4

我一直在开发一个版本为 2.5.11 的 Joomla 站点。该站点的流量将非常高。

我的问题是关于 MySQL 查询性能。数据库在内容表中包含大约 60000 行,下面看到的查询(核心 com_content 文章模型查询)执行时间约为 6 秒。非常慢。

SELECT a.id,
   a.title,
   a.alias,
   a.title_alias,
   a.introtext,
   a.checked_out,
   a.checked_out_time,
   a.catid,
   a.created,
   a.created_by,
   a.created_by_alias,
   CASE
     WHEN a.modified = 0 THEN a.created
     ELSE a.modified
   END as modified,
   a.modified_by,
   uam.name as modified_by_name,
   CASE
     WHEN a.publish_up = 0 THEN a.created
     ELSE a.publish_up
   END as publish_up,
   a.publish_down,
   a.images,
   a.urls,
   a.attribs,
   a.metadata,
   a.metakey,
   a.metadesc,
   a.access,
   a.hits,
   a.xreference,
   a.featured,
   LENGTH(a.fulltext) AS readmore,
   CASE
     WHEN badcats.id is not null THEN 0
     ELSE a.state
   END AS state,
   c.title AS category_title,
   c.path AS category_route,
   c.access AS category_access,
   c.alias AS category_alias,
   CASE
     WHEN a.created_by_alias > ' ' THEN a.created_by_alias
     ELSE ua.name
   END AS author,
   ua.email AS author_email,
   contact.id as contactid,
   parent.title as parent_title,
   parent.id as parent_id,
   parent.path as parent_route,
   parent.alias as parent_alias,
   ROUND(v.rating_sum / v.rating_count, 0) AS rating,
   v.rating_count as rating_count,
   c.published,
   CASE
     WHEN badcats.id is null THEN c.published
     ELSE 0
   END AS parents_published
  FROM #__content AS a
 LEFT JOIN #__content_frontpage AS fp ON fp.content_id = a.id
 LEFT JOIN #__categories AS c ON c.id = a.catid
 LEFT JOIN #__users AS ua ON ua.id = a.created_by
 LEFT JOIN #__users AS uam ON uam.id = a.modified_by
 LEFT JOIN 
 (SELECT contact.user_id, MAX(contact.id) AS id, contact.language FROM
  #__contact_details AS contact WHERE contact.published = 1 GROUP BY
   contact.user_id, contact.language) AS contact ON contact.user_id =
    a.created_by
 LEFT JOIN #__categories as parent ON parent.id = c.parent_id
 LEFT JOIN #__content_rating AS v ON a.id = v.content_id
 LEFT OUTER JOIN 
 (SELECT cat.id as id FROM #__categories AS cat JOIN #__categories AS
  parent ON cat.lft BETWEEN parent.lft AND parent.rgt WHERE parent.extension
   = 'com_content' AND parent.published != 1 GROUP BY cat.id) AS badcats ON
    badcats.id = c.id
  WHERE CASE
  WHEN badcats.id is null THEN a.state
  ELSE 0
 END = 1 AND
  a.featured = 0 AND
  a.id NOT IN (8921, 33722, 33728, 33729, 34187, 35047, 36784, 36236, 33724,
   19522) AND
  a.catid IN (8, 39, 40, 38, 72, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
   20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 36, 37, 41) AND
  (a.publish_up = '0000-00-00 00:00:00' OR
  a.publish_up <= '2013-06-12 06:44:44') AND
  (a.publish_down = '0000-00-00 00:00:00' OR
  a.publish_down >= '2013-06-12 06:44:44')
    GROUP BY a.id,
     a.title,
     a.alias,
     a.title_alias,
     a.introtext,
     a.checked_out,
     a.checked_out_time,
     a.catid,
     a.created,
     a.created_by,
     a.created_by_alias,
     a.created,
     a.modified,
     a.modified_by,
     uam.name,
     a.publish_up,
     a.attribs,
     a.metadata,
     a.metakey,
     a.metadesc,
     a.access,
     a.hits,
     a.xreference,
     a.featured,
     a.fulltext,
     a.state,
     a.publish_down,
     badcats.id,
     c.title,
     c.path,
     c.access,
     c.alias,
     uam.id,
     ua.name,
     ua.email,
     contact.id,
     parent.title,
     parent.id,
     parent.path,
     parent.alias,
     v.rating_sum,
     v.rating_count,
     c.published,
     c.lft,
     a.ordering,
     parent.lft,
     fp.ordering,
     c.id,
     a.images,
     a.urls
 ORDER BY publish_up DESC
 LIMIT 4,  4

例如当我改变

 (a.publish_up = '0000-00-00 00:00:00' OR
  a.publish_up <= '2013-06-12 06:44:44') AND
  (a.publish_down = '0000-00-00 00:00:00' OR
  a.publish_down >= '2013-06-12 06:44:44')

(a.publish_up >= DATE_SUB(NOW(), INTERVAL 2 MONTH))

查询执行时间约为 ~0.5 秒

查询说明:

在此处输入图像描述

不破解核心代码我能做什么,我永远不想破解核心代码......

MySQL服务器配置:

skip-external-locking
skip-name-resolve
key_buffer_size = 128M
max_connections=1024
max_allowed_packet = 16M
net_buffer_length = 8K
table_open_cache = 512
table_cache = 2048
join_buffer_size = 128M
read_buffer_size = 128M
sort_buffer_size = 128M
read_rnd_buffer_size=2M
myisam_sort_buffer_size=128M
thread_concurrency=16
query_cache_size = 128M  
query_cache_type=1
thread_cache_size = 1300
query_cache_limit=128M
max_heap_table_size = 1024M
tmp_table_size = 1024M
4

3 回答 3

4

我发现改进这个非常慢的查询的最好方法是删除对归档的引用(这破坏了归档状态的全部目的,即从列表结果中删除这些项目)。只需获取已发布(或已发布 + 未发布)的项目,因此 eiher state=1 或 state IN(0,1) 您可以查看可能使用http://extensions.joomla.org/extensions/style-a-design/template /15611

切换到 InnoDB 可能也会有所帮助,但拥有最新版本的 MySQL 是获得改进的最大方法。

关于“你现在能做什么”我很想看到你的拉取请求,为该查询添加日期限制。我们最近在 JDatabaseQuery 中添加了数据数学,所以我做起来相对简单,而且会是一个非常好的改进。还有一些其他悬而未决的问题会对性能产生重大影响,如果您对它们进行测试和评论,它们将进入下一个版本的代码库。例如:https ://github.com/joomla/joomla-cms/pull/1274

于 2013-06-13T19:34:32.690 回答
0

由于您必须破解核心才能对查询进行更改,因此最好的办法是向 com_content 添加一个新模型/视图。添加的内容不会被核心更新所覆盖。

于 2013-06-13T13:19:49.127 回答
0

您真的必须对所有这些列进行 GROUP BY 吗?一个简单的 GROUP BY a.id 就足够了吗?

如果这不起作用,请尝试增加变量 tmp_table_size 和 max_heap_table_size 的值。这可能就是您在“额外列”中获得“使用临时”的原因。此消息是因为 MySQL 在进行排序时用完了物理内存,需要先将数据写入磁盘,这样比较慢。

找出这些值是什么:
SELECT @@tmp_table_size, @@max_heap_table_size;

然后将它们设置为更高的值,直到“使用临时”消失。(希望你有足够的内存使用):
SET tmp_table_size = ###, @@max_heap_table_size = ###;

此页面上有更多信息:http: //dev.mysql.com/doc/refman/5.1/en/internal-temporary-tables.html

如果上述方法不起作用,您可以尝试将每个派生查询放入一个临时表中,该表在您要加入的每个列上都有索引,并加入每个新的临时表。但这会使它变慢,特别是如果您必须在每次请求期间都这样做。否则,如果数据没有更新,那么它可以工作。

PS我刚刚意识到你的评论“我能做什么而不破解核心代码”。希望会话变量有效。

于 2013-06-12T20:45:58.770 回答