技术分享-InnoDB-Handlerread-变量解释

24次阅读

共计 15292 个字符，预计需要花费 39 分钟才能阅读完成。

作者：高鹏
文章末尾有他著作的《深入理解 MySQL 主从原理 32 讲》，深入透彻理解 MySQL 主从，GTID 相关技术知识。
源码版本：percona 5.7.14
本文为学习记录，可能有误请谅解。

本文建议 PC 端观看，效果更佳。

一、Handler_read_* 值的实质

内部表示如下：

{"Handler_read_first",       (char*) offsetof(STATUS_VAR, ha_read_first_count),     SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},
{"Handler_read_key",         (char*) offsetof(STATUS_VAR, ha_read_key_count),       SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},
{"Handler_read_last",        (char*) offsetof(STATUS_VAR, ha_read_last_count),      SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},
{"Handler_read_next",        (char*) offsetof(STATUS_VAR, ha_read_next_count),      SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},
{"Handler_read_prev",        (char*) offsetof(STATUS_VAR, ha_read_prev_count),      SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},
{"Handler_read_rnd",         (char*) offsetof(STATUS_VAR, ha_read_rnd_count),       SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},
{"Handler_read_rnd_next",    (char*) offsetof(STATUS_VAR, ha_read_rnd_next_count),  SHOW_LONGLONG_STATUS,    SHOW_SCOPE_ALL},

实际上这些变量都是 MySQL 层定义出来的，因为 MySQL 可以包含多个存储引擎。因此这些值如何增加需要在引擎层的接口中自行实现，也就是说各个引擎都有自己的实现，在 MySQL 层进行汇总，因此这些值不是某个引擎特有的，打个比方如果有 Innodb 和 MyISAM 引擎，那么这些值是两个引擎的总和。本文将以 Innodb 为主要学习对象进行解释。

二、各个值的解释

1. Handler_read_key

  • 内部表示：ha_read_key_count
  • Innodb 更改接口：ha_innobase::index_read
  • 文档解释：The number of requests to read a row based on a key. If this value is high, it is a good indication that your tables are properly indexed for your queries.
  • 源码函数解释：Positions an index cursor to the index specified in the handle. Fetches the row if any.
  • 作者解释：这个函数是访问索引的时候定位到值所在的位置用到的函数，因为必须要知道读取索引的开始位置才能向下访问。

2. Handler_read_next

  • 内部表示：ha_read_next_count
  • Innodb 更改接口：ha_innobase::index_next_same ha_innobase::index_next
  • 文档解释：The number of requests to read the next row in key order. This value is incremented if you are

querying an index column with a range constraint or if you are doing an index scan.

  • 源码函数解释：

index_next – Reads the next row from a cursor, which must have previously been positioned using index_read.
index_next_same – Reads the next row matching to the key value given as the parameter.

  • 作者解释：访问索引的下一条数据封装的 ha_innobase::general_fetch 函数，index_next_same 和 index_next 不同在于访问的方式不一样，比如范围 range 查询需要用到和索引全扫描也会用到 index_next，而 ref 访问方式会使用 index_next_same

3. Handler_read_first

  • 内部表示：ha_read_first_count
  • Innodb 更改接口：ha_innobase::index_first
  • 文档解释：The number of times the first entry in an index was read. If this value is high, it suggests that the

server is doing a lot of full index scans; for example, SELECT col1 FROM foo, assuming that col1
is indexed

  • 源码函数解释：Positions a cursor on the first record in an index and reads the corresponding row to buf.
  • 作者解释：定位索引的第一条数据，实际上也是封装的 ha_innobase::index_read 函数(如全表扫描 / 全索引扫描调用)

4. Handler_read_rnd_next

  • 内部表示：ha_read_rnd_next_count
  • Innodb 更改接口：ha_innobase::rnd_next
  • 文档解释：The number of requests to read the next row in the data file. This value is high if you are doing a lot of table scans. Generally this suggests that your tables are not properly indexed or that your queries

are not written to take advantage of the indexes you have.

  • 源码函数解释：Reads the next row in a table scan (also used to read the FIRST row in a table scan).
  • 作者解释：全表扫描访问下一条数据，实际上也是封装的 ha_innobase::general_fetch，在访问之前会调用 ha_innobase::index_first

5. Handler_read_rnd

  • 内部表示：ha_read_rnd_count
  • Innodb 更改接口：ha_innobase::rnd_pos
  • Memory 更改接口：ha_heap::rnd_pos
  • 文档解释：The number of requests to read a row based on a fixed position. This value is high if you are doing a lot of queries that require sorting of the result. You probably have a lot of queries that require MySQL to scan entire tables or you have joins that do not use keys properly.
  • 作者解释：这个状态值在我测试期间只发现对临时表做排序的时候会用到，而且是 Memory 引擎的，具体只能按照文档理解了。

6. 其他

最后 2 个简单说一下

  • Handler_read_prev

Innodb 接口为 ha_innobase::index_prev 访问索引的上一条数据，实际上也是封装的 ha_innobase::general_fetch 函数，用于 ORDER BY DESC 索引扫描避免排序，内部状态值 ha_read_prev_count 增加。

  • Handler_read_last

Innodb 接口为 ha_innobase::index_last 访问索引的最后一条数据作为定位，实际上也是封装的 ha_innobase::index_read 函数，用于 ORDER BY DESC 索引扫描避免排序，内部状态值 ha_read_last_count 增加。

三、常用查询测试

1. 测试用例

mysql> show create table z1;
+-------+-------------------------------------------------------------------------------------------------------------------------------------------+
| Table | Create Table                                                                                                                              |
+-------+-------------------------------------------------------------------------------------------------------------------------------------------+
| z1    | CREATE TABLE `z1` (`a` int(11) DEFAULT NULL,
  `name` varchar(20) DEFAULT NULL,
  KEY `a` (`a`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 |
+-------+-------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.00 sec)

mysql> show create table z10;
+-------+------------------------------------------------------------------------------------------------------------------------------------------------+
| Table | Create Table                                                                                                                                   |
+-------+------------------------------------------------------------------------------------------------------------------------------------------------+
| z10   | CREATE TABLE `z10` (`a` int(11) DEFAULT NULL,
  `name` varchar(20) DEFAULT NULL,
  KEY `a_idx` (`a`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 |
+-------+------------------------------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.00 sec)
mysql> select count(*) from z1;
+----------+
| count(*) |
+----------+
|    56415 |
+----------+
1 row in set (5.27 sec)

mysql> select count(*) from z10;
+----------+
| count(*) |
+----------+
|       10 |
+----------+
1 row in set (0.00 sec)

2. 全表扫描

mysql> desc select * from z1;
+----+-------------+-------+------------+------+---------------+------+---------+------+-------+----------+-------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref  | rows  | filtered | Extra |
+----+-------------+-------+------------+------+---------------+------+---------+------+-------+----------+-------+
|  1 | SIMPLE      | z1    | NULL       | ALL  | NULL          | NULL | NULL    | NULL | 56650 |   100.00 | NULL  |
+----+-------------+-------+------------+------+---------------+------+---------+------+-------+----------+-------+
1 row in set, 1 warning (0.00 sec)

mysql> pager cat >>/dev/null
PAGER set to 'cat >>/dev/null'
mysql> flush status;
Query OK, 0 rows affected (0.10 sec)
mysql> select * from z1;
56415 rows in set (4.05 sec)
mysql> pager;
Default pager wasn't set, using stdout.
mysql> show status like 'Handler_read%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| Handler_read_first    | 1     |
| Handler_read_key      | 1     |
| Handler_read_last     | 0     |
| Handler_read_next     | 0     |
| Handler_read_prev     | 0     |
| Handler_read_rnd      | 0     |
| Handler_read_rnd_next | 56416 |
+-----------------------+-------+
7 rows in set (0.01 sec)

Handler_read_first 增加 1 次用于初次定位，Handler_read_key 增加 1 次，Handler_read_rnd_next 增加扫描行数。我们前面说过因为 ha_innobase::index_first 也是封装的 ha_innobase::index_read 因此都需要 +1。

3. 全索引扫描

mysql> desc select a from z1;
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-------------+
| id | select_type | table | partitions | type  | possible_keys | key  | key_len | ref  | rows  | filtered | Extra       |
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-------------+
|  1 | SIMPLE      | z1    | NULL       | index | NULL          | a    | 5       | NULL | 56650 |   100.00 | Using index |
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-------------+
1 row in set, 1 warning (0.00 sec)

mysql> flush status;
Query OK, 0 rows affected (0.12 sec)

mysql> pager cat >>/dev/null
PAGER set to 'cat >>/dev/null'
mysql> select a from z1;
56415 rows in set (4.57 sec)

mysql> pager
Default pager wasn't set, using stdout.
mysql> show status like 'Handler_read%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| Handler_read_first    | 1     |
| Handler_read_key      | 1     |
| Handler_read_last     | 0     |
| Handler_read_next     | 56415 |
| Handler_read_prev     | 0     |
| Handler_read_rnd      | 0     |
| Handler_read_rnd_next | 0     |
+-----------------------+-------+
7 rows in set (0.01 sec)

Handler_read_first 增加 1 次用于初次定位，Handler_read_key 增加 1 次，Handler_read_next 增加扫描行数用于连续访问接下来的行。我们前面说过因为 ha_innobase::index_first 也是封装的 ha_innobase::index_read 因此都需要 +1。

4. 索引 ref 访问

我这里因为是测试索引全是等于 10 的加上了 force index

mysql>  desc select  * from z1 force index(a) where a=10;
+----+-------------+-------+------------+------+---------------+------+---------+-------+-------+----------+-------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref   | rows  | filtered | Extra |
+----+-------------+-------+------------+------+---------------+------+---------+-------+-------+----------+-------+
|  1 | SIMPLE      | z1    | NULL       | ref  | a             | a    | 5       | const | 28325 |   100.00 | NULL  |
+----+-------------+-------+------------+------+---------------+------+---------+-------+-------+----------+-------+
1 row in set, 1 warning (0.01 sec)

mysql> flush status;
Query OK, 0 rows affected (0.13 sec)

mysql> pager cat >>/dev/null
PAGER set to 'cat >>/dev/null'
mysql> select  * from z1 force index(a) where a=10;
56414 rows in set (32.39 sec)
mysql> pager
Default pager wasn't set, using stdout.
mysql> show status like 'Handler_read%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| Handler_read_first    | 0     |
| Handler_read_key      | 1     |
| Handler_read_last     | 0     |
| Handler_read_next     | 56414 |
| Handler_read_prev     | 0     |
| Handler_read_rnd      | 0     |
| Handler_read_rnd_next | 0     |
+-----------------------+-------+
7 rows in set (0.06 sec)

Handler_read_key 增加 1 次这是用于初次定位，Handler_read_next 增加扫描行数次数用于接下来的数据访问。

5. 索引 range 访问

mysql> desc select  * from z1 force index(a) where a>9 and a<12;
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-----------------------+
| id | select_type | table | partitions | type  | possible_keys | key  | key_len | ref  | rows  | filtered | Extra                 |
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-----------------------+
|  1 | SIMPLE      | z1    | NULL       | range | a             | a    | 5       | NULL | 28325 |   100.00 | Using index condition |
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-----------------------+
1 row in set, 1 warning (0.00 sec)

mysql>  pager cat >>/dev/null
PAGER set to 'cat >>/dev/null'
mysql> select  * from z1 force index(a) where a>9 and a<12;
56414 rows in set (47.54 sec)
mysql> show status like 'Handler_read%';
7 rows in set (0.03 sec)
mysql>  pager
Default pager wasn't set, using stdout.
mysql> show status like 'Handler_read%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| Handler_read_first    | 0     |
| Handler_read_key      | 1     |
| Handler_read_last     | 0     |
| Handler_read_next     | 56414 |
| Handler_read_prev     | 0     |
| Handler_read_rnd      | 0     |
| Handler_read_rnd_next | 0     |
+-----------------------+-------+
7 rows in set (0.02 sec)

Handler_read_key 增加 1 次这是用于初次定位，Handler_read_next 增加扫描行数次数用于接下来的数据访问。

6. 被驱动表带索引访问

mysql> desc select * from z1 STRAIGHT_JOIN z10 force index(a_idx) on z1.a=z10.a;
+----+-------------+-------+------------+------+---------------+-------+---------+-----------+-------+----------+-------------+
| id | select_type | table | partitions | type | possible_keys | key   | key_len | ref       | rows  | filtered | Extra       |
+----+-------------+-------+------------+------+---------------+-------+---------+-----------+-------+----------+-------------+
|  1 | SIMPLE      | z1    | NULL       | ALL  | a             | NULL  | NULL    | NULL      | 56650 |   100.00 | Using where |
|  1 | SIMPLE      | z10   | NULL       | ref  | a_idx         | a_idx | 5       | test.z1.a |    10 |   100.00 | NULL        |
+----+-------------+-------+------------+------+---------------+-------+---------+-----------+-------+----------+-------------+
2 rows in set, 1 warning (0.01 sec)

mysql> flush status;
Query OK, 0 rows affected (0.47 sec)

mysql> pager cat >> /dev/null
PAGER set to 'cat >> /dev/null'
mysql>  select * from z1 STRAIGHT_JOIN z10 force index(a_idx) on z1.a=z10.a;
112828 rows in set (1 min 21.21 sec)

mysql> pager
Default pager wasn't set, using stdout.
mysql>  show status like 'Handler_read%';
+-----------------------+--------+
| Variable_name         | Value  |
+-----------------------+--------+
| Handler_read_first    | 1      |
| Handler_read_key      | 56416  |
| Handler_read_last     | 0      |
| Handler_read_next     | 112828 |
| Handler_read_prev     | 0      |
| Handler_read_rnd      | 0      |
| Handler_read_rnd_next | 56416  |
+-----------------------+--------+
7 rows in set (0.00 sec)

Handler_read_first 增加一次作为驱动表 z1 全表扫描定位的开始，接下来 Handler_read_rnd_next 扫描全部记录，每次扫描一次在 z10 表通过索引 a_idx 定位一次 Handler_read_key 增加 1 次，然后接下来进行索引 a_idx 进行数据查找 Handler_read_next 增加为扫描的行数。

7. 索引避免排序正向和反向

mysql>  flush status;
Query OK, 0 rows affected (0.05 sec)

mysql> pager cat >> /dev/null
PAGER set to 'cat >> /dev/null'
mysql> select * from z1 force index(a) order by a;
56415 rows in set (27.39 sec)

mysql> pager
Default pager wasn't set, using stdout.
mysql> show status like 'Handler_read%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| Handler_read_first    | 1     |
| Handler_read_key      | 1     |
| Handler_read_last     | 0     |
| Handler_read_next     | 56415 |
| Handler_read_prev     | 0     |
| Handler_read_rnd      | 0     |
| Handler_read_rnd_next | 0     |
+-----------------------+-------+
7 rows in set (0.01 sec)

mysql> flush status;
Query OK, 0 rows affected (0.10 sec)

mysql> desc  select * from z1 force index(a) order by a desc;
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-------+
| id | select_type | table | partitions | type  | possible_keys | key  | key_len | ref  | rows  | filtered | Extra |
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-------+
|  1 | SIMPLE      | z1    | NULL       | index | NULL          | a    | 5       | NULL | 56650 |   100.00 | NULL  |
+----+-------------+-------+------------+-------+---------------+------+---------+------+-------+----------+-------+
1 row in set, 1 warning (0.00 sec)

mysql> pager cat >> /dev/null
PAGER set to 'cat >> /dev/null'
mysql>  select * from z1 force index(a) order by a desc;
56415 rows in set (24.94 sec)

mysql> pager
Default pager wasn't set, using stdout.
mysql>  show status like 'Handler_read%';
+-----------------------+-------+
| Variable_name         | Value |
+-----------------------+-------+
| Handler_read_first    | 0     |
| Handler_read_key      | 1     |
| Handler_read_last     | 1     |
| Handler_read_next     | 0     |
| Handler_read_prev     | 56415 |
| Handler_read_rnd      | 0     |
| Handler_read_rnd_next | 0     |
+-----------------------+-------+
7 rows in set (0.01 sec)

不用过多解释，可以看到 Handler_read_last 和 Handler_read_prev 的用途。

四、总结

  • Handler_read_rnd_next 通常代表着全表扫描。
  • Handler_read_first 通常代表着全表或者全索引扫描。
  • Handler_read_next 通常代表着合理的使用了索引或者全索引扫描。
  • Handler_read_key 不管全表全索引或者正确使用的索引实际上都会增加，只是一次索引定位而已。
  • Innodb 中全表扫描也是主键的全索引扫描。

顺序访问的一条记录实际上都是调用 ha_innobase::general_fetch 函数，另外一个功能 innodb_thread_concurrency 参数的功能就在里面实现，下次再说。

五、参考栈帧

全表扫描

mysql> desc select * from z1 ;
+----+-------------+-------+------------+------+---------------+------+---------+------+-------+----------+-------+
| id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | filtered | Extra |
+----+-------------+-------+------------+------+---------------+------+---------+------+-------+----------+-------+
| 1 | SIMPLE | z1 | NULL | ALL | NULL | NULL | NULL | NULL | 56650 | 100.00 | NULL |
+----+-------------+-------+------------+------+---------------+------+---------+------+-------+----------+-------+
1 row in set, 1 warning (0.00 sec)

第一次:
#0 row_search_mvcc (buf=0x7fff2ccc9380 "\377", mode=PAGE_CUR_G, prebuilt=0x7fff2cd4bb40, match_mode=0, direction=0)
at /root/mysql5.7.14/percona-server-5.7.14-7/storage/innobase/row/row0sel.cc:4479
#1 0x00000000019b3051 in ha_innobase::index_read (this=0x7fff2cd32480, buf=0x7fff2ccc9380 "\377", key_ptr=0x0, key_len=0, find_flag=HA_READ_AFTER_KEY)
at /root/mysql5.7.14/percona-server-5.7.14-7/storage/innobase/handler/ha_innodb.cc:9104
#2 0x00000000019b4374 in ha_innobase::index_first (this=0x7fff2cd32480, buf=0x7fff2ccc9380 "\377")
at /root/mysql5.7.14/percona-server-5.7.14-7/storage/innobase/handler/ha_innodb.cc:9551
#3 0x00000000019b462c in ha_innobase::rnd_next (this=0x7fff2cd32480, buf=0x7fff2ccc9380 "\377")
at /root/mysql5.7.14/percona-server-5.7.14-7/storage/innobase/handler/ha_innodb.cc:9656
#4 0x0000000000f66fa2 in handler::ha_rnd_next (this=0x7fff2cd32480, buf=0x7fff2ccc9380 "\377") at /root/mysql5.7.14/percona-server-5.7.14-7/sql/handler.cc:3099
#5 0x00000000014c61b6 in rr_sequential (info=0x7fff2c0026a0) at /root/mysql5.7.14/percona-server-5.7.14-7/sql/records.cc:520
#6 0x000000000155f2a4 in join_init_read_record (tab=0x7fff2c002650) at /root/mysql5.7.14/percona-server-5.7.14-7/sql/sql_executor.cc:2481
#7 0x000000000155c381 in sub_select (join=0x7fff2c001f70, qep_tab=0x7fff2c002650, end_of_records=false)
at /root/mysql5.7.14/percona-server-5.7.14-7/sql/sql_executor.cc:1271
#8 0x000000000155bd

最后推荐高鹏的专栏《深入理解 MySQL 主从原理 32 讲》，想要透彻了解学习 MySQL 主从原理的朋友不容错过。

正文完
innodb mysql 数据库
发表至：无分类
2019-10-31
 0