matt on Thursday, July 15, 2004, 11:58:31 AM, wrote: >>Consider replicating to some slave servers and dividing reads among them.
mr> I already replicate to slaves, and sites will do read only queries off mr> these slaves mr> 99.9 % of the tables are read only anyway, the only tables we update or mr> insert into, are very very small and fast. mr> These big tables are daily extracts from IBM DB2 sites, in ebcdic mr> format, we archive the data and users then query our site which is mr> faster, unless they start doing multiple query options, then things get mr> slow. mr> If you query only one feild its FAST, but if you query two feilds, its mr> slow, very slow, need multiple key per query support in mysql. One thing that can help (at least for me) is working with temporary tables, spliting one complex query in many littles. A simple example: 2 tables: header and details CREATE TEMPORARY TABLE tmp TYPE=HEAP SELECT hdr_code FROM header WHERE ...; ALTER TABLE tmp ADD PRIMARY KEY(hdr_code); SELECT d.* FROM details d INNER JOIN tmp ON d.hdr_code=tmp.hdr_code WHERE ...; DROP TABLE TMP; -- Alejandro D. Burne Departamento de Sistemas Asociación Médica Rosario España 401 - S2000SBI Rosario - Santa Fe - Argentina +54-0341-4252313 Interno 145 -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
