MySql manual says:
========================
5.7.2.10 Table Optimization <cid:part...@terra.es>
<cid:part...@terra.es>
To coalesce fragmented records and eliminate wasted space resulting from
deleting or updating records, run @command{myisamchk} in recovery mode:
shell> myisamchk -r tbl_name
=============================
I don't know if you are using myisam or innodb tables. Search for
"fragment" in MySql Manual.
Of course, if disk space is fragmented too you will need to get rid of
that too.
Greg Quinn escribió:
Its definately not a network prob, cached queries of 3000 rows are returned
almost immediately.
Its just the initial read that is very slow.
How do you mean physical space? You mean if hard drive space is fragmented?
A direct table scan can take longer than an index scan if phisical
space is too fragmented.
Can you test your query in the box where mysql lives? Just to be sure is
not a net problem. A faulty net adapter can be losing information,
5 messages in com.mysql.lists.win32Re: Re : 3000 rows, 33 seconds, then ...