慣用手法: 擷取不同長度,並測試其區分度,

<span style="font-size:18px;">select count(distinct left(word,6))/count(*) from dict; 
+---------------------------------------+
| count(distinct left(word,6))/count(*) |
+---------------------------------------+
|                                0.9992 |
+---------------------------------------+</span>
 

擷取word欄位長度，從1開始擷取，計算字元字首沒有重複的字元佔全部資料的比例

對於一般的系統應用: 區別度能達到0.1,索引的效能就可以接受.

3:多列索引
 3.1 多列索引的考慮因素---  
列的查詢頻率 , 列的區分度, 
以ecshop商城為例, goods表中的cat_id,brand_id,做多列索引
從區分度看,Brand_id區分度更高, 

select count(distinct cat_id) / count(*) from  goods;
+-----------------------------------+
| count(distinct cat_id) / count(*) |
+-----------------------------------+
|                            0.2903 |
+-----------------------------------+
1 row in set (0.00 sec)

mysql> select count(distinct brand_id) / count(*) from  goods;
+-------------------------------------+
| count(distinct brand_id) / count(*) |
+-------------------------------------+
|                              0.3871 |
+-------------------------------------+
1 row in set (0.00 sec)
 

+------+-----------+------+
| id   | name      | pid  |
+------+-----------+------+
| .... | .... | .... |
| 1584 | 日照市 | 1476 |
| 1586 | 東港區 | 1584 |
| 1587 | 五蓮縣 | 1584 |
| 1588 | 莒縣    | 1584 |
+------+-----------+------+


1: 不加任何索引,自身連線查詢

explain select s.id,s.name from it_area as p inner join it_area as s on p.id=s.pid   where p.name='日照市' \G
*************************** 1. row ***************************
           id: 1
  select_type: SIMPLE
        table: p
         type: ALL
possible_keys: NULL
          key: NULL
      key_len: NULL
          ref: NULL
         rows: 3263
        Extra: Using where
*************************** 2. row ***************************
           id: 1
  select_type: SIMPLE
        table: s
         type: ALL
possible_keys: NULL
          key: NULL
      key_len: NULL
          ref: NULL
         rows: 3263
        Extra: Using where; Using join buffer
2 rows in set (0.00 sec)

explain select s.id,s.name from it_area as p inner join it_area as s on p.id=s.pid   where p.name='日照市' \G
*************************** 1. row ***************************
           id: 1
  select_type: SIMPLE
        table: p
         type: ref
possible_keys: name
          key: name
      key_len: 93
          ref: const
         rows: 1
        Extra: Using where
*************************** 2. row ***************************
           id: 1
  select_type: SIMPLE
        table: s
         type: ALL
possible_keys: NULL
          key: NULL
      key_len: NULL
          ref: NULL
         rows: 3243
        Extra: Using where; Using join buffer
2 rows in set (0.00 sec)

explain select s.id,s.name from it_area as p inner join it_area as s on p.id=s.pid   where p.name='日照市' \G
*************************** 1. row ***************************
           id: 1
  select_type: SIMPLE
        table: p
         type: ref
possible_keys: name
          key: name
      key_len: 93
          ref: const
         rows: 1
        Extra: Using where
*************************** 2. row ***************************
           id: 1
  select_type: SIMPLE
        table: s
         type: ref
possible_keys: pid
          key: pid
      key_len: 5
          ref: big_data.p.id
         rows: 4
        Extra: Using where
2 rows in set (0.00 sec)