It has been my (unfortunate) experience that OLAP type applications are not MySQL's strong point. Large dataset applications involving queries that perform aggregations, and scan most/all of the dataset tend to take a very very long time to execute on MySQL even when using a star-schema (although that *does* help -- if you try such on a normalized schema be prepared for queries that take days or more to run).
I don't know how MySQL handles very large temporary tables on systems that don't support large files, so unless someone can chime in with reassurances that it can work around such a limitation I'd be sure that you use a large-file capable system regardless of the table type you choose. -JF > -----Original Message----- > From: Nigel Edwards [mailto:[EMAIL PROTECTED]] > Sent: Tuesday, April 30, 2002 7:19 AM > To: [EMAIL PROTECTED] > Subject: Large Tables > > > I need to use some large tables using MySQL under Linux and would welcome > suggestions from anyone with any experience. Currently one months data > creates a table of 3M records about 750Mb in size. I should like to > consolidate 12 Months data (which is expected to grow to say 50M > records per > month) so I could be looking at 15Gb table plus indexes. This is for > 'warehousing' so transactional security and frequent updates are of little > concern. > > The options I can see include: > Merge Tables/INNODB or MyIsam using large file capable file-system (XFS?) > > Please offer ideas and experience. I don't want to reinvent the wheel. > > Thanks > > Nigel > > > > --------------------------------------------------------------------- > Before posting, please check: > http://www.mysql.com/manual.php (the manual) > http://lists.mysql.com/ (the list archive) > > To request this thread, e-mail <[EMAIL PROTECTED]> > To unsubscribe, e-mail > <[EMAIL PROTECTED]> > Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php > > --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php