Re: 4 Minute Full Text Index searches

[Ueberbach JP]

use sphinx instead. join the pk's from the result with additional information you might need ( when group by needed ( i.e. showing category hits from current ft search result in desc order ), use synchronized (triggers) memory heap tables containing the values used in the group by ). http://sphinxsearch.com

Generally, it's a standalone search engine, meant to provide fast, size-efficient and relevant fulltext search functions to other applications. Sphinx was specially designed to integrate well with SQL databases and scripting languages. Currently built-in data sources support fetching data either via direct connection to MySQL or PostgreSQL, or using XML pipe mechanism (a pipe to indexer in special XML-based format which Sphinx recognizes).

As for the name, Sphinx is an acronym which is officially decoded as SQL Phrase Index. Yes, I know about CMU's Sphinx project.