原本只是想驗證一下選擇不同索引對innodb count(*)查詢速度的影響。
各位順道可參考下這篇文章 [InnoDB系列] -- innodb表如何更快得到count(*)結果。
測試過程中沒想到同樣的一條sql語句僅僅是增加了force index後查詢速度幾乎快了一倍。
select count(*) from http_log_3 force index(time) where time >= 000000 //1 row in set (11 min 19.35 sec)
select count(*) from http_log_3 where time >= 000000 //1 row in set (20 min 5.86 sec)
但實際上通過explain分析可知其實這兩條sql語句使用的都是time索引，完全一樣！
在這個特例當中使用force index(time)後影響的並不是索引key的選擇（優化器默認也使用time索引），而是type及rows.
很想知道這是爲什麼，rows是如何被估算出來的，可有公式？
測試環境：
數據庫 mysql 5.1.34，innodb引擎，使用innodb_file_per_table選項。
使用表分區方式創建數據表（按日分區共十個），表中一共有5000萬數據，即每個分區各500萬。
測試輸出：
--------------
explain partitions select count(*) from http_log_3 force index(time) where time >= 000000
--------------
+----+-------------+------------+-----------------------------------+-------+---------------+------+---------+------+----------+--------------------------+
| id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-----------------------------------+-------+---------------+------+---------+------+----------+--------------------------+
| 1 | SIMPLE | http_log_3 | p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,p10 | range | time | time | 3 | NULL | 25000141 | Using where; Using index |
+----+-------------+------------+-----------------------------------+-------+---------------+------+---------+------+----------+--------------------------+
1 row in set (0.01 sec)
--------------
explain partitions select count(*) from http_log_3 where time >= 000000
--------------
+----+-------------+------------+-----------------------------------+-------+--------------------------+------+---------+------+----------+--------------------------+
| id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-----------------------------------+-------+--------------------------+------+---------+------+----------+--------------------------+
| 1 | SIMPLE | http_log_3 | p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,p10 | index | time_ip,time_domain,time | time | 3 | NULL | 50000291 | Using where; Using index |
+----+-------------+------------+-----------------------------------+-------+--------------------------+------+---------+------+----------+--------------------------+
1 row in set (0.01 sec)
--------------
explain partitions select count(*) from http_log_3
--------------
+----+-------------+------------+-----------------------------------+-------+---------------+------+---------+------+----------+-------------+
| id | select_type | table | partitions | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+------------+-----------------------------------+-------+---------------+------+---------+------+----------+-------------+
| 1 | SIMPLE | http_log_3 | p0,p1,p2,p3,p4,p5,p6,p7,p8,p9,p10 | index | NULL | time | 3 | NULL | 50000291 | Using index |
+----+-------------+------------+-----------------------------------+-------+---------------+------+---------+------+----------+-------------+
1 row in set (0.00 sec)
--------------
select count(*) from http_log_3 force index(time) where time >= 000000
--------------
+----------+
| count(*) |
+----------+
| 50000000 |
+----------+
1 row in set (11 min 19.35 sec)
--------------
select count(*) from http_log_3 where time >= 000000
--------------
+----------+
| count(*) |
+----------+
| 50000000 |
+----------+
1 row in set (20 min 5.86 sec)
--------------
select count(*) from http_log_3
--------------
+----------+
| count(*) |
+----------+
| 50000000 |
+----------+
1 row in set (20 min 6.32 sec)