MySQL · 特性分析 · 内部临时表
lirain115
8年前
<h2>MySQL中的两种临时表</h2> <h3>外部临时表</h3> <p>通过CREATE TEMPORARY TABLE 创建的临时表,这种临时表称为外部临时表。这种临时表只对当前用户可见,当前会话结束的时候,该临时表会自动关闭。这种临时表的命名与非临时表可以同名(同名后非临时表将对当前会话不可见,直到临时表被删除)。</p> <h3>内部临时表</h3> <p>内部临时表是一种特殊轻量级的临时表,用来进行性能优化。这种临时表会被MySQL自动创建并用来存储某些操作的中间结果。这些操作可能包括在优化阶段或者执行阶段。这种内部表对用户来说是不可见的,但是通过EXPLAIN或者SHOW STATUS可以查看MYSQL是否使用了内部临时表用来帮助完成某个操作。内部临时表在SQL语句的优化过程中扮演着非常重要的角色, MySQL中的很多操作都要依赖于内部临时表来进行优化。但是使用内部临时表需要创建表以及中间数据的存取代价,所以用户在写SQL语句的时候应该尽量的去避免使用临时表。</p> <p>内部临时表有两种类型:一种是HEAP临时表,这种临时表的所有数据都会存在内存中,对于这种表的操作不需要IO操作。另一种是OnDisk临时表,顾名思义,这种临时表会将数据存储在磁盘上。OnDisk临时表用来处理中间结果比较大的操作。如果HEAP临时表存储的数据大于MAX_HEAP_TABLE_SIZE(详情请参考MySQL手册中系统变量部分),HEAP临时表将会被自动转换成OnDisk临时表。OnDisk临时表在5.7中可以通过INTERNAL_TMP_DISK_STORAGE_ENGINE系统变量选择使用MyISAM引擎或者InnoDB引擎。</p> <p>本篇文章主要介绍哪些操作可能会利用到内部临时表。如果用户在书写SQL语句的时候能够尽量少的使用内部临时表进行查询优化,将有效的提高查询执行的效率。</p> <p>首先我们定义一个表t1,</p> <pre> <code class="language-sql">CREATE TABLE t1( a int, b int); INSERT INTO t1 VALUES(1,2),(3,4);</code></pre> <p>下面所有的操作都是基于表t1进行举例的。</p> <ul> <li>在SQL语句中使用SQL_BUFFER_RESULT hint</li> </ul> <p>SQL_BUFFER_RESULT主要用来让MySQL尽早的释放表上的锁。因为如果数据量很大的话,需要较长时间将数据发送到客户端,通过将数据缓冲到临时表中可以有效的减少读锁对表的占用时间。例如:</p> <pre> <code class="language-sql"> mysql> explain format=json select SQL_BUFFER_RESULT * from t1; EXPLAIN { "query_block": { "select_id": 1, "cost_info": { "query_cost": "2.00" }, "buffer_result": { "using_temporary_table": true, "table": { "table_name": "t1", "access_type": "ALL", ...</code></pre> <ul> <li>如果SQL语句中包含了DERIVED_TABLE。</li> </ul> <p>在5.7中,由于采用了新的优化方式,我们需要使用 set optimizer_switch=’derived_merge=off’来禁止derived table合并到外层的Query中。例如:</p> <pre> <code class="language-sql">mysql> explain format=json select * from (select * from t1) as tt; EXPLAIN { "query_block": { "select_id": 1, "cost_info": { "query_cost": "2.40" }, "table": { "table_name": "tt", "access_type": "ALL", ... "materialized_from_subquery": { "using_temporary_table": true, ...</code></pre> <ul> <li>如果我们查询系统表的话,系统表的数据将被存储到内部临时表中。</li> </ul> <p>我们当前不能使用EXPLAIN来查看是否读取系统表数据需要利用到内部临时表,但是可以通过SHOW STATUS来查看是否利用到了内部临时表。例如:</p> <pre> <code class="language-sql">mysql> select * from information_schema.character_sets; mysql> show status like 'CREATE%';</code></pre> <ul> <li>如果DISTINCT语句没有被优化掉,即DISTINCT语句被优化转换为GROUP BY操作或者利用UNIQUE INDEX消除DISTINCT, 内部临时表将会被使用。</li> <li> <pre> <code class="language-sql">mysql> explain format=json select distinct a from t1; EXPLAIN { { "query_block": { "select_id": 1, "cost_info": { "query_cost": "1.60" }, "duplicates_removal": { "using_temporary_table": true, ...</code></pre> </li> </ul> <ul> <li>如果查询带有ORDER BY语句,并且不能被优化掉。下面几种情况会利用到内部临时表缓存中间数据,然后对中间数据进行排序。</li> </ul> <p>1)如果连接表使用BNL(Batched Nestloop)/BKA(Batched Key Access)例如:</p> <p>1))BNL默认是打开的</p> <pre> <code class="language-sql">mysql> explain format=json select * from t1, t1 as t2 order by t1.a; EXPLAIN { "query_block": { "select_id": 1, "cost_info": { "query_cost": "22.00" }, "ordering_operation": { "using_temporary_table": true, ...</code></pre> <p>2))关掉BNL后,ORDER BY将直接使用filesort。</p> <pre> <code class="language-sql">mysql> set optimizer_switch='block_nested_loop=off'; Query OK, 0 rows affected (0.00 sec) mysql> explain format=json select * from t1, t1 as t2 order by t1.a; EXPLAIN { "query_block": { "select_id": 1, "cost_info": { "query_cost": "25.00" }, "ordering_operation": { "using_filesort": true, ...</code></pre> <p>2)ORDER BY的列不属于执行计划中第一个连接表的列。例如:</p> <pre> <code class="language-sql">mysql> explain format=json select * from t as t1, t as t2 order by t2.a; EXPLAIN { "query_block": { "select_id": 1, "cost_info": { "query_cost": "25.00" }, "ordering_operation": { "using_temporary_table": true, ...</code></pre> <p>3)如果ORDER BY的表达式是个复杂表达式。</p> <p>那么什么样的ORDER BY表达式,MySQL认为是复杂表达式呢?</p> <p>1))如果排序表达式是SP或者UDF。例如:</p> <pre> <code class="language-sql">drop function if exists func1; delimiter | create function func1(x int) returns int deterministic begin declare z1, z2 int; set z1 = x; set z2 = z1+2; return z2; end| delimiter ; explain format=json select * from t1 order by func1(a); { "query_block": { "select_id": 1, "cost_info": { "query_cost": "2.20" }, "ordering_operation": { "using_temporary_table": true, ...</code></pre> <p>2))ORDER BY的列包含聚集函数</p> <p>为了简化执行计划,我们利用INDEX来优化GROUP BY语句。</p> <p>例如:</p> <pre> <code class="language-sql"> create index idx1 on t1(a); explain format=json SELECt a FROM t1 group by a order by sum(a); | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "1.20" }, "ordering_operation": { "using_temporary_table": true, "using_filesort": true, "grouping_operation": { "using_filesort": false, ... drop index idx1 on t1;</code></pre> <p>3))ORDER BY的列中包含有SCALAR SUBQUERY,当然该SCALAR SUBQUERY没有被优化掉。例如:</p> <pre> <code class="language-sql">explain format=json select (select rand() from t1 limit 1) as a from t1 order by a; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "1.20" }, "ordering_operation": { "using_temporary_table": true, "using_filesort": true, ...</code></pre> <p>4) 如果查询既带有ORDER BY同时也有GROUP BY语句,但是两个语句使用的列不相同。</p> <p>注意: 如果是5.7,我们需要将sql_mode设置为非only_full_group_by模式,否则会报错。</p> <p>同样为了简化执行计划,我们利用INDEX来优化GROUP BY语句。</p> <p>例如:</p> <pre> <code class="language-sql">set sql_mode=''; create index idx1 on t1(b); explain format=json select t1.a from t1 group by t1.b order by 1; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "1.40" }, "ordering_operation": { "using_temporary_table": true, "using_filesort": true, "grouping_operation": { "using_filesort": false, ... drop index idx1 on t1;</code></pre> <ul> <li>如果查询带有GROUP BY语句,并且不能被优化掉。下面几种情况会利用到内部临时表缓存中间数据,然后对中间数据进行GROUP BY。</li> </ul> <p>1)如果连接表使用BNL(Batched Nestloop)/BKA(Batched Key Access)。例如:</p> <pre> <code class="language-sql">explain format=json select t2.a from t1, t1 as t2 group by t1.a; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "8.20" }, "grouping_operation": { "using_temporary_table": true, "using_filesort": true, "cost_info": { "sort_cost": "4.00" ...</code></pre> <p>2) 如果GROUP BY的列不属于执行计划中的第一个连接表。例如:</p> <pre> <code class="language-sql">explain format=json select t2.a from t1, t1 as t2 group by t2.a; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "8.20" }, "grouping_operation": { "using_temporary_table": true, "using_filesort": true, "nested_loop": [ ...</code></pre> <p>3) 如果GROUP BY语句使用的列与ORDER BY语句使用的列不同。例如:</p> <pre> <code class="language-sql">set sql_mode=''; explain format=json select t1.a from t1 group by t1.b order by t1.a; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "1.40" }, "ordering_operation": { "using_filesort": true, "grouping_operation": { "using_temporary_table": true, "using_filesort": false, ...</code></pre> <p>4) 如果GROUP BY带有ROLLUP并且是基于多表外连接。例如:</p> <pre> <code class="language-sql">explain format=json select sum(t1.a) from t1 left join t1 as t2 on true group by t1.a with rollup; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "7.20" }, "grouping_operation": { "using_temporary_table": true, "using_filesort": true, "cost_info": { "sort_cost": "4.00" }, ...</code></pre> <p>5) 如果GROUP BY语句使用的列来自于SCALAR SUBQUERY,并且没有被优化掉。例如:</p> <pre> <code class="language-sql">explain format=json select (select avg(a) from t1) as a from t1 group by a; | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "3.40" }, "grouping_operation": { "using_temporary_table": true, "using_filesort": true, "cost_info": { "sort_cost": "2.00" }, ...</code></pre> <ul> <li>IN表达式转换为semi-join进行优化<br> 1) 如果semi-join执行方式为Materialization<br> 例如:</li> <li> <pre> <code class="language-sql">set optimizer_switch='firstmatch=off,duplicateweedout=off'; explain format=json select * from t1 where a in (select b from t1); | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "5.60" }, "nested_loop": [ { "rows_examined_per_scan": 1, "materialized_from_subquery": { "using_temporary_table": true, "query_block": { "table": { "table_name": "t1", "access_type": "ALL", ... </code></pre> </li> </ul> <p>2) 如果semi-join执行方式为Duplicate Weedout例如:</p> <pre> <code class="language-sql">set optimizer_switch='firstmatch=off'; explain format=json select * from t1 where a in (select b from t1); | { "query_block": { "select_id": 1, "cost_info": { "query_cost": "4.80" }, "duplicates_removal": { "using_temporary_table": true, "nested_loop": [ { ...</code></pre> <ul> <li>如果查询语句带有UNION,MySQL将利用内部临时表帮助UNION操作消除重复。<br> 例如:</li> <li> <pre> <code class="language-sql">explain format=json select * from t1 union select * from t1; | { "query_block": { "union_result": { "using_temporary_table": true, "table_name": "<union1,2>", ...</code></pre> </li> </ul> <ul> <li>如果查询语句使用多表更新。<br> 这里Explain不能看到内部临时表被利用,所以需要查看status。<br> 例如:</li> <li> <pre> <code class="language-sql">update t1, t1 as t2 set t1.a=3; show status like 'CREATE%';</code></pre> </li> </ul> <ul> <li>如果聚集函数中包含如下函数,内部临时表也会被利用。</li> <li> <pre> <code class="language-sql">1) count(distinct *) 例如: explain format=json select count(distinct a) from t1; 2) group_concat 例如: explain format=json select group_concat(b) from t1;</code></pre> </li> </ul> <p>总之,上面列出了10种情况,MySQL将利用内部临时表进行中间结果缓存,如果数据量比较大的话,内部临时表将会把数据存储在磁盘上,这样显然会对性能有所影响。为了尽可能的减少性能损失,我们需要尽量避免上述情况的出现。</p> <p> </p> <p>来自: <a href="/misc/goto?guid=4959674841680323273" rel="nofollow">https://yq.aliyun.com/articles/55867</a></p>