gil*_*ilm 5 mysql performance index count mysql-5.5 query-performance
我有一个与此类似的表(简化):
CREATE TABLE books (
id INT AUTO_INCREMENT,
category INT NOT NULL,
PRIMARY KEY (id),
KEY (category)
);
Run Code Online (Sandbox Code Playgroud)
这张桌子已经结束了10,000,000 rows,大约在12 categories。所以每个类别的平均值为833,333 books.
查询计数时:
SELECT COUNT(*) FROM books WHERE category=1;
Run Code Online (Sandbox Code Playgroud)
即使它在查询时使用索引,这也需要很长时间才能完成(几秒钟)。您将如何优化它?
之前我每次插入书籍时都会增加一个数字(插入一个涉及类别->书籍数量的表。)但是我们的代码很复杂,很多地方插入或删除书籍。我知道可以用 解决这个问题EVENTS,但我在问也许我错过了一个 MySQL 功能。
category由于索引基数较低,查询会很慢。有 12 个类别,因此平均查询将读取索引的 1/12 部分。您无法改进此查询。
您原来的方法可以提高整体性能。只需book_count在 INSERT 和 DELETE 事件上创建触发器,而不是手动更新。
更新:证明查询将部分读取索引category
mysql> select count(*) from books;
+----------+
| count(*) |
+----------+
| 1000 |
+----------+
1 row in set (0.00 sec)
mysql> select category, count(*) from books group by 1;
+----------+----------+
| category | count(*) |
+----------+----------+
| 0 | 50 |
| 1 | 77 |
| 2 | 88 |
| 3 | 84 |
| 4 | 102 |
| 5 | 79 |
| 6 | 79 |
| 7 | 73 |
| 8 | 84 |
| 9 | 76 |
| 10 | 87 |
| 11 | 83 |
| 12 | 38 |
+----------+----------+
13 rows in set (0.01 sec)
mysql> flush status;
Query OK, 0 rows affected (0.00 sec)
mysql> select count(*) from books where category = 6;
+----------+
| count(*) |
+----------+
| 79 |
+----------+
1 row in set (0.00 sec)
mysql> show status like 'Hand%';
+----------------------------+-------+
| Variable_name | Value |
+----------------------------+-------+
| Handler_commit | 1 |
| Handler_delete | 0 |
| Handler_discover | 0 |
| Handler_external_lock | 2 |
| Handler_mrr_init | 0 |
| Handler_prepare | 0 |
| Handler_read_first | 0 |
| Handler_read_key | 1 |
| Handler_read_last | 0 |
| Handler_read_next | 79 |
| Handler_read_prev | 0 |
| Handler_read_rnd | 0 |
| Handler_read_rnd_next | 0 |
| Handler_rollback | 0 |
| Handler_savepoint | 0 |
| Handler_savepoint_rollback | 0 |
| Handler_update | 0 |
| Handler_write | 0 |
+----------------------------+-------+
18 rows in set (0.01 sec)
Run Code Online (Sandbox Code Playgroud)