mysql - mysql 表上的正确索引

Question

当我在一个有 22M 行的表上运行以下查询时，需要 20 秒才能运行:

select p.*,
(select avg(close)
from endOfDayData p2
where p2.symbol = p.symbol and
p2.date between p.date - interval 6 day and p.date
) as MvgAvg_X
from endOfDayData p
where p.symbol = 'AAPL'

表结构如下:

mysql> desc endOfDayData;
+--------+---------------+------+-----+---------+-------+
| Field  | Type          | Null | Key | Default | Extra |
+--------+---------------+------+-----+---------+-------+
| date   | date          | NO   | PRI | NULL    |       |
| symbol | varchar(14)   | NO   | PRI | NULL    |       |
| open   | decimal(10,4) | NO   |     | NULL    |       |
| high   | decimal(10,4) | NO   |     | NULL    |       |
| low    | decimal(10,4) | NO   |     | NULL    |       |
| close  | decimal(10,4) | NO   |     | NULL    |       |
| volume | int(11)       | NO   |     | NULL    |       |
+--------+---------------+------+-----+---------+-------+

存在以下索引:

mysql> show index from endOfDayData;
+--------------+------------+------------+--------------+-------------+-----------+-------------+----------+--------+------+------------+---------+---------------+
| Table        | Non_unique | Key_name   | Seq_in_index | Column_name | Collation | Cardinality | Sub_part | Packed | Null | Index_type | Comment | Index_comment |
+--------------+------------+------------+--------------+-------------+-----------+-------------+----------+--------+------+------------+---------+---------------+
| endOfDayData |          0 | PRIMARY    |            1 | date        | A         |      162294 |     NULL | NULL   |      | BTREE      |         |               |
| endOfDayData |          0 | PRIMARY    |            2 | symbol      | A         |    24019617 |     NULL | NULL   |      | BTREE      |         |               |
| endOfDayData |          1 | EOD_dates  |            1 | date        | A         |       50145 |     NULL | NULL   |      | BTREE      |         |               |
| endOfDayData |          1 | EOD_symbol |            1 | symbol      | A         |       14322 |     NULL | NULL   |      | BTREE      |         |               |
+--------------+------------+------------+--------------+-------------+-----------+-------------+----------+--------+------+------------+---------+---------------+
4 rows in set (0.00 sec)

该机器是一个专用盒子，配有 80GB 内存和双处理器。我觉得它应该在一秒钟内运行并具有正确的索引。谢谢

| id | select_type        | table | type | possible_keys                | key        | key_len | ref                | rows | Extra                 |  
+----+--------------------+-------+------+------------------------------+-------‌​-----+---------+--------------------+------+-----------------------+
|  1 | PRIMARY            | p     | ref  | EOD_symbol                   | EOD_symbol | 16      | const              | 8409 | Using index condition |  
|  2 | DEPENDENT SUBQUERY | p2    | ref  | PRIMARY,EOD_dates,EOD_symbol | EOD_symbol | 16      | financial.p.symbol | 1677 | Using index condition |

我创建了一个新表，以 ID int 作为主键，并在符号、日期上创建了索引。

CREATE INDEX EODDateSym ON endOfDayData_new (symbol, date) USING BTREE;

还有 17 秒。再次感谢所有的想法和帮助

我的 my.conf 是

[mysql]

# CLIENT #
port                           = 3306
socket                         = /var/lib/mysql/mysql.sock

[mysqld]

# GENERAL #
user                           = mysql
default-storage-engine         = InnoDB
socket                         = /var/lib/mysql/mysql.sock
pid-file                       = /var/lib/mysql/mysql.pid

# MyISAM #
key-buffer-size                = 32M
myisam-recover                 = FORCE,BACKUP

# SAFETY #
max-allowed-packet             = 16M
max-connect-errors             = 1000000

# DATA STORAGE #
datadir                        = /var/lib/mysql/

# BINARY LOGGING #
log-bin                        = /var/lib/mysql/mysql-bin
expire-logs-days               = 14
sync-binlog                    = 1
server_id              = 1

# CACHES AND LIMITS #
tmp-table-size                 = 32M
max-heap-table-size            = 32M
query-cache-type               = 0
query-cache-size               = 0
max-connections                = 500
thread-cache-size              = 50
open-files-limit               = 65535
table-definition-cache         = 4096
table-open-cache               = 4096

# INNODB #
innodb-flush-method            = O_DIRECT
innodb-log-files-in-group      = 2
innodb-log-file-size           = 512M
innodb-flush-log-at-trx-commit = 1
innodb-file-per-table          = 1
innodb-buffer-pool-size        = 68G

# LOGGING #
log-error                      = /var/lib/mysql/mysql-error.log
log-queries-not-using-indexes  = 1
slow-query-log                 = 1
slow-query-log-file            = /var/lib/mysql/mysql-slow.log

Answer 1

由于您要同时匹配 symbol 和 date，因此您需要在这两列上设置索引，(symbol, date)，以便其在您所表达的条件下有效。

MySQL 通常会为给定的表选择最适合作业的索引，并且不能以任何有意义的方式组合两个索引。

如果您将这两个作为主键，那就很奇怪并且可能会损害性能。 UNIQUE 索引与常规 INT AUTO_INCRMENT PRIMARY KEY ID 类型列配合使用效果更好。当主键尽可能紧凑时，MySQL 的性能最佳。