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

Re: improve select performance...

From: Steve Holdoway <steve(dot)holdoway(at)firetrust(dot)com>
To: pgsql-admin(at)postgresql(dot)org
Subject: Re: improve select performance...
Date: 2007-05-19 02:51:45
Message-ID: 20070519145145.0c16e6fc.steve.holdoway@firetrust.com (view raw or flat)
Thread:
Lists: pgsql-admin
I managed to get this working properly by dropping the index on the url column, and making it the primary key instead. Can anyone point me at docs that define the difference between a primary key and an unique index on the same field? I'm a bit confused here!

Cheers,  Steve

On Wed, 16 May 2007 11:08:10 +1200
Steve Holdoway <steve(dot)holdoway(at)firetrust(dot)com> wrote:

> Can anyone out there help me, I'm having a bit of a problem trying to improve the performance of a php script that does a data load?
> 
> At the moment, the script checks a database to see if the entry is present, and, if not, it adds it. The table has about 400,000 rows, and I'm checking a bulk load of about 50,000 entries. The table is defined as follows (not my design!):
> 
> Welcome to psql 8.2.4, the PostgreSQL interactive terminal.
> 
> Type:  \copyright for distribution terms
>        \h for help with SQL commands
>        \? for help with psql commands
>        \g or terminate with semicolon to execute query
>        \q to quit
> 
> db=> \d badurls
>                      Table "public.badurls"
>    Column    |            Type             |     Modifiers
> -------------+-----------------------------+--------------------
>  url         | character varying           | not null
>  ip          | character(20)               |
>  dateadded   | timestamp without time zone | not null
>  type        | character(1)                |
>  publishdate | timestamp without time zone |
>  status      | integer                     | not null default 1
>  version     | integer                     | default 0
>  edited      | integer                     | default 0
>  category    | character(4)[]              |
> Indexes:
>     "idxbadurls_url" hash (url)
>     "idxbadurls_version" btree (version)
> 
> 
> and I'm accessing the url column. (I've tried a btree index as well... not much difference)
> 
> In an attempt to improve the performance, I'm getting a list of data that's already there in a single hit, and not attempting to insert them ( I'm still checking the rest first, just in case... ).
> 
> So, I'm trying to run a select url from badurls where url in ( .... ) to get this list. However, this is also desperately slow...
> 
> 1000 in the list: 14 sec
> 5000 in the list: 2:40
> 10000 in the list: 5:50
> 
> ( run from a client machine over a 100mbit lan ).
> 
> Anyway, that's enough to prove to me that I need to rejig stuff to get this to work!!!
> 
> Server's a dual xeon machine, with hardware raid, and running debian. It's got 4GB of memory, 1GB is currently not even used for io buffers.
> 
> The postgres instance is configured as standard, except for increasing the shared_buffers to 200MB.
> 
> Explain plan for about 50 entries in the 'in' clause returns
> 
> Bitmap Heap Scan on badurls  (cost=4531.22..15748.67 rows=60462 width=32)
>    Recheck Cond: ((url)::text = ANY (('{.....}'::character varying[])::text[]))
>    ->  Bitmap Index Scan on idxbadurls_url  (cost=0.00..4516.10 rows=60462 width=0)
>          Index Cond: ((url)::text = ANY (('{.....}'::character varying[])::text[]))
> (4 rows)
> 
> 
> Can anyone suggest what I should be reconfiguring???
> 
> Cheers,
> 
> 
> Steve
> 
> ---------------------------(end of broadcast)---------------------------
> TIP 7: You can help support the PostgreSQL project by donating at
> 
>                 http://www.postgresql.org/about/donate

In response to

Responses

pgsql-admin by date

Next:From: Tom LaneDate: 2007-05-19 17:58:53
Subject: Re: improve select performance...
Previous:From: Joshua D. DrakeDate: 2007-05-18 19:42:21
Subject: Re: PostgreSQL Training?

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group