如果对于Fork/Join不了解的话可以参考这里：http://www.infoq.com/cn/articles/fork-join-introduction、http://ifeve.com/fork-join-1/、http://www.iteye.com/topic/643724 。主要代码如下：
package com.zkn.fullstacktraining.seventh; import com.google.common.collect.Lists; import javafx.util.Pair; import java.io.IOException; import java.nio.file.Files; import java.nio.file.Path; import java.nio.file.Paths; import java.util.List; import java.util.Optional; import java.util.concurrent.ExecutionException; import java.util.concurrent.ForkJoinPool; import java.util.concurrent.Future; import java.util.concurrent.RecursiveTask; import java.util.stream.Collectors; import java.util.stream.IntStream; /** * Created by wb-zhangkenan on 2017/2/6. * 通过Fork/Join的方式来统计某个字符串在某个文件夹下的文件中出现的次数 */ public class SearchStringByForkJoin extends RecursiveTask<List<Pair<String, Integer>>> { //任务的阀值 private static final int THRESHOLD = 5; //任务区间的开始值 private int start; //任务区间的结束值 private int end; //所有的路径信息 private List<Path> allPaths; public SearchStringByForkJoin(int start, int end, List<Path> allPaths) { this.start = start; this.end = end; this.allPaths = allPaths; } public static void main(String[] args) { ForkJoinPool forkJoinPool = new ForkJoinPool(); try { List<Path> pathList = Files.walk(Paths.get("D:\\CUST\\WORK\\Exercises\\FullStackTraining\\src\\main\\java\\com\\zkn"), 4) .filter(file -> !Files.isDirectory(file) && file.toString().endsWith("java")) .collect(Collectors.toList()); SearchStringByForkJoin searchStringByForkJoin = new SearchStringByForkJoin(0, pathList.size(), pathList); Future<List<Pair<String, Integer>>> result = forkJoinPool.submit(searchStringByForkJoin); List<Pair<String, Integer>> pairList = result.get(); pairList.stream() .filter(pair -> pair.getValue() > 0) .sorted((s1, s2) -> Integer.compare(s2.getValue(), s1.getValue())) .forEach(pair -> System.out.println(String.format("%d次出现在%s文件中", pair.getValue(), pair.getKey().toString()))); forkJoinPool.shutdown(); } catch (IOException e) { e.printStackTrace(); } catch (InterruptedException e) { e.printStackTrace(); } catch (ExecutionException e) { e.printStackTrace(); } } /** * The main computation performed by this task. * @return the result of the computation */ @Override protected List<Pair<String, Integer>> compute() { List<Pair<String, Integer>> pairList = Lists.newArrayList(); boolean canCompute = end - start <= THRESHOLD; if (canCompute) { pairList = IntStream.range(start, end + 1).mapToObj(i -> { //到集合的结尾了 if (i == allPaths.size()) return new Pair<>("", 0); int sum = 0; try { Optional optional = Files.lines(allPaths.get(i)).map(file -> { int count = 0; int flag = file.indexOf("main"); if (flag >= 0) { do { count++; } while ((flag = file.indexOf("main", flag + 1)) > 0); } return count; }).reduce(Integer::sum); sum = optional.isPresent() ? (int) optional.get() : 0; } catch (IOException e) { e.printStackTrace(); } return new Pair<>(allPaths.get(i).toString(), sum); }).collect(Collectors.toList()); } else { //取中间值 int middle = (end + start) / 2; //创建Fork子任务 SearchStringByForkJoin leftFork = new SearchStringByForkJoin(start, middle, allPaths); //创建Fork子任务 SearchStringByForkJoin rightFork = new SearchStringByForkJoin(middle + 1, end, allPaths); //执行子任务 leftFork.fork(); rightFork.fork(); //合并子任务的结果集 pairList.addAll(leftFork.join()); pairList.addAll(rightFork.join()); } return pairList; } } 相关资源：敏捷开发V1.0.pptx