Why do I need a thread pool? What is pooling technology?

2022 The global C++ And system software technology conference | 3 month 11-12 Japan · Shanghai Click to learn more 》>>>

stay Java In language , There are two ways to improve the execution efficiency of the program , One is to use threads 、 The other is to use thread pool . And in a production environment , We usually use the latter . Why is this so ？ Today, let's talk about the advantages of thread pool , And pool technology and its application .

1. Pool technology

Pool technology refer to Prepare in advance Some resources , You can Reuse These pre prepared resources . There are two main advantages of pooling technology ： Prepare and reuse in advance . With Java Object creation in language as an example , When creating an object, you need to go through the following steps ：

1. according to new Parameter after identifier , Find the symbolic reference of the class in the constant pool ;
2. If no symbol application is found （ Class is not loaded ）, Load the class 、 analysis 、 Initialization etc. ;
3. The virtual machine allocates memory for objects in the heap , And initialize the allocated memory to 0, For object headers , Establish the corresponding description structure （ The long-running ： You need to find the free area in the heap , Modify memory allocation status, etc ）;
4. Invokes the initialization method of the object （ The long-running ： User's complex logic verification and other operations , Such as IO、 Whether the numerical calculation meets the regulations, etc ）.

From the above process, we can see , Creating a class requires complex and time-consuming operations , therefore We should try to reuse existing classes , To ensure the efficient operation of the program , Of course, it would be great if you could create these classes in advance , The realization of these functions depends on pooling technology .

2. Application of pooling Technology

Common pool technology applications are ： Thread pool 、 Memory pool 、 Database connection pool 、HttpClient Connection pool, etc , Next , Let's look at .

2.1 Thread pool

The principle of thread pool is very simple , Similar to the concept of buffer in the operating system . A number of threads will be started in the thread pool , These threads are all asleep . When the client has a new request , It will wake up a sleeping thread in the thread pool , Let it handle the request on the client side , When the request is processed , The thread is in sleep again . Thread pool can greatly improve the performance of programs . For example, there is a bank network center with a large concentration of provincial data , The number of concurrent client requests per second during peak periods exceeds 100, If you create a new thread for each client request , The cost of CPU Time and memory are amazing , If you take a possession 200 A pool of threads , That will save a lot of system resources , Make more CPU Time and memory are used to deal with practical business applications , Instead of frequent thread creation and destruction .

 

2.2 Memory pool

How to better manage application memory usage , At the same time, improve the frequency of memory use , This is a question worth pondering by every developer . Memory pool （Memory Pool） It provides a more feasible solution . During the creation of memory pool , Will allocate enough memory in advance , Form a preliminary memory pool . Then every time the user requests memory , It will return a piece of free memory in the memory pool , And set the flag of this memory as used . When the memory is used up and the memory is released , Not really free or delete The process of , It's the process of putting memory back into the memory pool , And in the process of putting it back, the flag should be set to idle . Last , When the application ends, the memory pool will be destroyed , Release each piece of memory in the memory pool . Advantages of memory pool ：

• Reduce memory fragmentation , This advantage can be seen in the process of creating memory pools , When we are creating a memory pool , All the allocated memory blocks are relatively regular memory blocks , Reduce memory fragmentation .
• Increased memory usage . This can be seen from the process of allocating and freeing memory . Each allocation and release is not to call the functions or operators provided by the system to operate the actual memory , Instead, reuse the memory in the memory pool .

Disadvantages of memory pool ： It's a waste of memory , Because to use the memory pool, you need to allocate a large piece of idle memory at the beginning , And not all of this memory is used .

2.3 Database connection pool

The basic idea of database connection pool is to store the database connection as an object in memory during system initialization , When the user needs to access the database , It's not about building a new connection , Instead, take an established free connection object out of the connection pool . After use , The user is not closing the connection , Instead, put the connection back into the connection pool , For next request access , And the establishment of these connections 、 Disconnection is managed by the connection pool itself . meanwhile , You can also set the parameters of the connection pool to control the initial number of connections in the connection pool 、 The upper and lower limits of connections and the maximum number of uses of each connection 、 Maximum free time, etc . Of course , You can also monitor the number of connections through the management mechanism of the connection pool itself 、 Usage, etc .

2.4 HttpClient Connection pool

HttpClient We often use it for HTTP Service access . In our project, there will be a function to obtain the execution status of tasks HttpClient, Ask once a second , Often Conection Reset abnormal . After analysis, we found that , The problem is HttpClient Each request will create a new connection , When the frequency of creating connections is higher than that of closing connections , It will lead to a large number of problems in the system TIME_CLOSED State connection , At this time, using connection pool to reuse connections can solve this problem .

3. Introduction to thread pool

Thread pool is a mode used by threads , It separates the concepts of thread and task , Use threads to perform tasks , It also provides a unified implementation method of thread management and task management , Avoid the performance overhead caused by frequent creation and destruction of threads .

4. Analysis of the advantages of thread pool

Compared with threads, thread pool , It does not require frequent creation and destruction of threads , Once the thread is created , By default, it will remain in the thread pool , When a mission comes , Then use these existing threads to perform tasks , As shown in the figure below ：

 

advantage 1： Reuse threads , Reduce resource consumption

When creating a thread, you need to open up the virtual machine stack 、 Native Method Stack 、 Memory space of private threads such as program counters , And when they are destroyed, they have to recycle these private space resources , As shown in the figure below ：

After the thread pool creates a thread, it will be placed in the thread pool , Therefore, compared with threads, thread pool , The first advantage is You can reuse threads 、 Reduce the consumption of system resources .

advantage 2： Improve response time

Thread pool is to reuse existing threads to perform tasks , Threads are created only when there are tasks , So compared to threads , Thread pool can respond to and execute tasks faster .

advantage 3： Control the number of threads and tasks

Thread pool provides more management functions , Here, the management function is mainly reflected in the following two aspects ：

1. Control the maximum concurrent number ： The thread pool can create a fixed number of threads , This avoids the problem of unlimited thread creation . When too many threads are created , Will cause the system to execute slowly , because CPU The number of cores is certain 、 The number of tasks that can be processed at the same time is also certain , When there are too many threads, it will cause malicious contention and frequent thread switching , This causes the program to execute slowly , Therefore, the appropriate number of threads is the key to high-performance operation .
2. Control the maximum number of tasks ： If there are unlimited tasks , When there is not enough memory , It will cause the program execution to report an error , The thread pool can control the maximum number of tasks , When the task exceeds a certain number , Will use the rejection strategy to deal with the extra tasks , So as to ensure the healthy operation of the system .

advantage 4： More enhancements

Thread pools provide more functionality than threads , Such as regular execution and periodic execution .

summary

Pooling technology refers to preparing some resources in advance , These pre prepared resources can be reused when needed . There are two main advantages of pooling technology ： Prepare and reuse in advance . Thread pool is a typical scenario of pooling technology , The main advantages of thread pool are 4 spot ：1. Reuse threads , Reduced resource consumption ;2. Improve response time ;3. It provides the ability to manage the number of threads and tasks ;4. More enhancements .

Last

If you think this article is of little help to you , Point a praise . Or you can join my development communication group ：1025263163 Learn from each other , We will have professional technical questions and answers

If you think this article is useful to you , Please give our open source project a little bit star:http://github.crmeb.net/u/defu Thank you for  ！

PHP Learning manual ：https://doc.crmeb.com
Technical exchange forum ：https://q.crmeb.com