Regarding performance regression on specific query

From: "Jung, Jinho" <jinho(dot)jung(at)gatech(dot)edu>
To: "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>
Subject: Regarding performance regression on specific query
Date: 2018-11-19 17:49:43
Message-ID: SN1PR07MB39341F7294904D78F291353FEED80@SN1PR07MB3934.namprd07.prod.outlook.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


Hello,

I am Jinho Jung, PhD student at Georgia Tech and I am conducting DB performance research. I am sending our evaluation result to get the feedback regarding our findings.

We found several performance regression queries. Attached files are three of them that we confirmed performance regression (in terms of query execution time) between v10.6 and v9.4.20. Hope you can test these queries and give us feedback. For your information, we are also sending the execution time measured on our machine.

Execution time
=============
1.sql
10.6 : 469 ms
9.4.20: 10 ms

4.sql
10.6 : 34019 ms
9.4.20: 0.4 ms

20.sql
10.6 : 2791 ms
9.4.20: 61 ms

Evaluation setup
==============
1) compile 9.4.20 and 10.6 from released source code (https://www.postgresql.org/ftp/source/)
2) without importing additional data, just run the attached queries

We appreciate you taking time for test!

Best regards,
Jinho Jung

Attachment Content-Type Size
1.sql application/sql 2.8 KB
4.sql application/sql 3.3 KB
20.sql application/sql 4.0 KB

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Banck 2018-11-19 18:11:19 Re: pgsql: Add TAP tests for pg_verify_checksums
Previous Message Tom Lane 2018-11-19 17:48:34 Re: Time to backpatch config/ax_pthread.m4 changes?