer v20171111 - 1 x 8000 MB RAM QEMU

Debian 11 - 5.12.8-051208-generic - GCC 10.2.1 20210110

10 Systems - 12 Benchmark Results

Intel Core i3-3110M - LENOVO - Intel 3rd Gen Core DRAM

OpenMandrivaLinux 4.2 - 5.11.12-desktop-1omv4002 - KDE Plasma 5.20.5

1 System - 60 Benchmark Results

Intel Core i7-4790 - ASUS BM6AE_BM1AE_BP1AE - Intel 4th Gen Core DRAM

Ubuntu 18.04 - 4.15.0-142-generic - X Server

5 Systems - 195 Benchmark Results

AMD EPYC - Hetzner vServer v20171111 - 1 x 8000 MB RAM QEMU

Debian 11 - 5.12.8-051208-generic - GCC 10.2.1 20210110

4 Systems - 195 Benchmark Results

AMD EPYC - Hetzner vServer v20171111 - 1 x 8000 MB RAM QEMU

Debian 10 - 4.19.0-16-amd64 - GCC 8.3.0

Most Popular Test Results

OpenBenchmarking.org Results Compare

3 Systems - 4 Benchmark Results

ARMv8 Cortex-A72 - BCM2835 Raspberry Pi 4 Model B Rev 1.4 - 8GB

Ubuntu 20.10 - 5.8.0-1015-raspi - GCC 10.2.0

3 Systems - 34 Benchmark Results

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

Ubuntu 20.10 - 5.8.0-44-generic - GNOME Shell 3.38.2

3 Systems - 131 Benchmark Results

Intel Core i5-10600K - ASUS PRIME Z490M-PLUS - Intel Comet Lake PCH

Ubuntu 20.04 - 5.4.0-45-generic - GNOME Shell 3.36.4

3 Systems - 114 Benchmark Results

AMD Ryzen 7 1800X Eight-Core - MSI X370 XPOWER GAMING TITANIUM - AMD 17h

Ubuntu 20.10 - 5.8.0-21-generic - GNOME Shell 3.38.0

Find More Test Results