Java

ThreadPoolExecutor

创建和管理线程池，减少内存消耗，提高执行效率。
任务执行过程
1.当线程数小于核心线程数时，创建线程；
2.当线程数大于等于核心线程数，且任务队列未满时，将任务放入任务队列；
3.当线程数大于等于核心线程数，且任务队列已满：

• 若线程数小于最大线程数，创建线程
• 若线程数等于最大线程数，执行拒绝策略(抛出异常、拒绝任务等)

任务执行API

public void execute(Runnable command)
public <T> Future<T> submit(Callable<T> task) //通过future对象的get方法获取任务的执行结果

详情参见：ThreadPoolExecutor
题外，Runtime.getRuntime().availableProcessors()获取当前设备CPU核心数。
此处提供一个Executors工具类，使用时直接@Autowired即可

@Bean
@Primary
public TaskExecutor taskExecutor() {
    ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
    executor.setCorePoolSize(corePoolSize);
    executor.setKeepAliveSeconds(keepAliveSeconds);
    executor.setMaxPoolSize(maxPoolSize);
    executor.setQueueCapacity(queueCapacity);
    ThreadPoolExecutor.CallerRunsPolicy callerRunsPolicy = new ThreadPoolExecutor.CallerRunsPolicy();
    executor.setRejectedExecutionHandler(callerRunsPolicy); //对拒绝task的处理策略
    executor.initialize();
    return executor;
}

拒绝策略支持自定义

private RejectedExecutionHandler getRejectedExecutionHandler(){
    return new RejectedExecutionHandler(){
        @Override
        public void rejectedExecution(Runnable r, ThreadPoolExecutor executor){
            if(!executor.isShutdown()){
                try{ executor.getQueue().put(r); }
                catch (Exception e){ e.getMessage(); }
            } }}; }