The problem:
I have 1 table of aprox 5000 rows called imported_cities
I have 1 table of aprox 800 000 rows called postal_codes containing postal codes cities
I need to validate each distinct city from imported_cities against the cities in postal codes table based on city name and its province. See tables structure below.
If they match exactly (yes, exactly. The rest of cities are manually validated) I have to update a column on imported_city and
enter both city from imported_cities and city from postal_codes (side by side) into a third table called imported_cities_equiv
What I have tried:
Adding indexes to tables and make query below. It takes forever... :(
explain SELECT DISTINCT ic.destinationCity, pc.city FROM (imported_cities ic, postalcodes pc)
WHERE LOWER(ic.destinationCity) = LOWER(pc.city)
the result
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE ip index NULL company_city 478 NULL 4221 Using index; Using temporary
1 SIMPLE pc index NULL city_prov 160 NULL 765407 Using where; Using index; Using join buffer (Block...
--
-- Table structure for table postalcodes
CREATE TABLE IF NOT EXISTS `postalcodes` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`code` varchar(11) NOT NULL,
`city` varchar(50) NOT NULL,
`province` varchar(50) NOT NULL,
`provinceISO` varchar(2) NOT NULL,
`latitude` decimal(17,13) NOT NULL,
`longitude` decimal(17,13) NOT NULL,
PRIMARY KEY (`id`),
KEY `code` (`code`),
KEY `city_prov` (`city`,`provinceISO`)
--
-- Table structure for table imported_cities
CREATE TABLE IF NOT EXISTS `imported_cities` (
`id` int(11) unsigned NOT NULL AUTO_INCREMENT,
`companyName` varchar(30) CHARACTER SET utf8 NOT NULL,
`destinationCity` varchar(128) CHARACTER SET utf8 NOT NULL,
`destinationProvince` varchar(20) CHARACTER SET utf8 NOT NULL,
`equivCity` varchar(128) CHARACTER SET utf8 DEFAULT NULL,
`minAmount` decimal(6,2) NOT NULL
PRIMARY KEY (`id`),
KEY `company_city` (`companyName`,`destinationCity`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci AUTO_INCREMENT=7933 ;
--
-- Table structure for table imported_cities_equiv
CREATE TABLE IF NOT EXISTS `imported_cities_equiv` (
`id` int(11) unsigned NOT NULL AUTO_INCREMENT,
`imported_city` varchar(128) CHARACTER SET utf8 NOT NULL,
`pc_city` varchar(128) CHARACTER SET utf8 NOT NULL,
`province` varchar(20) CHARACTER SET utf8 NOT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci AUTO_INCREMENT=149 ;
Any help or suggestion is appreciated. Thank you.
The query you want to get your information is:
SELECT ip.*, (pc.city is not null) as exact match
FROM imported_prices ip left join
postalcodes pc
on LOWER(ip.destinationCity) = LOWER(pc.city) and
lower(ip.province) = lower(pc.province);
However, this will have really bad performance. Getting rid of the lower() would help:
SELECT ip.*, (pc.city is not null) as exact match
FROM imported_prices ip left join
postalcodes pc
on(ip.destinationCity) =(pc.city) and
(ip.province) = (pc.province);
Because then you can add an index on postalcodes(city, province).
If you cannot use remove lower(), then alter the table to add new columns and put the lower-case values in those columns. Then build an index on the new columns and use them in the join.
Thank you all for pointing me on the right direction.
Some changes have been made following your advices:
added indexes on imported_cities table on destinationCity and destinationProvince columns
added indexes on postalcodes table on city and provinceISO columns
JOIN clause have only one side upper since the field ic.destinationCity is already in uppercase
limit query by province on WHERE for performance
The final SQL is:
SELECT DISTINCT pc.city, pc.provinceISO
FROM postalcodes pc
LEFT JOIN imported_cities ic
ON upper(pc.city) = ic.destinationCity AND
pc.provinceISO = ic.destinationProvince
WHERE ic.destinationProvince = 'QC';
AND the EXPLAIN
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE pc ref province province 8 const 278115 Using index condition; Using temporary
1 SIMPLE ip ref destinationCity,destinationProvince destinationCity 386 func 1 Using index condition; Using where; Distinct
Going forward I can now construct the INSERT query on PHP and make one INSERT query to insert all equivalent cities on the 3rd table. Thank you all.
Related
I have two huge innodb tables (page: +40M rows, +30Gb and stat: +45M rows, +10Gb). I have a query that selects rows from the join of these two tables and it used to take about a second for execution. Recently it's taking more than 20 seconds (sometime up to few minutes) for the exact same query to be completed. I suspected that with lot's of inserts and updates it might need an optimization. I ran OPTIMIZE TABLE on the table using phpMyAdmin but no improvements. I've Googled a lot but couldn't find any content helping me on this situation.
The query I mentioned earlier looks like below:
SELECT `c`.`unique`, `c`.`pub`
FROM `pages` `c`
LEFT JOIN `stat` `s` ON `c`.`unique`=`s`.`unique`
WHERE `s`.`isc`='1'
AND `s`.`haa`='0'
AND (`pubID`='24')
ORDER BY `eid` ASC LIMIT 0, 10
These are the tables structure:
CREATE TABLE `pages` (
`eid` int(10) UNSIGNED NOT NULL,
`ti` text COLLATE utf8_persian_ci NOT NULL,
`fat` text COLLATE utf8_persian_ci NOT NULL,
`de` text COLLATE utf8_persian_ci NOT NULL,
`fad` text COLLATE utf8_persian_ci NOT NULL,
`pub` varchar(100) COLLATE utf8_persian_ci NOT NULL,
`pubID` int(10) UNSIGNED NOT NULL,
`pubn` text COLLATE utf8_persian_ci NOT NULL,
`unique` tinytext COLLATE utf8_persian_ci NOT NULL,
`pi` tinytext COLLATE utf8_persian_ci NOT NULL,
`kw` text COLLATE utf8_persian_ci NOT NULL,
`fak` text COLLATE utf8_persian_ci NOT NULL,
`te` text COLLATE utf8_persian_ci NOT NULL,
`fae` text COLLATE utf8_persian_ci NOT NULL,
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_persian_ci;
ALTER TABLE `pages`
ADD PRIMARY KEY (`eid`),
ADD UNIQUE KEY `UNIQ` (`unique`(128)),
ADD KEY `pub` (`pub`),
ADD KEY `unique` (`unique`(128)),
ADD KEY `pubID` (`pubID`) USING BTREE;
ALTER TABLE `pages` ADD FULLTEXT KEY `faT` (`fat`);
ALTER TABLE `pages` ADD FULLTEXT KEY `faA` (`fad`,`fae`);
ALTER TABLE `pages` ADD FULLTEXT KEY `faK` (`fak`);
ALTER TABLE `pages` ADD FULLTEXT KEY `pubn` (`pubn`);
ALTER TABLE `pages` ADD FULLTEXT KEY `faTAK` (`fat`,`fad`,`fak`,`fae`);
ALTER TABLE `pages` ADD FULLTEXT KEY `ab` (`de`,`te`);
ALTER TABLE `pages` ADD FULLTEXT KEY `Ti` (`ti`);
ALTER TABLE `pages` ADD FULLTEXT KEY `Kw` (`kw`);
ALTER TABLE `pages` ADD FULLTEXT KEY `TAK` (`ti`,`de`,`kw`,`te`);
ALTER TABLE `pages`
MODIFY `eid` int(10) UNSIGNED NOT NULL AUTO_INCREMENT;
CREATE TABLE `stat` (
`sid` int(10) UNSIGNED NOT NULL,
`unique` tinytext COLLATE utf8_persian_ci NOT NULL,
`haa` tinyint(1) UNSIGNED NOT NULL,
`isc` tinyint(1) NOT NULL,
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_persian_ci;
ALTER TABLE `stat`
ADD PRIMARY KEY (`sid`),
ADD UNIQUE KEY `Unique` (`unique`(128)),
ADD KEY `isc` (`isc`),
ADD KEY `haa` (`haa`),
ALTER TABLE `stat`
MODIFY `sid` int(10) UNSIGNED NOT NULL AUTO_INCREMENT;
The following query took only 0.0126 seconds with 38685601 total results as said by phpMyAdmin:
SELECT `sid` FROM `stat` WHERE `s`.`isc`='1' AND `s`.`haa`='0'
and this one took 0.0005 seconds with 5159484 total results
SELECT `eid`, `unique`, `pubn`, `pi` FROM `pages` WHERE `pubID`='24'
Am I missing something? Can anybody help?
The slowdown is probably due to scanning so many rows, and that is now more than can fit in cache. So, let's try to improve the query.
Replace INDEX(pubID) with INDEX(pubID, eid) -- This may allow both the WHERE and ORDER BY to be handled by the index, thereby avoiding a sort.
Replace TINYTEXT with VARCHAR(255) or some smaller limit. This may speed up tmp tables.
Don't use prefix index on eid -- its an INT !
Don't say UNIQUE with prefixing -- UNIQUE(x(128)) only checks the uniqueness of the first 128 columns !
Once you change to VARCHAR(255) (or less), you can apply UNIQUE to the entire column.
The biggest performance issue is filtering on two tables -- can you move the status flags into the main table?
Change LEFT JOIN to JOIN.
What does unique look like? If it is a "UUID", that could further explain the trouble.
If that is a UUID that is 39 characters, the string can be converted to a 16-byte column for further space savings (and speedup). Let's discuss this further if necessary.
5 million results in 0.5ms is bogus -- it was fetching from the Query cache. Either turn off the QC or run with SELECT SQL_NO_CACHE...
+1 to #RickJames answer, but following it I have done a test.
I would also recommend you do not use the name unique for a column name, because it's an SQL reserved word.
ALTER TABLE pages
CHANGE `unique` objectId VARCHAR(128) NOT NULL COMMENT 'Document Object Identifier',
DROP KEY pubId,
ADD KEY bktest1 (pubId, eid, objectId, pub);
ALTER TABLE stat
CHANGE `unique` objectId VARCHAR(128) NOT NULL COMMENT 'Document Object Identifier',
DROP KEY `unique`,
ADD UNIQUE KEY bktest2 (objectId, isc, haa);
mysql> explain SELECT `c`.`objectId`, `c`.`pub` FROM `pages` `c` JOIN `stat` `s` ON `c`.`objectId`=`s`.`objectId` WHERE `s`.`isc`='1' AND `s`.`haa`='0' AND (`pubID`='24') ORDER BY `eid` ASC LIMIT 0, 10;
+----+-------------+-------+------------+--------+-------------------------+---------+---------+-----------------------------+------+----------+--------------------------+
| id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra |
+----+-------------+-------+------------+--------+-------------------------+---------+---------+-----------------------------+------+----------+--------------------------+
| 1 | SIMPLE | c | NULL | ref | unique,unique_2,bktest1 | bktest1 | 4 | const | 1 | 100.00 | Using where; Using index |
| 1 | SIMPLE | s | NULL | eq_ref | bktest2,haa,isc | bktest2 | 388 | test.c.objectId,const,const | 1 | 100.00 | Using index |
+----+-------------+-------+------------+--------+-------------------------+---------+---------+-----------------------------+------+----------+--------------------------+
By creating the multi-column indexes, this makes them covering indexes, and you see "Using index" in the EXPLAIN report.
It's important to put eid second in the bktest1 index, so you avoid a filesort.
This is the best you can hope to optimize this query without denormalizing or partitioning the tables.
Next you should make sure your buffer pool is large enough to hold all the requested data.
I wish to know how I can create indexes in my database according to my data structure. most of my queries are fetching data against the ID and the name as well with two or three tables joining while pagination. please advise how to make indexes according to below queries.
Query:1
SELECT DISTINCT topic, type FROM books where type like 'Tutor-Books' order by topic
Explain:
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE books range faith faith 102 NULL 132 Using index condition; Using temporary; Using filesort
Query:2
SELECT books.name, books.name2, books.id, books.image, books.faith,
books.topic, books.downloaded, books.viewed, books.language,
books.size, books.author as author_id, authors.name as author_name,
authors.aid
from books
LEFT JOIN authors ON books.author = authors.aid
WHERE books.id = '".$id."'
AND status = 1
Explain:
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE books const PRIMARY PRIMARY 4 const 1 NULL
1 SIMPLE authors const aid aid 4 const 1 NULL
Can i use indexes for pagination in offset case where same query returns total:
SELECT SQL_CALC_FOUND_ROWS books.name, books.name2, books.id,
books.image, books.topic, books.author as author_id,
authors.name as author_name, authors.aid
from books
LEFT JOIN authors ON books.author = authors.aid
WHERE books.author = '$pid'
AND status = 1
ORDER BY books.name
LIMIT $limit OFFSET $offset
Do I need to update my queries after creating indexes. please also suggest what should be the table format.
SHOW CREATE TABLE books:
Table Create Table
books CREATE TABLE `books` (
`name` varchar(100) CHARACTER SET utf8 COLLATE utf8_unicode_ci NOT NULL,
`name2` varchar(150) CHARACTER SET utf8 COLLATE utf8_unicode_ci NOT NULL,
`author` int(100) NOT NULL,
`translator` int(120) NOT NULL,
`publisher` int(100) NOT NULL,
`pages` int(50) NOT NULL,
`date` varchar(50) CHARACTER SET latin1 NOT NULL,
`downloaded` int(100) NOT NULL,
`alt_lnk` text NOT NULL,
`viewed` int(100) NOT NULL,
`language` varchar(100) CHARACTER SET latin1 NOT NULL,
`image` varchar(200) CHARACTER SET latin1 NOT NULL,
`faith` varchar(100) CHARACTER SET latin1 NOT NULL,
`id` int(100) NOT NULL AUTO_INCREMENT,
`sid` varchar(1200) CHARACTER SET latin1 DEFAULT NULL,
`topic` varchar(100) CHARACTER SET latin1 NOT NULL,
`last_viewed` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
`size` double NOT NULL,
`status` int(2) NOT NULL DEFAULT '0',
`is_scroll` int(2) NOT NULL,
`is_downloaded` int(2) NOT NULL,
`pdf_not_found` int(2) NOT NULL,
PRIMARY KEY (`id`),
KEY `name` (`name`),
KEY `downloaded` (`downloaded`),
KEY `name2` (`name2`),
KEY `topic` (`topic`),
KEY `faith` (`faith`)
) ENGINE=InnoDB AUTO_INCREMENT=12962 DEFAULT CHARSET=utf8
where type like 'Tutor-Books' order by topic (or:)
where type = 'Tutor-Books' order by topic
--> INDEX(type, topic)
where type like '%Tutor-Books' order by topic
--> INDEX(topic) -- the leading % prevents indexing
LEFT JOIN authors ON books.author = authors.aid
--> PRIMARY KEY(aid)
Do you really need LEFT JOIN? If you can change it to JOIN, the optimizer might be able to start with authors. If it does, then
--> INDEX(author) -- in `books`
My cookbook for building indexes.
Other tips:
INT(100) and INT(2) are identical -- each is a 4-byte signed integer. Read about TINYINT UNSIGNED for numbers 0..255, etc. Use that for your flags (status, is_scroll, etc)
DATE is a datatype; using a VARCHAR is problematic if you ever want to compare or order.
Learn about composite indexes, such as my first example.
Your display widths are a little funky, but that wont cause a problem.
Query 1:
You're using the LIKE operator without a wildcard search %. You can likely swap this with an = operator.
I don't see the column type in your SHOW CREATE TABLE -- but it seems you don't have an index here, unless you renamed it to faith.
Do you need to type to be a string? could it be abstracted to a types table and then joined against using an integer? Or, if you have a fixed amount of types that's unlikely to change, could you use an enum?
Query 2:
You don't need to quote strings, also that's probably vulnerable to SQL injection. do ='.intval($id).' instead.
Make sure you have an index on authors.aid and that they're of the same type.
Having some real issues with a few queries, this one inparticular. Info below.
tgmp_games, about 20k rows
CREATE TABLE IF NOT EXISTS `tgmp_games` (
`g_id` int(8) NOT NULL AUTO_INCREMENT,
`site_id` int(6) NOT NULL,
`g_name` varchar(255) NOT NULL,
`g_link` varchar(255) NOT NULL,
`g_url` varchar(255) NOT NULL,
`g_platforms` varchar(128) NOT NULL,
`g_added` datetime NOT NULL,
`g_cover` varchar(255) NOT NULL,
`g_impressions` int(8) NOT NULL,
PRIMARY KEY (`g_id`),
KEY `g_platforms` (`g_platforms`),
KEY `site_id` (`site_id`),
KEY `g_link` (`g_link`),
KEY `g_release` (`g_release`),
KEY `g_genre` (`g_genre`),
KEY `g_name` (`g_name`),
KEY `g_impressions` (`g_impressions`)
) ENGINE=MyISAM DEFAULT CHARSET=latin1;
tgmp_reviews - about 200k rows
CREATE TABLE IF NOT EXISTS `tgmp_reviews` (
`r_id` int(8) NOT NULL AUTO_INCREMENT,
`site_id` int(6) NOT NULL,
`r_source` varchar(128) NOT NULL,
`r_date` date NOT NULL,
`r_score` int(3) NOT NULL,
`r_copy` text NOT NULL,
`r_link` text NOT NULL,
`r_int_link` text NOT NULL,
`r_parent` int(8) NOT NULL,
`r_platform` varchar(12) NOT NULL,
`r_impressions` int(8) NOT NULL,
PRIMARY KEY (`r_id`),
KEY `site_id` (`site_id`),
KEY `r_parent` (`r_parent`),
KEY `r_platform` (`r_platform`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 ;
Here is the query, takes 3 seconds ish
SELECT * FROM tgmp_games g
RIGHT JOIN tgmp_reviews r ON g_id = r.r_parent
WHERE g.site_id = '34'
GROUP BY g_name
ORDER BY g_impressions DESC LIMIT 15
EXPLAIN
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE r ALL r_parent NULL NULL NULL 201133 Using temporary; Using filesort
1 SIMPLE g eq_ref PRIMARY,site_id PRIMARY 4 engine_comp.r.r_parent 1 Using where
I am just trying to grab the 15 most viewed games, then grab a single review (doesnt really matter which, I guess highest rated would be ideal, r_score) for each game.
Can someone help me figure out why this is so horribly inefficient?
I don't understand what is the purpose of having a GROUP BY g_name in your query, but this makes MySQL performing aggregates on the columns selected, or all columns from both table. So please try to exclude it and check if it helps.
Also, RIGHT JOIN makes database to query tgmp_reviews first, which is not what you want. I suppose LEFT JOIN is a better choice here. Please, try to change the join type.
If none of the first options helps, you need to redesign your query. As you need to obtain 15 most viewed games for the site, the query will be:
SELECT g_id
FROM tgmp_games g
WHERE site_id = 34
ORDER BY g_impressions DESC
LIMIT 15;
This is the very first part that should be executed by the database, as it provides the best selectivity. Then you can get the desired reviews for the games:
SELECT r_parent, max(r_score)
FROM tgmp_reviews r
WHERE r_parent IN (/*1st query*/)
GROUP BY r_parent;
Such construct will force database to execute the first query first (sorry for the tautology) and will give you the maximal score for each of the wanted games. I hope you will be able to use the obtained results for your purpose.
Your MyISAM table is small, you can try converting it to see if that resolves the issue. Do you have a reason for using MyISAM instead of InnoDB for that table?
You can also try running an analyze on each table to update the statistics to see if the optimizer chooses something different.
i have a very simple query that im trying to optimize, its taking 2~5 secs to execute.
This is my CREATE TABLE
CREATE TABLE `artist` (
`id` INTEGER NOT NULL AUTO_INCREMENT,
`name` VARCHAR(100) character set utf8 NOT NULL,
`bio` MEDIUMTEXT character set utf8 DEFAULT NULL,
`hits` INTEGER NOT NULL,
PRIMARY KEY (`id`)
);
CREATE TABLE `album` (
`id` INTEGER NOT NULL AUTO_INCREMENT,
`artist_id` INTEGER NOT NULL,
`title` VARCHAR(100) character set utf8 NOT NULL,
`year` INTEGER,
`hits` INTEGER NOT NULL,
PRIMARY KEY (`id`),
KEY (`artist_id`)
);
CREATE TABLE `track` (
`id` INTEGER NOT NULL AUTO_INCREMENT,
`name` VARCHAR(100) character set utf8 NOT NULL,
`lyric` MEDIUMTEXT character set utf8,
`album_id` INTEGER NOT NULL,
`hits` INTEGER NOT NULL,
`date` datetime NOT NULL,
PRIMARY KEY (`id`),
KEY (`album_id`)
);
ALTER TABLE `album` ADD FOREIGN KEY (artist_id) REFERENCES `artist` (`id`);
ALTER TABLE `track` ADD FOREIGN KEY (album_id) REFERENCES `album` (`id`);
and this is the query im running
SELECT DISTINCT artist.name, track.name
FROM track
LEFT JOIN album ON track.album_id = album.id
LEFT JOIN artist ON album.artist_id = artist.id
ORDER BY track.hits DESC
LIMIT 5
Explain selects show this:
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE track ALL NULL NULL NULL NULL 103796 Using temporary; Using filesort
1 SIMPLE album eq_ref PRIMARY PRIMARY 4 lyrics.track.album_id 1
1 SIMPLE artist eq_ref PRIMARY PRIMARY 4 lyrics.album.artist_id 1
I'm new to MySQL but i guess using Using temporary; Using filesort is bad and thats why the query is very slow, can you guys hint me here? thanks!
update: The main problem here is that the very same song can be 5 times in the DB with different ID's, because the same song can be in different albums. If i dont use distinct, this doesnt happen, bust i must for this reason
This answer isn't 100% an answer for the original question. The original question is what came up when searching using the messages from my problem though, so just in case it helps someone else, I'll leave the solution for a problem that is closely related.
The "using temporary; using filesort" was actually a red herring and the index that was added was never getting used. The index was not getting used because one of the joined tables had a different character encoding on it than the other.
Converting all tables in the query so that they all used the same character encoding fixed it instantly.
(In our case converting a utf8 encoded table to a latin1 encoding)
Hope it helps someone.
You can get it to use an index by adding
create index idx_tracks_on_album_id_name_hits on track(album_id, name, hits);
And since you are doing a DISTINCT across two tables, there will be no index to possibly find the unique rows so it puts it into a temp table to get rid of the duplicates.
I think if you create an index on track.hits, you might get rid of "using temporary; using filesort", the reason for which might be because MySQL cannot find an index to do the sort.
ALTER TABLE `track`
ADD KEY `idx_hits` (`hits`);
Let me know if it worked.
why do you use DISTINCT? why do you use LEFT JOIN (insted of JOIN)?
This query:
explain
SELECT `Lineitem`.`id`, `Donation`.`id`, `Donation`.`order_line_id`
FROM `order_line` AS `Lineitem`
LEFT JOIN `donations` AS `Donation`
ON (`Donation`.`order_line_id` = `Lineitem`.`id`)
WHERE `Lineitem`.`session_id` = '1'
correctly uses the Donation.order_line_id and Lineitem.id indexes, shown in this EXPLAIN output:
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE Lineitem ref session_id session_id 97 const 1 Using where; Using index
1 SIMPLE Donation ref order_line_id order_line_id 4 Lineitem.id 2 Using index
However, this query, which simply includes another field:
explain
SELECT `Lineitem`.`id`, `Donation`.`id`, `Donation`.`npo_id`,
`Donation`.`order_line_id`
FROM `order_line` AS `Lineitem`
LEFT JOIN `donations` AS `Donation`
ON (`Donation`.`order_line_id` = `Lineitem`.`id`)
WHERE `Lineitem`.`session_id` = '1'
Shows that the Donation table does not use an index:
id select_type table type possible_keys key key_len ref rows Extra
1 SIMPLE Lineitem ref session_id session_id 97 const 1 Using where; Using index
1 SIMPLE Donation ALL order_line_id NULL NULL NULL 3
All of the _id fields in the tables are indexed, but I can't figure out how adding this field into the list of selected fields causes the index to be dropped.
As requested by James C, here are the table definitions:
CREATE TABLE `donations` (
`id` int(10) unsigned NOT NULL auto_increment,
`npo_id` int(10) unsigned NOT NULL,
`order_line_detail_id` int(10) unsigned NOT NULL default '0',
`order_line_id` int(10) unsigned NOT NULL default '0',
`created` datetime default NULL,
`modified` datetime default NULL,
PRIMARY KEY (`id`),
KEY `npo_id` (`npo_id`),
KEY `order_line_id` (`order_line_id`),
KEY `order_line_detail_id` (`order_line_detail_id`)
) ENGINE=InnoDB AUTO_INCREMENT=7 DEFAULT CHARSET=utf8
CREATE TABLE `order_line` (
`id` bigint(20) unsigned NOT NULL auto_increment,
`order_id` bigint(20) NOT NULL,
`npo_id` bigint(20) NOT NULL default '0',
`session_id` varchar(32) collate utf8_unicode_ci default NULL,
`created` datetime default NULL,
PRIMARY KEY (`id`),
KEY `order_id` (`order_id`),
KEY `npo_id` (`npo_id`),
KEY `session_id` (`session_id`)
) ENGINE=InnoDB AUTO_INCREMENT=23 DEFAULT CHARSET=utf8
I also did some reading about cardinality, and it looks like both the Donations.npo_id and Donations.order_line_id have a cardinality of 2. Hopefully this suggests something useful?
I'm thinking that a USE INDEX might solve the problem, but I'm using an ORM that makes this a bit tricky, and I don't understand why it wouldn't grab the correct index when the JOIN specifically names indexed fields?!?
Thanks for your brainpower!
The first explain has "uses index" at the end. This means that it was able to find the rows and return the result for the query by just looking at the index and not having to fetch/analyse any row data.
In the second query you add a row that's likely not indexed. This means that MySQL has to look at the data of the table. I'm not sure why the optimiser chose to do a table scan but I think it's likely that if the table is fairly small it's easier for it to just read everything than trying to pick out details for individual rows.
edit: I think adding the following indexes will improve things even more and let all of the join use indexes only:
ALTER TABLE order_line ADD INDEX(session_id, id);
ALTER TABLE donations ADD INDEX(order_line_id, npo_id, id)
This will allow order_line to to find the rows using session_id and then return id and also allow donations to join onto order_line_id and then return the other two columns.
Looking at the auto_increment values can I assume that there's not much data in there. It's worth noting that the amount of data in the tables will have an effect on the query plan and it's good practice to put some sample data in there to test things out. For more detail have a look in this blog post I made some time back: http://webmonkeyuk.wordpress.com/2010/09/27/what-makes-a-good-mysql-index-part-2-cardinality/