MySQL optimization

Catalog

MySQL Three aspects of optimization ：

answer ：1. Database design 2. System usage 3. System bottlenecks

1. Database design

2. The purpose of the system

3. The bottleneck of the system

---

 

 

  This article is my previous notes , If I saw it on the Nuggets, it's right , I'm Cheng Cheng ：  Cheng Cheng My personal homepage - dynamic - Nuggets This is the only one . If there is anything wrong, you are welcome to point out the problem , I am a little white .

The link of the last article is ：MySQL Note down _ Cheng Cheng is Xiaobai's blog -CSDN Blog   . This article is about optimization

If it is springboot If so, please see springboot Create project _ Cheng Cheng is Xiaobai's blog -CSDN Blog , This is from scratch, not finished springboot, Study together ！！！！

MySQL Three aspects of optimization ：

answer ：1. Database design 2. System usage 3. System bottlenecks

1. Database design

Try to make the database smaller and take up disk space .

1). Use smaller integer types when possible .(mediumint Just like int More appropriate ).
2). Define as many fields as possible not null, Unless this field needs null.
3). If you don't use variable length fields, for example varchar, Then use a fixed size record format, such as char.
4). The primary index of the table should be as short as possible . In this way, each record has a name tag and is more efficient .
5). Create only the indexes you really need . The index is good for retrieving records , But it's not good for keeping records quickly . If you always have to search in the combined fields of a table , So create an index on these fields . The first part of the index must be the most frequently used field . If you always need to use a lot of fields , First, you should copy more of these fields , Better compression of indexes .
6). All data has to be processed before it is saved to the database .
7). All fields have to have default values .
8). In some cases , It's much faster to divide a frequently scanned table into two . When scanning a dynamic format table to get relevant records , It may use a smaller static format table, especially if .

2. The purpose of the system

1). Try to use long connections .
2).explain complex SQL sentence .
3). If two related tables are to be compared , The fields to be compared must be of the same type and length .
4).LIMIT Try to follow order by perhaps distinct. You can avoid doing it once full table scan.
5). If you want to clear all records of the table , Suggest using truncate table tablename instead of delete from tablename.
6). Be able to use STORE PROCEDURE perhaps USER FUNCTION When .
7). In one insert Multiple record insertion format is used in the statement . And use load data infile To import a lot of data , It's more than just indert Much faster .
8). often OPTIMIZE TABLE To clean up the pieces .
9). And that is date If you want to make a comparison frequently, try to save it in unsigned int Faster type .

3. The bottleneck of the system

1). Disk search .
Parallel search , Separate data into multiple disks , This will speed up the search time .
2). Disk read and write (IO)
Data can be read from multiple media in parallel .
3).CPU cycle
The data is stored in the main memory . So you have to add CPU To process the data .
4). Memory bandwidth
When CPU To store more data in CPU From the cache , The bandwidth of memory becomes the bottleneck .

This article is my previous notes , If I saw it on the Nuggets, it's right , I'm Cheng Cheng ：  Cheng Cheng My personal homepage - dynamic - Nuggets This is the only one . If there is anything wrong, you are welcome to point out the problem , I am a little white .

The link of the last article is ：MySQL Note down _ Cheng Cheng is Xiaobai's blog -CSDN Blog   . This article is about optimization

If it is springboot If so, please see springboot Create project _ Cheng Cheng is Xiaobai's blog -CSDN Blog , This is from scratch, not finished springboot, Study together ！！！！