springboot创建线程池两种方式
1.使用static代码块创建
这样的方式创建的好处是当代码用到线程池的时候才会初始化核心线程数
具体代码如下：
public class HttpApiThreadPool {
static int cpuNums = Runtime.getRuntime().availableProcessors();
private static int corePoolSize = 10;
private static int maximumPoolSize = cpuNums * 5;
public static ExecutorService httpApiThreadPool = null;
static{
System.out.println("创建线程数:"+corePoolSize+",最大线程数:"+maximumPoolSize);
//建立10个核心线程，线程请求个数超过20，则进入队列等待
httpApiThreadPool = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, 0L,
TimeUnit.MILLISECONDS, new ArrayBlockingQueue<Runnable>(100),new ThreadFactoryBuilder().setNameFormat("PROS-%d").build());
}
}使用方法
public static void main(String[] args) {
HttpApiThreadPool.httpApiThreadPool.execute(()->System.out.println("测试"));
}注意：
1.不能使用Executors的方法创建线程池，这个是大量的生产事故得出来的结论
2.maximumPoolSize本程序使用的是cup数的5倍，你可以看你实际情况用
3.new ThreadFactoryBuilder().setNameFormat("PROS-%d").build() 给每个线程已名字，可以方便调试
2.使用@Configuration @bean注解，程序启动时创建
@Configuration
public class TreadPoolConfig {
private Logger logger = LoggerFactory.getLogger(TreadPoolConfig.class);
int cpuNums = Runtime.getRuntime().availableProcessors();
private int corePoolSize = 10;
private int maximumPoolSize = cpuNums * 5;
@Bean(value = "httpApiThreadPool")
public ExecutorService buildHttpApiThreadPool(){
logger.info("TreadPoolConfig创建线程数:"+corePoolSize+",最大线程数:"+maximumPoolSize);
ExecutorService pool = new ThreadPoolExecutor(corePoolSize, maximumPoolSize, 0L,
TimeUnit.MILLISECONDS, new ArrayBlockingQueue<Runnable>(100),new ThreadFactoryBuilder().setNameFormat("PROS-%d").build());
return pool ;
}
}使用方法
//注入
@Resource
private TreadPoolConfig treadPoolConfig;
//调用
public void test() {
treadPoolConfig.buildHttpApiThreadPool().execute(()->System.out.println("tre"));
}现在两种线程池的创建方法已经介绍完了。
springboot如何开启线程池
定义线程池
定义的位置，要在启动类的子包或者同级目录中
import lombok.Data;
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.ThreadPoolExecutor;
@Data
@Configuration
@EnableAsync //开启异步请求
public class ThreadPoolConfig {
private static final int corePoolSize = 10; // 核心线程数（默认线程数）
private static final int maxPoolSize = 100; // 最大线程数
private static final int keepAliveTime = 10; // 允许线程空闲时间（单位：默认为秒）
private static final int queueCapacity = 500; // 缓冲队列数
@Bean("taskExecutor")
public ThreadPoolTaskExecutor taskExecutor(){
ThreadPoolTaskExecutor pool = new ThreadPoolTaskExecutor();
pool.setThreadNamePrefix("--------------全局线程池-----------------");
pool.setCorePoolSize(corePoolSize);
pool.setMaxPoolSize(maxPoolSize);
pool.setKeepAliveSeconds(keepAliveTime);
pool.setQueueCapacity(queueCapacity);
// 直接在execute方法的调用线程中运行
pool.setRejectedExecutionHandler(new ThreadPoolExecutor.CallerRunsPolicy());
// 初始化
pool.initialize();
return pool;
}
}使用
直接在需要异步执行的方法上加注解
@Async("taskExecutor")以上为个人经验，希望能给大家一个参考，也希望大家多多支持编程网。
