Skip site navigation (1) Skip section navigation (2)

Re: Performance problem with correlated sub-query

From: "Howard, Steven (US - Tulsa)" <sthoward(at)DELOITTE(dot)com>
To: "Paul Thomas" <paul(at)tmsl(dot)demon(dot)co(dot)uk>
Cc: "pgsql-general (at) postgresql (dot) org" <pgsql-general(at)PostgreSQL(dot)org>
Subject: Re: Performance problem with correlated sub-query
Date: 2004-04-29 14:27:41
Message-ID: (view raw, whole thread or download thread mbox)
Lists: pgsql-general
I apologize for my mistake in the version. I have 7.3, and not 7.0.

However, I'll upgrade to 7.4. and work with some of the other
suggestions made also.

-----Original Message-----
From: Paul Thomas [mailto:paul(at)tmsl(dot)demon(dot)co(dot)uk] 
Sent: Thursday, April 29, 2004 9:20 AM
To: Howard, Steven (US - Tulsa)
Cc: pgsql-general @ postgresql . org
Subject: Re: [GENERAL] Performance problem with correlated sub-query

On 29/04/2004 14:34 "Howard, Steven (US - Tulsa)" wrote:
> I have created a web app that stores and displays all the messages
> my database maintenance jobs that run each night. The web app uses
> servlets and has PostgreSQL 7.0 as the back end.

7.0? That's positively ancient!
> When the user requests the first page, he gets a list of all the
> with maintenance records in the database, and a drop down list of all
> the dates of maintenance records. If the user chooses a date first,
> the app uses a prepared statement with the date contained in a
> parameter, and this executes very quickly - no problems.
> However, if the web page user does not choose a date, then the app
> a correlated sub-query to grab only the current (latest) day's
> maintenance records. The query that is executed is:
> select servername, databasename, message from messages o where
> o.date_of_msg =
>             (select max(date_of_msg) from messages i where
> = o.servername);
> And this is a dog. It takes 15 - 20 minutes to execute the query
> are about 200,000 rows in the table). I have an index on (servername,
> date_of_msg), but it doesn't seem to be used in this query.

PG doesn't use indexes for things like count(), max, min()...

You can avoid using max() by something like

select my_date from my_table order by my_date desc limit 1;

which will use the index.

> Is there a way to improve the performance on this query?

In addition to the above, I'd strongly recommend upgrading to 7.4 to
advantage of the last ~4 years of continuous improvements.

Paul Thomas
| Thomas Micro Systems Limited | Software Solutions for 
Business             |
| Computer Consultants         |   |

This message (including any attachments) contains confidential information intended for a specific individual and purpose, and is protected by law.  If you are not the intended recipient, you should delete this message.  Any disclosure, copying, or distribution of this message, or the taking of any action based on it, is strictly prohibited.

pgsql-general by date

Next:From: Richard WeltyDate: 2004-04-29 15:16:43
Subject: Re: Anyone using pgreplicator?
Previous:From: Paul ThomasDate: 2004-04-29 14:19:55
Subject: Re: Performance problem with correlated sub-query

Privacy Policy | About PostgreSQL
Copyright © 1996-2018 The PostgreSQL Global Development Group