之前寫過多線程累加計(jì)數(shù)，原理跟本篇類似，傳送門

累加計(jì)數(shù)比計(jì)算數(shù)組之和邏輯稍微簡單一點(diǎn)，如果對于這塊不熟悉的，可以先看下累加計(jì)數(shù)。

基本思想已經(jīng)在之前那篇文章里寫過了，這里就直接貼代碼了。

這里分別通過自己創(chuàng)建線程來實(shí)現(xiàn)功能，還有通過線程池來實(shí)現(xiàn)功能。思想都差不多。只是代碼寫法略有不同。僅供參考。

五個線程交替累加計(jì)算數(shù)組之和，這種方法其實(shí)不如單線程直接累加快，因?yàn)榻惶胬奂有枰耙粋€線程計(jì)算的結(jié)果。

package test; import java.util.concurrent.ExecutorService;import java.util.concurrent.Executors; public class FiveThreadCount { private int count=0; private int[] arr={1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28}; private int j=0; //定義一個任務(wù)，關(guān)鍵點(diǎn)所在 private class MyThread extends Thread{ @Override public void run() { super.run(); while(j<arr.length) { synchronized (MyThread.class) { if(j>=arr.length){ return; } count+=arr[j++]; try { Thread.sleep(100); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); } System.out.println(Thread.currentThread().getName()); } } } } //方法一 public void test1(){ for(int i=0;i<5;i++){ new MyThread().start(); }try { Thread.sleep(10000); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); }System.out.println(count); } //方法二 public void test2(){ Thread myThread=new MyThread(); for(int i=0;i<5;i++){ new Thread(myThread).start(); }try { Thread.sleep(10000); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); }System.out.println(count); } //方法一的線程池實(shí)現(xiàn)版 public void test3(){ ExecutorService service=Executors.newCachedThreadPool(); for(int i=0;i<5;i++){ service.execute(new MyThread()); }try { Thread.sleep(10000); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); }System.out.println(count); } //方法二的線程池實(shí)現(xiàn)版 public void test4(){ ExecutorService service=Executors.newCachedThreadPool(); Thread myThread=new MyThread(); for(int i=0;i<5;i++){ service.execute(myThread); }try { Thread.sleep(10000); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); }System.out.println(count); } }

上邊代碼中，用到了sleep方法的原因，sleep(100)是為了讓其他線程有時間執(zhí)行任務(wù)，如果不sleep的話，有可能一個線程就全部執(zhí)行完了。 最后的sleep（10000）是為了等所有線程執(zhí)行完后，打印最后的計(jì)算結(jié)果。

將數(shù)組分為5等分，讓每個線程計(jì)算自己負(fù)責(zé)的那份，并發(fā)計(jì)算，最后匯總結(jié)果。這種方式比代碼一速度會快些。因?yàn)榫€程獨(dú)立計(jì)算，不依賴其他線程的結(jié)果。最后幾個線程將總數(shù)累加即可。

使用Callable，F(xiàn)utureTask方式，來實(shí)現(xiàn)代碼：

package test; import java.util.concurrent.Callable;import java.util.concurrent.ExecutionException;import java.util.concurrent.ExecutorService;import java.util.concurrent.Executors;import java.util.concurrent.FutureTask; public class FiveThreadCount2 { private int[] arr={1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28}; private int total=0; public void test() throws InterruptedException, ExecutionException{ ExecutorService service=Executors.newFixedThreadPool(5); int length=arr.length; for(int j=0;j<length;j+=(length/5)){ FutureTask<Integer> task; if( (j+(length/5))<length){ task=new FutureTask<Integer>(new MyCallable(arr, j, j+(length/5))); }else{ task=new FutureTask<Integer>(new MyCallable(arr, j, length)); } service.execute(task); total+=task.get(); } service.shutdown(); System.out.println(total); } public class MyCallable implements Callable<Integer>{ int[] arr; int startIndex; int endIndex; public MyCallable(int[] arr,int startIndex,int endIndex){ this.arr=arr; this.startIndex=startIndex; this.endIndex=endIndex; } @Override public Integer call() throws Exception { int sum=0; for(int i=startIndex;i<endIndex;i++){ sum+=arr[i]; } System.out.println(Thread.currentThread().getName()); return sum; } } }

這個方式有一個缺點(diǎn)，看似5個線程異步執(zhí)行，其實(shí)是順序執(zhí)行，因?yàn)?task.get是要等待線程執(zhí)行完畢才會執(zhí)行下邊的代碼。所以效率不會高,可能換種寫法可以解決這個問題，這里就不深入研究。

通過java工具類CountDownLatch實(shí)現(xiàn)并發(fā)計(jì)算

package test; import java.util.concurrent.CountDownLatch; public class FiveThreadCount3 { private int[] arr={1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28}; private int total=0; public void test() throws InterruptedException{ int length=arr.length; CountDownLatch latch=new CountDownLatch(length%5==0?5:6); System.out.println(length); for(int j=0;j<length;j+=(length/5)){ MyThread task; if( (j+(length/5))<=length){ task=new MyThread(arr, j, j+(length/5), latch); }else{ task=new MyThread(arr, j, length, latch); } new Thread(task).start(); } latch.await(); System.out.println(total); } private class MyThread implements Runnable{ int[] arr; int startIndex; int endIndex; CountDownLatch latch; public MyThread(int[] arr,int startIndex,int endIndex,CountDownLatch latch){ this.arr=arr; this.startIndex=startIndex; this.endIndex=endIndex; this.latch=latch; } @Override public void run() { int sum=0; for(int i=startIndex;i<endIndex;i++){ sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } }}

對于CountDownLatch不熟悉的可以搜索下用法。

通過java工具類 CyclicBarrier實(shí)現(xiàn)并發(fā)計(jì)算。

package test; import java.util.concurrent.BrokenBarrierException;import java.util.concurrent.CyclicBarrier; public class FiveThreadCount1 { private int[] arr={1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28}; private int total=0; public void test() throws InterruptedException, BrokenBarrierException{ int length=arr.length; CyclicBarrier barrier=new CyclicBarrier((length%5==0?5:6)+1); System.out.println(length); for(int j=0;j<length;j+=(length/5)){ MyThread task; if( (j+(length/5))<=length){ task=new MyThread(arr, j, j+(length/5), barrier); }else{ task=new MyThread(arr, j, length, barrier); } new Thread(task).start(); } barrier.await(); System.out.println(total); } private class MyThread implements Runnable{ int[] arr; int startIndex; int endIndex; CyclicBarrier barrier; public MyThread(int[] arr,int startIndex,int endIndex,CyclicBarrier barrier){ this.arr=arr; this.startIndex=startIndex; this.endIndex=endIndex; this.barrier=barrier; } @Override public void run() { int sum=0; for(int i=startIndex;i<endIndex;i++){ sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } try { System.out.println(Thread.currentThread().getName()); barrier.await(); } catch (InterruptedException e) { // TODO Auto-generated catch block e.printStackTrace(); } catch (BrokenBarrierException e) { // TODO Auto-generated catch block e.printStackTrace(); } } }}總結(jié)

總體來說，代碼二的方式二、三，效率會高一點(diǎn)。以上代碼都是通過main方法調(diào)用示例代碼的test方法，輸出結(jié)果到控制臺。

到此這篇關(guān)于Java使用5個線程計(jì)算數(shù)組之和的文章就介紹到這了,更多相關(guān)Java 線程數(shù)組之和內(nèi)容請搜索好吧啦網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關(guān)文章希望大家以后多多支持好吧啦網(wǎng)！