Optimization steps of SQL statements (II) -- MySQL statement optimization

       mysql Database is a widely used database language , In order to improve the sql Statement execution efficiency , In our usual design process , The following aspects can be used for reference , Can effectively improve sql Statement execution efficiency .

1、 Mass insert data （sql files inserting ）

        adopt load When the command imports data , Appropriate settings can improve the efficiency of import

        about innoDB Tables of type are saved in the order of primary keys , So import the data to be imported in the order of the primary key , It can effectively improve the efficiency of data

   1） Primary key inserted successfully

Add ： head sql file name View file data memory

Data import format

load data local infile ' route   file name ' into table ' Table name '
            fields terminabled by ',' 
            lines terminated by '\n'

nto table :                  File data needs to be inserted into that table

feilds terminabled ： The field is divided by what symbol

lines terminabled:      What symbol is used to divide the row

Be careful ： When there is a primary key, the import speed will be faster , Therefore, it is recommended to use the primary key

   2） Turn off uniqueness check

        Turn off uniqueness check , After the data import is completed, the uniqueness verification can be started , It can also improve efficiency

 Turn off uniqueness check  set unique_checks=0
 Enable uniqueness verification  set unique_checks=1

 Be careful ：    Opening and closing must exist in pairs 

    3） Close the commit transaction manually

              Turn off auto submit manually , Then turn on auto submit after data import is completed , It will also improve efficiency

set auto_committed=0; 
set auto_committed=1;

2、 Optimize insert sentence

   1). Try to use a single insert Sentence block

       insert into student values ( , ) ,( , );

   2). Use manual commit of transactions

set auto_committed=0; // Turn off auto submit 
start transaction;   // Open transaction 
insert into...... //select Execute statement 
commit(); // Manual submission 
set auto_committed=1;  // Turn on auto submit 

3. Data is inserted in order （ The primary key is inserted orderly ）

// Insert in order by primary key 

insert into student values(1,);

insert into student values(2,);

3、order by Optimize

            How to optimize ,

                    Use index sorting （using index）

                    Use one scan to sort （filesort）

  3.1 Two ways of sorting

         filesort( File system sorting ) ： Not directly through the index to return results

         using index( Index sort ) ： Return results directly through the index （ The return fields are all index fields ）, High operating efficiency

1、 Single field sorting

2、 Multi field sorting ： Press asc Or both desc Sort , Inconsistencies occur filesort Sort

3、 Press order by Then there is the multi field sorting , Sort by index

      Such as idx_age_salary ,order by age,salary

   3.2 Filesort Optimize

        about filesort ,mysql There are two algorithms in

1. Two scan algorithm （mysql4.1 Before ）

            First, take out the sorting field and row pointer information according to the conditions , stay sort buffer Middle order , Next, read the record back to the table according to the row pointer , This operation may cause a large number of random IO The operation of , Low efficiency

2. One scan algorithm

      Take out all the fields that meet the conditions at once , Then in the sorting area sort buffer After sorting, output the result set directly , High memory overhead , But it's more efficient

       mysql How to choose ？

  Compare system variables max_length_for_sort_data and Query The total size of the fields extracted from the statement

 max_length_for_sort_data Big words , Will use the second algorithm

Optimization strategy

          Improve max_length_for_sort_data and sort_buffer_size Value , Increase the size of the sort area , Make it mandatory to use the second （ One scan algorithm ）, Improve the efficiency of sorting

 Set up max_length_for_sort_data 
show variables like 'max_length_for_sort_data'; 
set max_length_for_sort_data= value ; 

 Set up  sort_buffer_size 
show variables like 'sort_buffer_size';
 set sort_buffer_size= value ;

4、group by Grouping sorting

  Optimize ： Index field grouping sort , The grouping field needs to be set as index to improve efficiency

            because group by During operation , Also used order by Sort , Then you can use index fields to sort by groups .

let me put it another way ： Set the grouped field as the index field , So when grouping , Is to sort by index group , More efficient

5、 Optimize nested queries （ Subquery ）

      Optimize ： Use as few subqueries as possible , Change to table connection join Inquire about

6、OR The optimization of the

        Optimize ：OR The association conditions are all index fields

        Optimize ： Use union Join table optimization （ recommend ）

Use OR When it comes to efficiency , It should be noted that ：  

 1、 OR The associated conditions must all be index fields , Otherwise, it will not be impossible to use index query

  2、 Cannot be a composite index , Must be indexed with a single column

7、limit Paging query

    Optimize one ： Complete sort paging operation on Index , Finally, according to the primary key Association, return to the original table to query other column contents

// Before optimization 
select * from student limit 20000,10;

// After optimization 

select * from student a,(select id from student limit 20000,10) b where a.id=b.id;

Optimization II 、 Applicable to tables with self incrementing primary key ( A self increasing primary key cannot have a fault ), You can put limit The query is converted to a query in a certain location

select * from student where id >20000 limit 10;

8、 Use SQL Tips （ Index tips ）

       SQL Prompt is an important means to optimize database , Simply speaking , Is in the SQL Add some human prompts in the statement to optimize the operation

1、USE INDEX Which index is recommended ( stay from Use after the table name )

      After the table name in the query statement , add to use index To provide hope mysql To refer to the index list , You can make mysql No longer consider other available indexes

    select * from Table name use index( Indexes ) where ..... Case study

select * from student use index(name_index) where name=' Zhang San ';

2、IGNORE INDEX Ignore the index

   select * from Table name ignore index( Indexes ) where ..... Ignore that index

3、FORCE INDEX Force which index to use

    select * from Table name force index( Indexes ) where ..... Ignore that index

  Be careful ：use and ignore Just recommend , It's not mandatory ,force Is mandatory