4 messages in com.mysql.lists.perlRE: Table Sizes
FromSent OnAttachments
Bill Platt16 Feb 2006 14:47 
Logan, David (SST - Adelaide)16 Feb 2006 15:10 
Jan Kratochvil16 Feb 2006 15:28 
Greg Meckes16 Feb 2006 18:27 
Subject:RE: Table Sizes
From:Logan, David (SST - Adelaide) (Davi@hp.com)
Date:02/16/2006 03:10:36 PM
List:com.mysql.lists.perl

Hi Bill,

You don't need to write your own software, you could explore the use of MERGE tables (available at 4.?) but if you can hang out for a little while, version 5.1 has a new scheme named partitioning that has greater functionality and maybe what you are looking for.

For the moment though, http://dev.mysql.com/doc/refman/5.0/en/merge-storage-engine.html may be the way to go. This is totally independent from the DBI/DBD perl modules so shouldn't affect any of your scripting. This is also available at version 4.?

Regards

---------------------------------------------------------------

-----Original Message----- From: Bill Platt [mailto:bpl@thephantomwriters.com] Sent: Friday, 17 February 2006 9:17 AM To: pe@lists.mysql.com Subject: Table Sizes

I am wondering which is most effective.

Setting up a table with smaller pieces and writing the software to search smaller tables, OR setting up one big table with all of the related data inside of it?

I have a mysql database I am working with that has 32 million records in it. I am wondering if I would be better off to somehow break that data into much smaller pieces so that there are fewer rows, or whether it is just as effective to keep everything in one table.

Maybe I should break it down alphabetically or something.

The reason for my concern about this is when I access the database through phpMyadmin, every request takes long time to load up. Even minor changes seem to be a real pain in the you know what.

Speed within the site is of vast importance to me.

Thanks for your advice.