MySQL index

Indexes

#  Basic concepts 

-    Indexes ： Data structure to optimize query efficiency 
-    shortcoming ：
    1.   Maintaining indexes consumes database resources 
    2.   Take up disk space 
    3.   Maintain the index when adding, deleting and modifying 
-    principle : Index common search fields 
-    Index classification 
    -   InnoDB（ Must be ）   Primary key    only    Single column    Reunite with 
        1.   primary key （ Cluster index ）
        2.   Single index （ Single value , Ordinary ）
        3.   unique index （ The value of the column is unique , It can be for null）
        4.   Composite index （ Multiple columns in one index ）
    -   MyISAM（ Unsupported transaction ）
        -   Full Text  :  Full-text index 

#  Basic operation of index ：
-  Query index 
	show index from tableName
-  Creation of ordinary index  key()
    1.  Create... When creating a table   
        create table t user(id varchar(20) primary key, name varchar(20), key(column));
    2.  After creating a table, create 
        create index indexName on table(tableColumnName)
-  unique index 
	1.  Create... When creating a table  
	create table t user(id varchar(20) primary key, name varchar(20), unique(column));
	2.  After creating a table, create 
    create unique index indexName on table(tableColumnName)
-  Composite index 
	1.  Create... When creating a table  
	create table t user(id varchar(20) primary key, name varchar(20), key(column1,column2));
	2.  After creating a table, create 
    create index indexName on table(col1,col2)
    3.  principle  ：（ interview ）
    	1.  Leftmost prefix principle （ Left containment principle ）
    	 Create order, such as (name,age), that name age bir Sure ,age name bir Can not be ()
    	2.  Dynamically adjust the query order , In order to make better use of queries 	    
 ##  The index name cannot be specified when creating a table , After creating the table, you can specify the index name    

#  Underlying data structure （B+ Tree）   It's usually 2-4 layer 
##  The phenomenon 
1.  Insert id disorder 
2.  Automatically sort when querying （ The primary key index is sorted ）
##  explain 
1.  Index auto sort , Convenient for quick query in the future 
##  other 
1.  Page based management , Add the principle that the index will automatically sort 
2.  By default, every page is 16KB
3.  A page table （ Store the first pointer of each page , You can link directly to the corresponding page ）
4.  Top level pages reside in memory 

#  Clustered index and non clustered index 
-  Cluster index ：（ The default is primary key , If there is no primary key, you will find a unique and non empty index ）
	 Put the data and index together , The leaf node of the index structure holds the row data （B+）
-  Nonclustered index ：
	 Separate data from index , The leaf node of the index structure points to the corresponding position of the data 
	 The primary key of the data is searched , Need to search again is the real data 
	 The reason why there is no address is ： The storage address needs to be changed to the address of the auxiliary index , If you use a primary key, you don't have to consider modifying the address .
##  The primary key index must be a clustered index , Otherwise, it doesn't work .
##  A table can only have one clustered index , Others are non clustered or auxiliary indexes .

## InnoDB  Clustering advantages 
	1.  Save the primary key value , It can prevent the maintenance of non clustered trees when modifying addresses , It can also reduce disk consumption 
	2.  Clustered indexes store primary key and row data , Find the leaf node and you can return data 
## MyISAM Nonclustered index , It's the address 

#  Why is it recommended that the primary key be self incremented ？
1.  If at this time a page   1 2 3 4 5 6 7 8 10 11, If you want a primary key 9 Data insertion , According to the principle of sorting by primary key , Will plug into 8 Back , But if this page is full , That will cause page splitting and record shifting , Cause performance loss .

#  Index failure scenario 
1.  The best left prefix rule （ important ）
2.  It's not equal to (!=  perhaps <>) Cause index to fail 
3.  Primary key insertion order  （ Why is it recommended that the primary key be self incremented ？）
4.  Calculation 、 function 、 Index invalidation due to type conversion 
5.  The column index to the right of the range condition fails （ Multiple composite indexes are in different order ）
6. is null You can use index ,is not null Index not available 
7. like Use wildcards % Invalid start index （ important ）
8. OR  As long as there are non indexed columns before and after , Will lead to index invalidation  
9.  Database and table character sets are used uniformly utf8mb4

## INNODB Maximum ID Record in memory , Restart the database or OPTIMIZE The operation will be lost .
	- optimize You can spread it out (fragmented) The stored data is moved back together (defragmentation), Clear away the debris , Reclaim idle database space 
	-  This action locks the table 

# HEAP  Tables exist in memory , For temporary high-speed storage .
# CHAR_LENGTH  It's the number of characters , and  LENGTH  Is the number of bytes .

#  SELECT VERSION(); Used to get the current  Mysql  Version of .
#   Create table time  TIMESTAMP  Column use  Zero  to update . As long as other fields in the table change ,
#  UPDATE CURRENT_TIMESTAMP  The Modifier updates the timestamp field to the current time 
#  shell Sign in MySQL
#  [mysql dir]/bin/mysql -h hostname -u <UserName> -p <password>
#  MySQL  A single watch exceeds 500 The capacity of ten thousand rows or single table exceeds 2GB Divide the database or table 
#  Like  Fuzzy query ：％ Corresponding to  0  Characters or more ,_ It's just  LIKE  A character in a statement .
#  Mysql stay V5.1 The previous default storage engine was MyISAM; After that, the default storage engine is InnoDB
#   Any standard table can be built at most 16 Index columns 
# NUMERIC  and  DECIMAL Used to save data with high progress , For example, the financial industry 

MySQL Business

• MySQL The default is autocommit Mode , Transaction is not supported by default , All database update operations will be submitted immediately .
• If the engine is InnoDB, Then you can set support transactions .Set AutoCommit = 0 , The update operation in this mode requires manual commit Will take effect , Use Rollback Perform transaction rollback .

• #  Open transaction 
  START TRANSACTION;
  SELECT @A:=SUM(salary) FROM table1 WHERE type=1;
  UPDATE table2 SET summmary=@A WHERE type=1;
  COMMIT;
  

MySQL lock