Execution process of MySQL query request - underlying principle

One 、 Connection management

Client processes can adopt TCP/IP、 Named pipes or shared memory 、UNIX Domain socket And other communication methods to establish a connection with the server process . Whenever a client process connects to a server process , The server process will create a thread to deal with the interaction with the client ; When the client exits, it will disconnect from the server , The server does not immediately destroy the threads that interact with the client , Instead, cache it , When another new client connects , The cached thread will be allocated to the new client . This eliminates the need to create and destroy threads frequently , So it saves money . From this we can also see ,MySQL The server will assign a thread to each connected client , However, too many threads will seriously affect the system performance , So we also need to limit the number of clients that can connect to the server at the same time . When the client program initiates a connection , Need to carry host information 、 user name 、 Password and other information , The server program authenticates the information provided by the client program . If authentication fails , The server program will reject the connection . in addition , If the client program and the server program do not run on the same computer , We can also adopt Transport layer security (TransportLayer Security,TLS) The protocol encrypts the connection , So as to ensure the security of data transmission .

When the connection is established , The server thread associated with the client will always wait for the request sent by the client .MySQL The request received by the server is just a text message , The text message has to go through various processes . What if you want to know what happened , Please read on .

Two 、 Analysis and optimization

Up to now ,MySQL The server has obtained the request in text form , Then I have to go through “ Ninety-eight is difficult ” To deal with , Some of the more important parts are The query cache 、 Syntax parsing and Query optimization . Let's take a closer look at .

2.1 The query cache

If I ask you 9 + 8 * 16- 3 * 2 * 17 What's the value of , You may use a calculator to calculate , Or more powerful, direct mental arithmetic , Finally, we got the result 357. If I ask you again 9 + 8 * 16 - 3 * 2 * 17 What's the value of , Will you still calculate foolishly ? We have just calculated , Just say the answer directly .

MySQL The same is true for the server program to process query requests , It will cache the query requests and results that have just been processed . If you have the same request next time , Just look up the results directly from the cache , There is no need to look up in the underlying table . This query cache can be shared between different clients . in other words . If the client A Just sent a query request , And the client B Then the same query request is sent , So the client B You can directly use the data in the query cache for this query .

Of course ,MySOL The server is not as smart as anyone , If there is any character difference between the two query requests ( for example , Space 、 notes 、 Case write ), Will cause the cache to miss . in addition , If the query request contains some system functions 、 User defined variables and functions 、 The system tables , Such as myql、information_schema、performance_schema Tables in the database , Then the request will not be cached . Take some system functions as an example , Two calls to the same function may produce different results . Like functions NOW, Each call will generate the latest current time . If this function is called in two query requests , Even if the text information of the query request is the same , Then two queries at different times should also get different results . If the results are cached at the first query , Using the result of the first query directly in the second query is wrong !

But since it's caching , Then there is a time when the cache fails .MySQL Our caching system monitors every table involved , As long as the structure or data of the table is modified , For example, we used INSERT、UPDATE、DELETE、TRUNCATE TABLE、ALTER TABLE、DROP TABLE or DROP DATABASE sentence , Then all query caches related to the table will become invalid and be deleted from the query cache !

Although query caching can sometimes improve system performance , But it also has to incur some overhead due to maintaining this cache . For example, we need to search in the query cache every time , After the query request is processed, the query cache needs to be updated , The memory area corresponding to the query cache needs to be maintained 、 from MySQL 5.7.20 Start , Query caching is not recommended , stay MySQL 8.0 Delete it directly in .

2.2 Syntax parsing

If the query cache misses , Next, you need to enter the formal query stage . Because the request sent by the client program is just a piece of text , therefore MySQL The server program first analyzes this text , Determine whether the syntax of the request is correct , Then the table to be queried from the text 、 All kinds of query conditions are extracted and put into MySQL Some data structures used inside the server .

essentially , Extracting the required information from the specified text is a compilation process , Involving lexical analysis 、 Syntax analysis 、 Semantic analysis and other stages . These issues do not fall within the scope of our discussion , You just need to understand that this step is required in the process of processing requests .

2.3 Query optimization

After parsing , The server program gets the information it needs , For example, what are the tables and columns to query 、 What are the search criteria . But that's not enough , Because we write MySQL Statement execution efficiency may not be very high ,MySQL Our optimization program will optimize our statements , If external connection is converted to internal connection 、 Expression simplification 、 Sub query into a bunch of things such as connections . The result of optimization is to generate an execution plan , This execution plan indicates which indexes should be used to execute the query , And what is the connection order between tables . We can use EXPLAIN Statement to view the execution plan of a statement .

3、 ... and 、 Storage engine

Until the server program completes the query optimization , You haven't really accessed the data in the real table .MySQL The server encapsulates the data storage and extraction operations into a module called storage engine . We know , A table consists of row by row records , But it's just a logical concept . How to physically represent records , How to read data from the table and how to write data to specific physical memory , It's all the responsibility of the storage engine . In order to achieve different functions ,MySQL Provides a wide range of storage engines , Tables managed by different storage engines may have different storage structures , The access algorithm may also be different .

Why is it called an engine ? Maybe this name is more popular , In fact, this was called table processor before storage engine , Later, people may think it's too rustic , It becomes a storage engine . Its function is to receive instructions from the upper layer , Then read or write the data in the table .

MySQL Supported storage engines ：

Storage engine support for some functions ：

From 《MySQL How it works —— Understand from the root MySQL》.

I hope my sharing above will help you , thank you ！