多线程批处理---Fork/Join框架

Fork/Join框架是Java7提供的一个用于并行执行任务的框架， 是一个把大任务分割成若干个小任务，最终汇总每个小任务结果后得到大任务结果的框架。

下列代码场景是启动5个线程同时处理存1w条记录的 list，最后把这个List遍历的的内容拼接起来。

package threadTest;

import java.util.ArrayList;
import java.util.List;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ForkJoinPool;
import java.util.concurrent.Future;
import java.util.concurrent.RecursiveTask;

/**
 * @Auther: xzs
 * @Date: 2018/5/31 10:58
 * @Description:
 */
public class ThreadList {
    private List<String> list;
    public static void main(String[] args) {
        System.out.println(new ThreadList().run().length());
    }

    private void init(){
        //模拟初始化1W条数据
        list = new ArrayList<String>();
        for (int i = 0; i < 10000; i++) {
            list.add(i + "");
        }
    }

    public String run(){
        init();
        ForkJoinPool pool = new ForkJoinPool(5);
        Task task = new Task(list);
        Future<String> result =  pool.submit(task);
        String str = "";
        try {
            str = result.get();
        } catch (InterruptedException e) {
            e.printStackTrace();
        } catch (ExecutionException e) {
            e.printStackTrace();
        }
        return str;
    }

    class Task extends RecursiveTask<String> {
        private int size = 2000;
        private List<String> data;
        public Task(List<String> data){
            this.data = data;
        }
        @Override
        protected String compute() {
            StringBuffer sb = new StringBuffer();
            if(data.size() <= size){
                System.out.println("******************************** size:" + data.size());
                for (String str : data) {
                    sb.append(str);
                }
            }else{
                //细分成小任务
                List<Task> tasks = new ArrayList<Task>();
                for (int index = 0; index * size < data.size(); index++) {
                    Task task;
                    if((index + 1) * size > data.size()){
                        task = new Task(data.subList(index * size, data.size()));
                    }else{
                        task = new Task(data.subList(index * size, (index + 1) * size));
                    }
                    task.fork();
                    tasks.add(task);
                }
                for (Task task : tasks) {
                    sb.append(task.join());
                }
            }

            return sb.toString();
        }

    }
}

----- 借鉴大神整理