Mysql optimization for select query with IN() clause inside where clause (explain output given) - optimization

Question

I have this query:-
SELECT SUM(DISTINCT( ttagrels.id_tag IN ( 1816, 2642, 1906, 1398,
2436, 2940, 1973, 2791, 1389 ) )) AS
key_1_total_matches,
IF(( od.id_od > 0 ), COUNT(DISTINCT( od.id_od )), 0) AS
tutor_popularity,
td.*,
u.*
FROM tutor_details AS td
JOIN users AS u
ON u.id_user = td.id_user
JOIN all_tag_relations AS ttagrels
ON td.id_tutor = ttagrels.id_tutor
LEFT JOIN learning_packs AS lp
ON ttagrels.id_lp = lp.id_lp
LEFT JOIN learning_packs_categories AS lpc
ON lpc.id_lp_cat = lp.id_lp_cat
LEFT JOIN learning_packs_categories AS lpcp
ON lpcp.id_lp_cat = lpc.id_parent
LEFT JOIN learning_pack_content AS lpct
ON ( lp.id_lp = lpct.id_lp )
LEFT JOIN webclasses AS wc
ON ttagrels.id_wc = wc.id_wc
LEFT JOIN learning_packs_categories AS wcc
ON wcc.id_lp_cat = wc.id_wp_cat
LEFT JOIN learning_packs_categories AS wccp
ON wccp.id_lp_cat = wcc.id_parent
LEFT JOIN order_details AS od
ON td.id_tutor = od.id_author
LEFT JOIN orders AS o
ON od.id_order = o.id_order
WHERE ( u.country = 'IE'
OR u.country IN ( 'INT' ) )
AND u.status = 1
AND CASE
WHEN ( lp.id_lp > 0 ) THEN lp.id_status = 1
AND lp.published = 1
AND lpcp.status = 1
AND ( lpcp.country_code = 'IE'
OR lpcp.country_code IN ( 'INT' )
)
ELSE 1
END
AND CASE
WHEN ( wc.id_wc > 0 ) THEN wc.wc_api_status = 1
AND wc.id_status = 1
AND wc.wc_type = 0
AND
wc.class_date > '2010-06-16 11:44:40'
AND wccp.status = 1
AND ( wccp.country_code = 'IE'
OR wccp.country_code IN ( 'INT' )
)
ELSE 1
END
AND CASE
WHEN ( od.id_od > 0 ) THEN od.id_author = td.id_tutor
AND o.order_status = 'paid'
AND CASE
WHEN ( od.id_wc > 0 ) THEN od.can_attend_class = 1
ELSE 1
END
ELSE 1
END
AND ( ttagrels.id_tag IN ( 1816, 2642, 1906, 1398,
2436, 2940, 1973, 2791, 1389 ) )
GROUP BY td.id_tutor
HAVING key_1_total_matches = 1
ORDER BY tutor_popularity DESC,
u.surname ASC,
u.name ASC
LIMIT 0, 20
The numbers inside the IN() are actually ids of another table called Tags which matched the search keywords entered by users. In this example the user has searched "class".
See the explain output of this query here:-
http://www.test.examvillage.com/Screenshot.png
The time taken by this query is 0.0536 sec
But, if the number of values in the ttagrels.id_tag in () increases (as user enters more search keywords), the execution time rises to around 1-5 seconds and more.For example, if user searched "class Available to tutors and students 3 times a day"
the execution time is 4.2226 sec. The explain query output for this query contains 2513 under rows.
There are a total of 6,152 records in the All_Tag_Relations table. Is any further optimization possible?

Answer 1

I don't know your databse, so I cannot give a definitive answer.
I notice in your query the Distinct keyword twice SUM(DISTINCT( and COUNT(DISTINCT(
This implies that your select statement returns several rows with the same id_tag or id_od. This means that your intermediate result before grouping contains to many rows. You must narrow down your where clauses to limit the rows.
Insert count(*) in your query, and observe the number of rows. If it is larger than expected, this explains your problem.