Mysql：为什么我在定义表时使用适当的数据大小并选择最佳数据长度

Question

I have a table which stores the details of users. I can hardly have 10K values. Its ID field is defined as bigint(20), which can hold even big big data ranges.

Now, changing it to SMALLINT will bring any favour wither in terms of performance or storage...? Shall some one plase explain how will it.

i create two small tables
one with ID as INT(10)
another with ID as INT(100)

I inserted 513 rows into each. when i see the show create table of each of them, i did not see any change in the data sizes or index size. they are MYISAM tables. Then whats the better thing in choosing SMALLINT than int(100) or INT(10)

Here is that info

| id    | int(10) | NO   | PRI | NULL    | auto_increment |
| size  | int(10) | YES  |     | NULL    |                |
Data_length: 4617
Index_length: 8192


| id    | int(100) | NO   | PRI | NULL    | auto_increment |
| size  | int(10)  | YES  |     | NULL    |                |
Data_length: 4617
Index_length: 8192

Answer 1

There's an important distinction here between the data type and the "length".

int(10) and int(100) are actually the same data type, so they will both take up 4 bytes. The "10" and "100" just affect the way the data is displayed, not the way it is stored.

Choosing the data type is a trade-off between storage efficiency and the flexibility to store a greater range of values.

Here's a helpful chart from the manual :

Type    Storage     Minimum Value       Maximum Value
        (Bytes)     (Signed/Unsigned)   Signed/Unsigned)
TINYINT     1   -128    127
                0   255
SMALLINT    2   -32768  32767
                0   65535
MEDIUMINT   3   -8388608    8388607
                0   16777215
INT     4   -2147483648     2147483647
                0   4294967295
BIGINT  8   -9223372036854775808    9223372036854775807
                0   18446744073709551615

Answer 2

One reason to be more specific in the field sizes is simply because it reduces the amount of mistakes. The more you limit someones particular freedom, the more accurate the information will be. That's why on a lot of sites when you register and it has drop down lists for what country, or what state you live in. This reduces the amount of options the user has, thus maintaining safe practices. If a site allowed you to type in your state, think of every example you would get from different users. For example, Florida would be:

• Florida
• FL
• fl
• Folrida

If you notice, I misspelt Florida wrong in the last example. Now imagine trying to run a query on users from Florida, which one would you choose? You would have to take into account every option that a user could input. Limiting their freedom so you KNOW what is inside the field is the best way.