18、 MySQL -- index

1. What is index ？

         Indexes It's a separate 、 Physical to database tables A column of or Multiple columns The value of the to Sort A kind of Storage structure .

         It's equivalent to a book catalog , The function is to improve SELECT Statement efficiency .

         Since it is equivalent to a directory , The catalogue is also the content of the book , The stored content is in the form of data , therefore Index is also data .

2. Why index is needed ？

         We need to search data from the table , The common method is full table scanning , That is, search from the first data block of the table , Find the last data block in order . When the amount of data is small , Yes MySQL Service requirements are not high , But when the amount of data is large MySQL High performance consumption , Inefficient search , So is there an efficient way ？

When we read books looking for specific content , You can find the target content from the directory first , Then go to the specific number of pages to find the content , The index came into being .

3.MySQL Index in

        innodb Storage engine support 3 Class index

                b+ Tree index

                 Hash index

                 Full table index

        among b+ Tree index is used most , Therefore, this article only discusses b+ Tree index

b+ A tree is a data structure , Data used to store indexes

b+ Trees (BTREE) It is divided into Clustered index (clustered index) and Secondary index (secondary index or non-clustered index)

        Clustered index

        Definition ： According to each table Primary key Build a tree b+ Trees , At the same time, what is stored in the leaf node is the The whole line of data .

        The leaf node of the clustered index is also called Data pages , Index organization table （ That is, the data in the table is stored in the order of the primary key ） The data in is also part of the index . Because the actual data page can only follow one b+ Trees sort , Therefore, a table can only have One Clustered index . Query optimizer Tends to use clustered indexes , Because the clustered index can find data directly on the leaf node of the index .

 # Because the index has sorting function , So when using query statements to query rows with indexes , There is no need to add sorting statements

        Secondary index

         Leaf node No Containing rows All data . Leaf nodes contain the index key values of rows , Each index row also contains a primary key value , The primary key value tells innodb Storage engine , Where can I find the data corresponding to the index

        # Contains data from other columns and pointers to the primary key index

         Single index

        The number of key values of the index is 1

         Composite index

        Unlike a single column index, the number of keys in a composite index is not 1, But greater than or equal to 2

        # Query a single leading column in the composite index , You can also use this composite index , However, compound queries cannot be used for queries with non leading columns .

        Create index

        CREATE INDEX id_name_idx ON IG(id,name);
        # In the table IG To create id As the leader name Is a non leading composite index id_name_idx

        Look at the index

        show index from IG;
        # See the table IG All the indexes of

        Some parameters in the table ：

                Key_name --》 Index name

                Non_unique--》 by 0 Expressed as a unique index , by 1 Expressed as a non unique index

                Collation --》 How to save column index as .A Expressed as ascending sort ,null It means no sorting

                Cardinality--》 The closer the value is to the number of rows , The richer the fields , The more suitable for indexing , On the contrary, it is not suitable for indexing

        Delete index

        DROP INDEX id_name_idx on IG;

        # Delete table IG Index in id_name_idx

        

  Two sides of index

        advantage ： Improve query efficiency

        shortcoming ：  Modified the data ,MySQL I will modify the index , Lead to MySQL Performance degradation . The cost of maintaining indexes is relatively large