I’m having real difficulties optimising a MySQL query. I have to use the existing database structure, but I am getting an extremely slow response under certain circumstances.
My query is:
SELECT
`t`.*,
`p`.`trp_name`,
`p`.`trp_lname`,
`trv`.`trv_prosceslevel`,
`trv`.`trv_id`,
`v`.`visa_destcountry`,
`track`.`track_id`,
`track`.`track_datetoembassy`,
`track`.`track_expectedreturn`,
`track`.`track_status`,
`track`.`track_comments`
FROM
(SELECT
*
FROM
`_transactions`
WHERE
DATE(`tr_datecreated`) BETWEEN DATE('2011-07-01 00:00:00') AND DATE('2011-08-01 23:59:59')) `t`
JOIN
`_trpeople` `p` ON `t`.`tr_id` = `p`.`trp_trid` AND `p`.`trp_name` = 'Joe' AND `p`.`trp_lname` = 'Bloggs'
JOIN
`_trvisas` `trv` ON `t`.`tr_id` = `trv`.`trv_trid`
JOIN
`_visas` `v` ON `trv`.`trv_visaid` = `v`.`visa_code`
JOIN
`_trtracking` `track` ON `track`.`track_trid` = `t`.`tr_id` AND `p`.`trp_id` = `track`.`track_trpid` AND `trv`.`trv_id` = `track`.`track_trvid` AND `track`.`track_status` IN ('New','Missing_Info',
'En_Route',
'Ready_Pickup',
'Received',
'Awaiting_Voucher',
'Sent_Client',
'Closed')
ORDER BY `tr_id` DESC
The results of an explain statement on the above is:
id select_type table type possible_keys key key_len ref rows Extra
1 PRIMARY <derived2> ALL NULL NULL NULL NULL 164 Using temporary; Using filesort
1 PRIMARY track ALL status_index NULL NULL NULL 4677 Using where
1 PRIMARY p eq_ref PRIMARY PRIMARY 4 db.track.track_trpid 1 Using where
1 PRIMARY trv eq_ref PRIMARY PRIMARY 4 db.track.track_trvid 1 Using where
1 PRIMARY v eq_ref visa_code visa_code 4 db.trv.trv_visaid 1
2 DERIVED _transactions ALL NULL NULL NULL NULL 4276 Using where
The query times are acceptable until the value of ‘Closed’ is included in the very last track.track_status IN clause. The length of time is then increased about 10 to 15 times the other queries.
This makes sense as the ‘Closed’ status refers to all the clients whose transactions have been dealt with, wihich corresponds to about 90% to 95% of the database.
The issue is, is that in some cases, the search is taking about 45 seconds which is rediculous. I’m sure MySQL can do much better than that and it’s just my query at fault, even if the tables do have 4000 rows, but I can’t work out how to optimise this statement.
I’d be grateful for some advice about where I’m going wrong and how I should be implementing this query to produce a faster result.
Many thanks
Try this: