2 Commits

Author SHA1 Message Date
wonder 9458d09f69 🔄Update: 创建线程池 2025-11-10 21:32:43 +08:00
wonder b2ab34c344 📝Docs: 更新 README 2025-11-10 19:31:05 +08:00
5 changed files with 29 additions and 370 deletions
+1 -262
View File
@@ -1,262 +1 @@
# 使用 `@Async` 实现线程池
```mermaid
graph TD
A[调用者调用异步方法] --> B[Spring AOP 代理拦截方法调用]
B --> C{方法是否为 @Async?}
C -->|是| D[封装为 FutureTask 或 Runnable]
C -->|否| E[按正常方式执行]
D --> F[提交到线程池 TaskExecutor]
F --> G[线程池分配一个线程]
G --> H[JVM 创建线程并执行任务]
H --> I[任务执行完毕,返回结果]
I --> J[调用者通过 Future 获取结果]
```
## 前置知识
### **`@Async`**
`@Async` 是 Spring 框架提供的一个注解,用于支持**异步方法调用**。它允许我们标记一个方法为异步方法,这样该方法将在**单独的线程**中执行,而不是阻塞当前线程。
#### 优点:
- 提高应用程序的响应速度
- 避免长时间运行的操作阻塞主线程
- 适用于处理耗时操作、并行任务、异步通知等场景
#### 使用场景:
- 异步发送邮件、短信
- 异步日志记录
- 异步计算任务
- 异步调用第三方服务
---
## 一、开启异步支持
要使用 `@Async` 注解,必须在 Spring 应用中开启**异步支持**。可以通过在主类或配置类上添加 `@EnableAsync` 注解来实现。
```java
@Configuration
@EnableAsync
public class AppConfig {
}
```
或者在 Spring Boot 主类上添加:
```java
@SpringBootApplication
@EnableAsync
public class Application {
public static void main(String[] args) {
SpringApplication.run(Application.class, args);
}
}
```
---
## 二、配置线程池(推荐)
Spring 提供了多种方式来配置异步任务执行的线程池,推荐使用 `TaskExecutor` 来自定义线程池,以控制并发行为。
### 1. 使用 `@Bean` 配置自定义线程池
```java
@Configuration
@EnableAsync
public class AsyncConfig {
@Bean(name = "customTaskExecutor")
public Executor customTaskExecutor() {
ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
executor.setCorePoolSize(5);
executor.setMaxPoolSize(10);
executor.setQueueCapacity(100);
executor.setThreadNamePrefix("Custom-Async-");
executor.initialize();
return executor;
}
}
```
### 2. 指定使用哪个线程池
可以通过 `@Async("customTaskExecutor")` 指定使用自定义的线程池。
```java
@Service
public class MyService {
@Async("customTaskExecutor")
public void asyncMethod() {
// 异步执行的操作
}
}
```
---
## 三、使用默认线程池
如果未配置自定义线程池,Spring 会使用默认的 `SimpleAsyncTaskExecutor`,但这个线程池不推荐用于生产环境,因为它**每次都会创建新线程**,容易引发资源耗尽问题。
可以使用 `@Async` 并不指定线程池名称,Spring 会自动使用默认的线程池(需确保 `@EnableAsync` 已开启):
```java
@Service
public class MyService {
@Async
public void defaultAsyncMethod() {
// 默认线程池执行
}
}
```
---
## 四、异步方法调用注意事项
- `@Async` 注解的方法不能和调用它的方法在同一个类中,否则不会生效。因为 Spring 的 AOP 机制需要通过代理来实现异步调用,若在同一个类中调用,代理无法生效。
- 异步方法不能返回 `void`,否则无法捕获异常。**建议返回 `Future<T>` 或 `CompletableFuture<T>`**。
- 异步方法的异常处理:如果希望捕获异步方法中的异常,可以在方法中使用 `try-catch`,或者通过 `Future` 返回异常信息。
- 如果异步方法中涉及事务控制,需注意事务的传播机制。
---
## 五、示例代码
### 示例 1:简单异步方法
```java
@Service
public class AsyncService {
@Async
public void sendNotification(String message) {
System.out.println("Sending notification: " + message);
// 模拟耗时操作
try {
Thread.sleep(1000);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
```
### 示例 2:使用自定义线程池并返回 Future
```java
@Service
public class AsyncService {
@Async("customTaskExecutor")
public Future<String> asyncTask(String input) {
try {
Thread.sleep(2000);
return new AsyncResult<>("Processed: " + input);
} catch (InterruptedException e) {
return new AsyncResult<>("Error occurred");
}
}
}
```
在调用处使用:
```java
@Autowired
private AsyncService asyncService;
public void callAsyncTask() {
Future<String> result = asyncService.asyncTask("test");
try {
System.out.println(result.get());
} catch (Exception e) {
e.printStackTrace();
}
}
```
---
## 六、线程池配置属性说明
| 属性名 | 描述 |
|-----------------------|--------------------------------------------|
| `corePoolSize` | 核心线程数(常驻线程数量) |
| `maxPoolSize` | 最大线程数(允许创建的最大线程数) |
| `queueCapacity` | 任务队列容量(缓存未被立即执行的任务) |
| `keepAliveSeconds` | 非核心线程的空闲时间(超过后会被销毁) |
| `threadNamePrefix` | 线程名称前缀(用于调试和日志区分) |
| `allowCoreThreadTimeOut` | 是否允许核心线程超时(默认 false) |
---
## 七、异步方法异常处理
由于异步方法在独立线程中执行,不能直接抛出异常。可使用 `Future` 或 `CompletableFuture` 来捕获和处理异常。
### 使用 `Future`
```java
@Async
public Future<String> asyncMethod() {
try {
// …
return new AsyncResult<>("Success");
} catch (Exception e) {
return new AsyncResult<>("Error: " + e.getMessage());
}
}
```
### 使用 `CompletableFuture`
```java
@Async
public CompletableFuture<String> asyncMethod() {
return CompletableFuture.supplyAsync(() -> {
try {
// …
return "Success";
} catch (Exception e) {
throw new RuntimeException(e);
}
});
}
```
---
## 八、常见问题与排查
| 问题 | 解决方案 |
|------|----------|
| `@Async` 方法未执行 | 检查是否开启 `@EnableAsync`,是否在不同类中调用 |
| 线程池配置无效 | 确保 `@Bean` 配置正确,且 `@Async` 指定了正确的 bean 名称 |
| 线程池资源耗尽 | 调整线程池大小和队列容量,增加 `queueCapacity` |
| 异常未被捕获 | 使用 `Future` 或 `CompletableFuture` 返回结果,避免 `void` 返回 |
---
## 九、相关类与接口
- `org.springframework.scheduling.annotation.Async`
- `org.springframework.scheduling.annotation.EnableAsync`
- `org.springframework.core.task.AsyncTaskExecutor`
- `org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor`
- `org.springframework.util.concurrent.AsyncTask`
---
## 十、最佳实践
- 为不同业务场景配置不同的线程池
- 使用 `CompletableFuture` 来处理异步任务中的返回结果和异常
- 避免在异步方法中使用共享变量,防止线程安全问题
- 控制线程池的大小和任务队列,防止资源浪费或阻塞
- 结合 `@Scheduled` 注解实现定时异步任务
> 注意:Java ExecutorService 使用的默认等待队列是无限大小的 `LinkedBlockingQueue`,容易 OOM,不推荐使用
@@ -1,11 +1,34 @@
package cn.hezhaohui.threadpool;
import org.springframework.boot.SpringApplication;
import org.springframework.boot.autoconfigure.SpringBootApplication;
import lombok.extern.slf4j.Slf4j;
@SpringBootApplication
import java.util.Random;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.TimeUnit;
@Slf4j
public class ThreadPoolApplication {
public static void main(String[] args) {
SpringApplication.run(ThreadPoolApplication.class, args);
public static void main(String[] args) throws InterruptedException {
ExecutorService executorService = Executors.newFixedThreadPool(2);
for (int i = 0; i < 100000000; i++) {
executorService.execute(() -> {
StringBuilder builder = new StringBuilder();
int random = new Random().nextInt(1000);
for (int j = 1; j < random; j++) {
builder.append(j);
}
try {
TimeUnit.HOURS.sleep(1);
} catch (InterruptedException e) {
log.error("[Error]");
}
log.info(builder.toString());
});
}
executorService.shutdown();
executorService.awaitTermination(1, TimeUnit.HOURS);
}
}
@@ -1,29 +0,0 @@
package cn.hezhaohui.threadpool.conifg;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.scheduling.annotation.EnableAsync;
import org.springframework.scheduling.concurrent.ThreadPoolTaskExecutor;
import java.util.concurrent.Executor;
import java.util.concurrent.ThreadPoolExecutor;
/**
* 异步配置类
*/
@Configuration
@EnableAsync
public class AsyncConfig {
@Bean(name = "customThreadPool")
public Executor customThreadPool() {
ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
executor.setCorePoolSize(5);
executor.setMaxPoolSize(10);
executor.setQueueCapacity(100);
executor.setThreadNamePrefix("Custom-Async-");
executor.initialize();
executor.setRejectedExecutionHandler(new ThreadPoolExecutor.DiscardOldestPolicy());
return executor;
}
}
@@ -1,45 +0,0 @@
package cn.hezhaohui.threadpool.service;
import lombok.extern.slf4j.Slf4j;
import org.springframework.scheduling.annotation.Async;
import org.springframework.stereotype.Service;
import java.util.Random;
import java.util.UUID;
import java.util.concurrent.CountDownLatch;
@Service
@Slf4j
public class AsyncService {
@Async("customThreadPool")
public void asyncPrint() {
String id = UUID.randomUUID().toString().substring(0, 5);
int sleepSeconds = new Random().nextInt(15);
log.info("[Thread {}] Called, waiting {} seconds", id, sleepSeconds);
try {
Thread.sleep(sleepSeconds * 1000L);
log.info("[Thread {}] Wait completed after {} seconds", id, sleepSeconds);
} catch (InterruptedException e) {
log.error("[Thread {}] Interrupted during wait", id, e);
Thread.currentThread().interrupt();
}
}
@Async("customThreadPool")
public void asyncPrint(CountDownLatch latch) {
String id = UUID.randomUUID().toString().substring(0, 5);
int sleepSeconds = new Random().nextInt(15);
log.info("[Thread {}] ⚠️ Called, waiting {} seconds", id, sleepSeconds);
try {
Thread.sleep(sleepSeconds * 1000L);
log.info("[Thread {}] ✅ Wait completed after {} seconds", id, sleepSeconds);
} catch (InterruptedException e) {
log.error("[Thread {}] Interrupted during wait", id, e);
Thread.currentThread().interrupt();
} finally {
latch.countDown();
log.info("⭐ {}", latch.getCount());
}
}
}
@@ -1,29 +0,0 @@
package cn.hezhaohui.threadpool.service;
import jakarta.annotation.Resource;
import lombok.extern.slf4j.Slf4j;
import org.junit.jupiter.api.Test;
import org.springframework.boot.test.context.SpringBootTest;
import java.util.concurrent.CountDownLatch;
import static org.junit.jupiter.api.Assertions.*;
@SpringBootTest
@Slf4j
class AsyncServiceTest {
@Resource
private AsyncService asyncService;
@Test
void asyncPrint() throws InterruptedException {
CountDownLatch latch = new CountDownLatch(100);
for (int i = 0; i < 100; i++) {
asyncService.asyncPrint(latch);
}
log.info("\n[+++ All Tasks Commited +++]\n");
latch.await();
log.info("\n[=== All Tasks Completed ===]\n");
}
}