Skip site navigation (1)
Skip section navigation (2)
Peripheral Links
Header And Logo
|
Site Navigation
Re: efficient pattern queries (using LIKE, ~)
- From: Mario Weilguni <mweilguni(at)sime(dot)com>
- To: Fernan Aguero <fernan(at)unsam(dot)edu(dot)ar>
- Cc: PostgreSQL Performance <pgsql-performance(at)postgresql(dot)org>
- Subject: Re: efficient pattern queries (using LIKE, ~)
- Date: Tue, 11 Sep 2007 21:08:37 +0200
Fernan Aguero schrieb:
Hi, I have a table containing some ~13 million rows. Queries on indexed fields run fast, but unanchored pattern queries on a text column are slooooow. Indexing the column doesn't help (this is already mentioned in the manual). http://www.postgresql.org/docs/8.2/interactive/indexes-types.html However, no alternative solution is mentioned for indexingand/or optimizing queries based on unanchored patterns: i.e. description LIKE '%kinase%'.
Maybe trigram search might help you? Never tried it myself, but it seems to be able to handle substring searches.
- References:
- efficient pattern queries (using LIKE, ~)
- From: Fernan Aguero
- efficient pattern queries (using LIKE, ~)
- Prev by Date: efficient pattern queries (using LIKE, ~)
- Next by Date: Re: random_page_costs - are defaults of 4.0 realistic for SCSI RAID 1
- Previous by thread: efficient pattern queries (using LIKE, ~)
- Next by thread: pg_dump blocking create database?
- Indexes: