我们常用ThreadPoolExecutor提供的线程池服务,springboot框架提供了@Async注解,帮助我们更方便的将业务逻辑提交到线程池中异步执行,今天我们就来实战体验这个线程池服务;
本文地址:http://blog.csdn.net/boling_cavalry/article/details/79120268
实战环境
- windowns10;
- jdk1.8;
- springboot 1.5.9.RELEASE;
- 开发工具:IntelliJ IDEA;
实战源码
本次实战的源码可以在我的GitHub下载,地址:git@github.com:zq2599/blog_demos.git,项目主页:https://github.com/zq2599/blog_demos
这里面有多个工程,本次用到的工程为threadpooldemoserver,如下图红框所示:
实战步骤梳理
本次实战的步骤如下:
1. 创建springboot工程;
2. 创建Service层的接口和实现;
3. 创建controller,开发一个http服务接口,里面会调用service层的服务;
4. 创建线程池的配置;
5. 将Service层的服务异步化,这样每次调用都会都被提交到线程池异步执行;
6. 扩展ThreadPoolTaskExecutor,在提交任务到线程池的时候可以观察到当前线程池的情况;
创建springboot工程
用IntelliJ IDEA创建一个springboot的web工程threadpooldemoserver,pom.xml内容如下:
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<groupId>com.bolingcavalry</groupId>
<artifactId>threadpooldemoserver</artifactId>
<version>0.0.1-SNAPSHOT</version>
<packaging>jar</packaging>
<name>threadpooldemoserver</name>
<description>Demo project for Spring Boot</description>
<parent>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-parent</artifactId>
<version>1.5.9.RELEASE</version>
<relativePath/> <!-- lookup parent from repository -->
</parent>
<properties>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding>
<java.version>1.8</java.version>
</properties>
<dependencies>
<dependency>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-starter-web</artifactId>
</dependency>
</dependencies>
<build>
<plugins>
<plugin>
<groupId>org.springframework.boot</groupId>
<artifactId>spring-boot-maven-plugin</artifactId>
</plugin>
</plugins>
</build>
</project>
创建Service层的接口和实现
创建一个service层的接口AsyncService,如下:
public interface AsyncService {
/**
* 执行异步任务
*/
void executeAsync();
}
对应的AsyncServiceImpl,实现如下:
@Service
public class AsyncServiceImpl implements AsyncService {
private static final Logger logger = LoggerFactory.getLogger(AsyncServiceImpl.class);
@Override
public void executeAsync() {
logger.info("start executeAsync");
try{
Thread.sleep(1000);
}catch(Exception e){
e.printStackTrace();
}
logger.info("end executeAsync");
}
}
这个方法做的事情很简单:sleep了一秒钟;
创建controller
创建一个controller为Hello,里面定义一个http接口,做的事情是调用Service层的服务,如下:
@RestController
public class Hello {
private static final Logger logger = LoggerFactory.getLogger(Hello.class);
@Autowired
private AsyncService asyncService;
@RequestMapping("/")
public String submit(){
logger.info("start submit");
//调用service层的任务
asyncService.executeAsync();
logger.info("end submit");
return "success";
}
}
至此,我们已经做好了一个http请求的服务,里面做的事情其实是同步的,接下来我们就开始配置springboot的线程池服务,将service层做的事情都提交到线程池中去处理;
springboot的线程池配置
创建一个配置类ExecutorConfig,用来定义如何创建一个ThreadPoolTaskExecutor,要使用@Configuration和@EnableAsync这两个注解,表示这是个配置类,并且是线程池的配置类,如下所示:
@Configuration
@EnableAsync
public class ExecutorConfig {
private static final Logger logger = LoggerFactory.getLogger(ExecutorConfig.class);
@Bean
public Executor asyncServiceExecutor() {
logger.info("start asyncServiceExecutor");
ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
//配置核心线程数
executor.setCorePoolSize(5);
//配置最大线程数
executor.setMaxPoolSize(5);
//配置队列大小
executor.setQueueCapacity(99999);
//配置线程池中的线程的名称前缀
executor.setThreadNamePrefix("async-service-");
// rejection-policy:当pool已经达到max size的时候,如何处理新任务
// CALLER_RUNS:不在新线程中执行任务,而是有调用者所在的线程来执行
executor.setRejectedExecutionHandler(new ThreadPoolExecutor.CallerRunsPolicy());
//执行初始化
executor.initialize();
return executor;
}
}
注意,上面的方法名称为asyncServiceExecutor,稍后马上用到;
将Service层的服务异步化
打开AsyncServiceImpl.java,在executeAsync方法上增加注解@Async(“asyncServiceExecutor”),asyncServiceExecutor是前面ExecutorConfig.java中的方法名,表明executeAsync方法进入的线程池是asyncServiceExecutor方法创建的,如下:
@Override
@Async("asyncServiceExecutor")
public void executeAsync() {
logger.info("start executeAsync");
try{
Thread.sleep(1000);
}catch(Exception e){
e.printStackTrace();
}
logger.info("end executeAsync");
}
验证效果
- 将这个springboot运行起来(pom.xml所在文件夹下执行mvn spring-boot:run);
- 在浏览器输入:http://localhost:8080;
- 在浏览器用F5按钮快速多刷新几次;
- 在springboot的控制台看见日志如下:
2018-01-21 22:43:18.630 INFO 14824 --- [nio-8080-exec-8] c.b.t.controller.Hello : start submit
2018-01-21 22:43:18.630 INFO 14824 --- [nio-8080-exec-8] c.b.t.controller.Hello : end submit
2018-01-21 22:43:18.929 INFO 14824 --- [async-service-1] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:18.930 INFO 14824 --- [async-service-1] c.b.t.service.impl.AsyncServiceImpl : start executeAsync
2018-01-21 22:43:19.005 INFO 14824 --- [async-service-2] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:19.006 INFO 14824 --- [async-service-2] c.b.t.service.impl.AsyncServiceImpl : start executeAsync
2018-01-21 22:43:19.175 INFO 14824 --- [async-service-3] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:19.175 INFO 14824 --- [async-service-3] c.b.t.service.impl.AsyncServiceImpl : start executeAsync
2018-01-21 22:43:19.326 INFO 14824 --- [async-service-4] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:19.495 INFO 14824 --- [async-service-5] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:19.930 INFO 14824 --- [async-service-1] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:20.006 INFO 14824 --- [async-service-2] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 22:43:20.191 INFO 14824 --- [async-service-3] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
如上日志所示,我们可以看到controller的执行线程是”nio-8080-exec-8”,这是tomcat的执行线程,而service层的日志显示线程名为“async-service-1”,显然已经在我们配置的线程池中执行了,并且每次请求中,controller的起始和结束日志都是连续打印的,表明每次请求都快速响应了,而耗时的操作都留给线程池中的线程去异步执行;
扩展ThreadPoolTaskExecutor
虽然我们已经用上了线程池,但是还不清楚线程池当时的情况,有多少线程在执行,多少在队列中等待呢?这里我创建了一个ThreadPoolTaskExecutor的子类,在每次提交线程的时候都会将当前线程池的运行状况打印出来,代码如下:
public class VisiableThreadPoolTaskExecutor extends ThreadPoolTaskExecutor {
private static final Logger logger = LoggerFactory.getLogger(VisiableThreadPoolTaskExecutor.class);
private void showThreadPoolInfo(String prefix){
ThreadPoolExecutor threadPoolExecutor = getThreadPoolExecutor();
if(null==threadPoolExecutor){
return;
}
logger.info("{}, {},taskCount [{}], completedTaskCount [{}], activeCount [{}], queueSize [{}]",
this.getThreadNamePrefix(),
prefix,
threadPoolExecutor.getTaskCount(),
threadPoolExecutor.getCompletedTaskCount(),
threadPoolExecutor.getActiveCount(),
threadPoolExecutor.getQueue().size());
}
@Override
public void execute(Runnable task) {
showThreadPoolInfo("1. do execute");
super.execute(task);
}
@Override
public void execute(Runnable task, long startTimeout) {
showThreadPoolInfo("2. do execute");
super.execute(task, startTimeout);
}
@Override
public Future<?> submit(Runnable task) {
showThreadPoolInfo("1. do submit");
return super.submit(task);
}
@Override
public <T> Future<T> submit(Callable<T> task) {
showThreadPoolInfo("2. do submit");
return super.submit(task);
}
@Override
public ListenableFuture<?> submitListenable(Runnable task) {
showThreadPoolInfo("1. do submitListenable");
return super.submitListenable(task);
}
@Override
public <T> ListenableFuture<T> submitListenable(Callable<T> task) {
showThreadPoolInfo("2. do submitListenable");
return super.submitListenable(task);
}
}
如上所示,showThreadPoolInfo方法中将任务总数、已完成数、活跃线程数,队列大小都打印出来了,然后Override了父类的execute、submit等方法,在里面调用showThreadPoolInfo方法,这样每次有任务被提交到线程池的时候,都会将当前线程池的基本情况打印到日志中;
修改ExecutorConfig.java的asyncServiceExecutor方法,将ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor()改为ThreadPoolTaskExecutor executor = new VisiableThreadPoolTaskExecutor(),如下所示:
@Bean
public Executor asyncServiceExecutor() {
logger.info("start asyncServiceExecutor");
//使用VisiableThreadPoolTaskExecutor
ThreadPoolTaskExecutor executor = new VisiableThreadPoolTaskExecutor();
//配置核心线程数
executor.setCorePoolSize(5);
//配置最大线程数
executor.setMaxPoolSize(5);
//配置队列大小
executor.setQueueCapacity(99999);
//配置线程池中的线程的名称前缀
executor.setThreadNamePrefix("async-service-");
// rejection-policy:当pool已经达到max size的时候,如何处理新任务
// CALLER_RUNS:不在新线程中执行任务,而是有调用者所在的线程来执行
executor.setRejectedExecutionHandler(new ThreadPoolExecutor.CallerRunsPolicy());
//执行初始化
executor.initialize();
return executor;
}
再次启动该工程,再浏览器反复刷新http://localhost:8080,看到的日志如下:
2018-01-21 23:04:56.113 INFO 15580 --- [nio-8080-exec-1] c.b.t.e.VisiableThreadPoolTaskExecutor : async-service-, 2. do submit,taskCount [99], completedTaskCount [85], activeCount [5], queueSize [9]
2018-01-21 23:04:56.113 INFO 15580 --- [nio-8080-exec-1] c.b.t.controller.Hello : end submit
2018-01-21 23:04:56.225 INFO 15580 --- [async-service-1] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 23:04:56.225 INFO 15580 --- [async-service-1] c.b.t.service.impl.AsyncServiceImpl : start executeAsync
2018-01-21 23:04:56.240 INFO 15580 --- [nio-8080-exec-2] c.b.t.controller.Hello : start submit
2018-01-21 23:04:56.240 INFO 15580 --- [nio-8080-exec-2] c.b.t.e.VisiableThreadPoolTaskExecutor : async-service-, 2. do submit,taskCount [100], completedTaskCount [86], activeCount [5], queueSize [9]
2018-01-21 23:04:56.240 INFO 15580 --- [nio-8080-exec-2] c.b.t.controller.Hello : end submit
2018-01-21 23:04:56.298 INFO 15580 --- [async-service-2] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 23:04:56.298 INFO 15580 --- [async-service-2] c.b.t.service.impl.AsyncServiceImpl : start executeAsync
2018-01-21 23:04:56.372 INFO 15580 --- [nio-8080-exec-3] c.b.t.controller.Hello : start submit
2018-01-21 23:04:56.373 INFO 15580 --- [nio-8080-exec-3] c.b.t.e.VisiableThreadPoolTaskExecutor : async-service-, 2. do submit,taskCount [101], completedTaskCount [87], activeCount [5], queueSize [9]
2018-01-21 23:04:56.373 INFO 15580 --- [nio-8080-exec-3] c.b.t.controller.Hello : end submit
2018-01-21 23:04:56.444 INFO 15580 --- [async-service-3] c.b.t.service.impl.AsyncServiceImpl : end executeAsync
2018-01-21 23:04:56.445 INFO 15580 --- [async-service-3] c.b.t.service.impl.AsyncServiceImpl : start executeAsync
注意这一行日志:2. do submit,taskCount [101], completedTaskCount [87], activeCount [5], queueSize [9]
这说明提交任务到线程池的时候,调用的是submit(Callable task)这个方法,当前已经提交了101个任务,完成了87个,当前有5个线程在处理任务,还剩9个任务在队列中等待,线程池的基本情况一路了然;
至此,springboot线程池服务的实战就完成了,希望能帮您在工程中快速实现异步服务;