我一直试图弄清楚为什么这个查询需要这么长时间。之前，它在大约150毫秒到200毫秒内执行，但现在它需要25秒或更长时间。这是在昨晚和这个mroning之间。唯一改变的是向表中添加数据。
根据下面的解释输出以及我提供的sql，有什么突出的东西可以解释为什么查询需要这么长时间吗？
更新：根据halfer的评论，我将订单更改为ORDER BY pins.pin_id，并将执行时间延长至172毫秒。
另外，我想提一下，表格引脚有2000万条记录，董事会有339,000条记录，用户有352738条记录。
+----+--------------------+---------------+-----------------+----------------------------------------------+---------+---------+--------------------------------+------+--------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+--------------------+---------------+-----------------+----------------------------------------------+---------+---------+--------------------------------+------+--------------------------+
| 1 | PRIMARY | <derived11> | system | NULL | NULL | NULL | NULL | 1 | |
| 1 | PRIMARY | pins | index | user_id,user_id_2,latitude,longitude,lat_lon | vip | 1 | NULL | 646 | Using where |
| 1 | PRIMARY | users | eq_ref | PRIMARY | PRIMARY | 8 | skoovy_prd.pins.user_id | 1 | |
| 1 | PRIMARY | boards | eq_ref | PRIMARY | PRIMARY | 8 | skoovy_prd.pins.board_id | 1 | |
| 1 | PRIMARY | via | eq_ref | PRIMARY | PRIMARY | 8 | skoovy_prd.pins.via | 1 | |
| 12 | DEPENDENT SUBQUERY | users | unique_subquery | PRIMARY | PRIMARY | 8 | func | 1 | Using where |
| 11 | DERIVED | NULL | NULL | NULL | NULL | NULL | NULL | NULL | No tables used |
| 10 | DEPENDENT SUBQUERY | users_avatars | ref | pin_id | pin_id | 14 | skoovy_prd.users.user_id,const | 1 | Using where |
| 9 | DEPENDENT SUBQUERY | users_avatars | ref | pin_id | pin_id | 14 | skoovy_prd.users.user_id,const | 1 | Using where |
| 8 | DEPENDENT SUBQUERY | users_avatars | ref | pin_id | pin_id | 14 | skoovy_prd.users.user_id,const | 1 | Using where |
| 7 | DEPENDENT SUBQUERY | pins_images | ref | pin_id | pin_id | 14 | skoovy_prd.pins.pin_id,const | 1 | Using where |
| 6 | DEPENDENT SUBQUERY | pins_images | ref | pin_id | pin_id | 14 | skoovy_prd.pins.pin_id,const | 1 | Using where |
| 5 | DEPENDENT SUBQUERY | pins_images | ref | pin_id | pin_id | 14 | skoovy_prd.pins.pin_id,const | 1 | Using where |
| 4 | DEPENDENT SUBQUERY | pins_images | ref | pin_id | pin_id | 14 | skoovy_prd.pins.pin_id,const | 1 | Using where |
| 3 | DEPENDENT SUBQUERY | pins_reports | index | user_id | user_id | 115 | NULL | 1 | Using where; Using index |
| 2 | DEPENDENT SUBQUERY | pins_likes | ref | pin_id,pin_id_2,user_id | pin_id | 16 | skoovy_prd.pins.pin_id,const | 1 | Using index |
+----+--------------------+---------------+-----------------+----------------------------------------------+---------+---------+--------------------------------+------+--------------------------+
SQL：
SELECT
`pins`.`pin_id` AS `pin_pin_id`,
`pins`.`check_unique_id` AS `pin_check_unique_id`,
`pins`.`category_id` AS `pin_category_id`,
`pins`.`board_id` AS `pin_board_id`,
`pins`.`user_id` AS `pin_user_id`,
`pins`.`date_added` AS `pin_date_added`,
`pins`.`date_modified` AS `pin_date_modified`,
`pins`.`likes` AS `pin_likes`,
`pins`.`comments` AS `pin_comments`,
`pins`.`repins` AS `pin_repins`,
`pins`.`description` AS `pin_description`,
`pins`.`title` AS `pin_title`,
`pins`.`image` AS `pin_image`,
`pins`.`price` AS `pin_price`,
`pins`.`price_currency_code` AS `pin_price_currency_code`,
`pins`.`price_type` AS `pin_price_type`,
`pins`.`price_from` AS `pin_price_from`,
`pins`.`price_to` AS `pin_price_to`,
`pins`.`event_type` AS `pin_event_type`,
`pins`.`event_start` AS `pin_event_start`,
`pins`.`event_end` AS `pin_event_end`,
`pins`.`from` AS `pin_from`,
`pins`.`from_md5` AS `pin_from_md5`,
`pins`.`from_repin` AS `pin_from_repin`,
`pins`.`is_video` AS `pin_is_video`,
`pins`.`views` AS `pin_views`,
`pins`.`latest_comments` AS `pin_latest_comments`,
`pins`.`source_id` AS `pin_source_id`,
`pins`.`via` AS `pin_via`,
`pins`.`repin_from` AS `pin_repin_from`,
`pins`.`public` AS `pin_public`,
`pins`.`ub_id` AS `pin_ub_id`,
`pins`.`total_views` AS `pin_total_views`,
`pins`.`delete_request` AS `pin_delete_request`,
`pins`.`pinmarklet` AS `pin_pinmarklet`,
`pins`.`vip` AS `pin_vip`,
`pins`.`store` AS `pin_store`,
`pins`.`width` AS `pin_width`,
`pins`.`height` AS `pin_height`,
`pins`.`ext` AS `pin_ext`,
`pins`.`location` AS `pin_location`,
`pins`.`latitude` AS `pin_latitude`,
`pins`.`longitude` AS `pin_longitude`,
IF(pins.price_type != ''
AND (pins.price_type = 'CUSTOM_RANGE'
AND pins.price_from <> 0.00
AND pins.price_to <> 0.00
AND pins.price_from < pins.price_to)
OR (pins.price_from <> 0.00),
1,
0) AS `pin_show_pricing`,
IF(pins.price_type = 'CUSTOM_RANGE',
CONCAT('$',
pins.price_from,
'-',
'$',
pins.price_to),
CONCAT('$', pins.price_from)) AS `pin_price_text`,
IF((pins.event_start IS NOT NULL
|| pins.event_end IS NOT NULL)
&& (pins.event_start != '0000-00-00 00:00:00'
|| pins.event_end != '0000-00-00 00:00:00')
&& (event_type = 'multiple'
|| event_type = 'single'),
1,
0) AS `pin_show_event`,
IF(pins.event_type = 'multiple',
CONCAT(DATE_FORMAT(pins.event_start, '%Y-%m-%d'),
' - ',
DATE_FORMAT(pins.event_end, '%Y-%m-%d')),
DATE_FORMAT(pins.event_start, '%Y-%m-%d')) AS `pin_event_text`,
(SELECT
COUNT(like_id)
FROM
`pins_likes`
WHERE
(pin_id = pins.pin_id)
AND (user_id = 57610)
LIMIT 1) AS `pin_is_liked`,
(SELECT
COUNT(pr_id)
FROM
`pins_reports`
WHERE
(pin_id = pins.pin_id) AND (checked = 0)
AND (user_id = '57610'
OR user_ip = '44021cb8')
LIMIT 1) AS `pin_is_reported`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`pins_images`
WHERE
(pin_id = pins.pin_id) AND (size = '_A')
LIMIT 1) AS `pin_thumb_a`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`pins_images`
WHERE
(pin_id = pins.pin_id) AND (size = '_B')
LIMIT 1) AS `pin_thumb_b`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`pins_images`
WHERE
(pin_id = pins.pin_id) AND (size = '_C')
LIMIT 1) AS `pin_thumb_c`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`pins_images`
WHERE
(pin_id = pins.pin_id) AND (size = '_D')
LIMIT 1) AS `pin_thumb_d`,
`users`.`user_id` AS `user_user_id`,
`users`.`source` AS `user_source`,
`users`.`source_uuid` AS `user_source_uuid`,
`users`.`is_business` AS `user_is_business`,
`users`.`username` AS `user_username`,
`users`.`password` AS `user_password`,
`users`.`email` AS `user_email`,
`users`.`new_email` AS `user_new_email`,
`users`.`new_email_key` AS `user_new_email_key`,
`users`.`business_name` AS `user_business_name`,
`users`.`business_phone` AS `user_business_phone`,
`users`.`firstname` AS `user_firstname`,
`users`.`lastname` AS `user_lastname`,
`users`.`groups` AS `user_groups`,
`users`.`status` AS `user_status`,
`users`.`last_action_datetime` AS `user_last_action_datetime`,
`users`.`last_login` AS `user_last_login`,
`users`.`ip_address` AS `user_ip_address`,
`users`.`following` AS `user_following`,
`users`.`followers` AS `user_followers`,
`users`.`avatar` AS `user_avatar`,
`users`.`facebook_id` AS `user_facebook_id`,
`users`.`twitter_id` AS `user_twitter_id`,
`users`.`twitter_username` AS `user_twitter_username`,
`users`.`is_admin` AS `user_is_admin`,
`users`.`is_developer` AS `user_is_developer`,
`users`.`gender` AS `user_gender`,
`users`.`location` AS `user_location`,
`users`.`latitude` AS `user_latitude`,
`users`.`longitude` AS `user_longitude`,
`users`.`website` AS `user_website`,
`users`.`date_added` AS `user_date_added`,
`users`.`new_password` AS `user_new_password`,
`users`.`new_password_key` AS `user_new_password_key`,
`users`.`facebook_session` AS `user_facebook_session`,
`users`.`boards` AS `user_boards`,
`users`.`pins` AS `user_pins`,
`users`.`likes` AS `user_likes`,
`users`.`latest_pins` AS `user_latest_pins`,
`users`.`description` AS `user_description`,
`users`.`facebook_connect` AS `user_facebook_connect`,
`users`.`facebook_timeline` AS `user_facebook_timeline`,
`users`.`twitter_connect` AS `user_twitter_connect`,
`users`.`dont_search_index` AS `user_dont_search_index`,
`users`.`delete_account` AS `user_delete_account`,
`users`.`delete_account_date` AS `user_delete_account_date`,
`users`.`groups_pin_email` AS `user_groups_pin_email`,
`users`.`comments_email` AS `user_comments_email`,
`users`.`likes_email` AS `user_likes_email`,
`users`.`repins_email` AS `user_repins_email`,
`users`.`follows_email` AS `user_follows_email`,
`users`.`email_interval` AS `user_email_interval`,
`users`.`digest_email` AS `user_digest_email`,
`users`.`news_email` AS `user_news_email`,
`users`.`first_login` AS `user_first_login`,
`users`.`store` AS `user_store`,
`users`.`width` AS `user_width`,
`users`.`height` AS `user_height`,
`users`.`instagram_connect` AS `user_instagram_connect`,
`users`.`instagram_profile_id` AS `user_instagram_profile_id`,
`users`.`instagram_token` AS `user_instagram_token`,
`users`.`enable_follow` AS `user_enable_follow`,
`users`.`public` AS `user_public`,
users.username AS `user_fullname`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`users_avatars`
WHERE
(user_id = users.user_id)
AND (size = '_A')
LIMIT 1) AS `user_avatar_a`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`users_avatars`
WHERE
(user_id = users.user_id)
AND (size = '_B')
LIMIT 1) AS `user_avatar_b`,
(SELECT
CONCAT_WS('|||',
image,
width,
height,
original,
mime)
FROM
`users_avatars`
WHERE
(user_id = users.user_id)
AND (size = '_C')
LIMIT 1) AS `user_avatar_c`,
`boards`.`board_id` AS `board_board_id`,
`boards`.`category_id` AS `board_category_id`,
`boards`.`user_id` AS `board_user_id`,
`boards`.`image` AS `board_image`,
`boards`.`date_added` AS `board_date_added`,
`boards`.`date_modified` AS `board_date_modified`,
`boards`.`sort_order` AS `board_sort_order`,
`boards`.`title` AS `board_title`,
`boards`.`description` AS `board_description`,
`boards`.`followers` AS `board_followers`,
`boards`.`pins` AS `board_pins`,
`boards`.`public` AS `board_public`,
`boards`.`latest_pins` AS `board_latest_pins`,
`boards`.`cover` AS `board_cover`,
`boards`.`delete_request` AS `board_delete_request`,
`boards`.`views` AS `board_views`,
`boards`.`total_views` AS `board_total_views`,
`via`.`user_id` AS `via_user_id`,
`via`.`source` AS `via_source`,
`via`.`source_uuid` AS `via_source_uuid`,
`via`.`is_business` AS `via_is_business`,
`via`.`username` AS `via_username`,
`via`.`password` AS `via_password`,
`via`.`email` AS `via_email`,
`via`.`new_email` AS `via_new_email`,
`via`.`new_email_key` AS `via_new_email_key`,
`via`.`business_name` AS `via_business_name`,
`via`.`business_phone` AS `via_business_phone`,
`via`.`firstname` AS `via_firstname`,
`via`.`lastname` AS `via_lastname`,
`via`.`groups` AS `via_groups`,
`via`.`status` AS `via_status`,
`via`.`last_action_datetime` AS `via_last_action_datetime`,
`via`.`last_login` AS `via_last_login`,
`via`.`ip_address` AS `via_ip_address`,
`via`.`following` AS `via_following`,
`via`.`followers` AS `via_followers`,
`via`.`avatar` AS `via_avatar`,
`via`.`facebook_id` AS `via_facebook_id`,
`via`.`twitter_id` AS `via_twitter_id`,
`via`.`twitter_username` AS `via_twitter_username`,
`via`.`is_admin` AS `via_is_admin`,
`via`.`is_developer` AS `via_is_developer`,
`via`.`gender` AS `via_gender`,
`via`.`location` AS `via_location`,
`via`.`latitude` AS `via_latitude`,
`via`.`longitude` AS `via_longitude`,
`via`.`website` AS `via_website`,
`via`.`date_added` AS `via_date_added`,
`via`.`new_password` AS `via_new_password`,
`via`.`new_password_key` AS `via_new_password_key`,
`via`.`facebook_session` AS `via_facebook_session`,
`via`.`boards` AS `via_boards`,
`via`.`pins` AS `via_pins`,
`via`.`likes` AS `via_likes`,
`via`.`latest_pins` AS `via_latest_pins`,
`via`.`description` AS `via_description`,
`via`.`facebook_connect` AS `via_facebook_connect`,
`via`.`facebook_timeline` AS `via_facebook_timeline`,
`via`.`twitter_connect` AS `via_twitter_connect`,
`via`.`dont_search_index` AS `via_dont_search_index`,
`via`.`delete_account` AS `via_delete_account`,
`via`.`delete_account_date` AS `via_delete_account_date`,
`via`.`groups_pin_email` AS `via_groups_pin_email`,
`via`.`comments_email` AS `via_comments_email`,
`via`.`likes_email` AS `via_likes_email`,
`via`.`repins_email` AS `via_repins_email`,
`via`.`follows_email` AS `via_follows_email`,
`via`.`email_interval` AS `via_email_interval`,
`via`.`digest_email` AS `via_digest_email`,
`via`.`news_email` AS `via_news_email`,
`via`.`first_login` AS `via_first_login`,
`via`.`store` AS `via_store`,
`via`.`width` AS `via_width`,
`via`.`height` AS `via_height`,
`via`.`instagram_connect` AS `via_instagram_connect`,
`via`.`instagram_profile_id` AS `via_instagram_profile_id`,
`via`.`instagram_token` AS `via_instagram_token`,
`via`.`enable_follow` AS `via_enable_follow`,
`via`.`public` AS `via_public`,
via.username AS `via_fullname`,
`row` . *,
@curRow:=@curRow + 1 AS `pin_pin_row`
FROM
`pins`
LEFT JOIN
`users` ON pins.user_id = users.user_id
LEFT JOIN
`boards` ON pins.board_id = boards.board_id
LEFT JOIN
`users` AS `via` ON pins.via = via.user_id
INNER JOIN
(SELECT @curRow:=- 1) AS `row`
WHERE
(pins.user_id = 57610
OR IF(pins.user_id = 57610, 1, pins.public) = 1)
AND (pins.latitude <= 40.78666477795
AND pins.latitude >= 40.64193322205
AND pins.longitude <= - 73.910342436525
AND pins.longitude >= - 74.101288363475)
AND (pins.user_id = 57610
OR pins.user_id IN (SELECT
user_id
FROM
users
WHERE
public = 1))
ORDER BY pins.vip DESC , pins.pin_id DESC
LIMIT 30
答案 0 :(得分：1)
您可以尝试更改此条件：
AND (pins.user_id = 57610
OR pins.user_id IN (SELECT
user_id
FROM
users
WHERE
public = 1))
人：
AND (pins.user_id = 57610
OR users.public = 1))
答案 1 :(得分：0)
我让两位分析师检查了sql以及使用的表。 sql和所有索引的结果非常好。两张表中有近3亿条记录。主（引脚）表现在有2800万条记录。
所以，我的解决方案是实施sphinx search。我在数据库的专用服务器上，因此可以轻松安装和调整。一旦我设置了sphinx，创建了我的初始索引，然后设置了一个预定的差异索引＆amp;合并过程中，我在很多情况下能够在100毫秒甚至60毫秒内获得结果。