Flask接口签名sign原理与实例代码浅析
295
2022-10-24
Java使用5个线程计算数组之和
前言
之前写过多线程累加计数,原理跟本篇类似,传送门
累加计数比计算数组之和逻辑稍微简单一点,如果对于这块不熟悉的,可以先看下累加计数。
基本思想已经在之前那篇文章里写过了,这里就直接贴代码了。
这里分别通过自己创建线程来实现功能,还有通过线程池来实现功能。思想都差不多。只是代码写法略有不同。仅供参考。
代码一:
五个线程交替累加计算数组之和,这种方法其实不如单线程直接累加快,因为交替累加需要前一个线程计算的结果。
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;
//定义一个任务,关键点所在
private class MyThread extends Thread{
@Override
public void run() {
super.run();
while(j { 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); } //方法一的线程池实现版 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); } //方法二的线程池实现版 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); } } 上边代码中,用到了sleep方法的原因,sleep(100)是为了让其他线程有时间执行任务,如果不sleep的话,有可能一个线程就全部执行完了。 最后的sleep(10000)是为了等所有线程执行完后,打印最后的计算结果。 代码二: 将数组分为5等分,让每个线程计算自己负责的那份,并发计算,最后汇总结果。这种方式比代码一速度会快些。因为线程独立计算,不依赖其他线程的结果。最后几个线程将总数累加即可。 方式一: 使用Callable,FutureTask方式,来实现代码: 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 FutureTask if( (j+(length/5)) task=new FutureTask }else{ task=new FutureTask } service.execute(task); total+=task.get(); } service.shutdown(); System.out.println(total); } public class MyCallable implements Callable 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 sum+=arr[i]; } System.out.println(Thread.currentThread().getName()); return sum; } } } 这个方式有一个缺点,看似5个线程异步执行,其实是顺序执行,因为 task.get是要等待线程执行完毕才会执行下边的代码。所以效率不会高,可能换种写法可以解决这个问题,这里就不深入研究。 方式二: 通过java工具类CountDownLatch实现并发计算 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 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 sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } } } 对于CountDownLatch不熟悉的可以搜索下用法。 方式三: 通过java工具类 CyclicBarrier实现并发计算。 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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
{
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);
}
//方法一的线程池实现版
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);
}
//方法二的线程池实现版
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);
}
}
上边代码中,用到了sleep方法的原因,sleep(100)是为了让其他线程有时间执行任务,如果不sleep的话,有可能一个线程就全部执行完了。 最后的sleep(10000)是为了等所有线程执行完后,打印最后的计算结果。
代码二:
将数组分为5等分,让每个线程计算自己负责的那份,并发计算,最后汇总结果。这种方式比代码一速度会快些。因为线程独立计算,不依赖其他线程的结果。最后几个线程将总数累加即可。
方式一:
使用Callable,FutureTask方式,来实现代码:
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 FutureTask if( (j+(length/5)) task=new FutureTask }else{ task=new FutureTask } service.execute(task); total+=task.get(); } service.shutdown(); System.out.println(total); } public class MyCallable implements Callable 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 sum+=arr[i]; } System.out.println(Thread.currentThread().getName()); return sum; } } } 这个方式有一个缺点,看似5个线程异步执行,其实是顺序执行,因为 task.get是要等待线程执行完毕才会执行下边的代码。所以效率不会高,可能换种写法可以解决这个问题,这里就不深入研究。 方式二: 通过java工具类CountDownLatch实现并发计算 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 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 sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } } } 对于CountDownLatch不熟悉的可以搜索下用法。 方式三: 通过java工具类 CyclicBarrier实现并发计算。 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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
FutureTask
if( (j+(length/5)) task=new FutureTask }else{ task=new FutureTask } service.execute(task); total+=task.get(); } service.shutdown(); System.out.println(total); } public class MyCallable implements Callable 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 sum+=arr[i]; } System.out.println(Thread.currentThread().getName()); return sum; } } } 这个方式有一个缺点,看似5个线程异步执行,其实是顺序执行,因为 task.get是要等待线程执行完毕才会执行下边的代码。所以效率不会高,可能换种写法可以解决这个问题,这里就不深入研究。 方式二: 通过java工具类CountDownLatch实现并发计算 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 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 sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } } } 对于CountDownLatch不熟悉的可以搜索下用法。 方式三: 通过java工具类 CyclicBarrier实现并发计算。 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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
task=new FutureTask
}else{
task=new FutureTask
}
service.execute(task);
total+=task.get();
}
service.shutdown();
System.out.println(total);
}
public class MyCallable implements Callable
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 sum+=arr[i]; } System.out.println(Thread.currentThread().getName()); return sum; } } } 这个方式有一个缺点,看似5个线程异步执行,其实是顺序执行,因为 task.get是要等待线程执行完毕才会执行下边的代码。所以效率不会高,可能换种写法可以解决这个问题,这里就不深入研究。 方式二: 通过java工具类CountDownLatch实现并发计算 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 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 sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } } } 对于CountDownLatch不熟悉的可以搜索下用法。 方式三: 通过java工具类 CyclicBarrier实现并发计算。 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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
sum+=arr[i];
}
System.out.println(Thread.currentThread().getName());
return sum;
}
}
}
这个方式有一个缺点,看似5个线程异步执行,其实是顺序执行,因为 task.get是要等待线程执行完毕才会执行下边的代码。所以效率不会高,可能换种写法可以解决这个问题,这里就不深入研究。
方式二:
通过java工具类CountDownLatch实现并发计算
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 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 sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } } } 对于CountDownLatch不熟悉的可以搜索下用法。 方式三: 通过java工具类 CyclicBarrier实现并发计算。 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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
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 sum+=arr[i]; } synchronized (MyThread.class) { total+=sum; } System.out.println(Thread.currentThread().getName()); latch.countDown(); } } } 对于CountDownLatch不熟悉的可以搜索下用法。 方式三: 通过java工具类 CyclicBarrier实现并发计算。 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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
sum+=arr[i];
}
synchronized (MyThread.class) {
total+=sum;
}
System.out.println(Thread.currentThread().getName());
latch.countDown();
}
}
}
对于CountDownLatch不熟悉的可以搜索下用法。
方式三:
通过java工具类 CyclicBarrier实现并发计算。
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 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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
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 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(); } } } } 总结 总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
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();
}
}
}
}
总结
总体来说,代码二的方式二、三,效率会高一点。以上代码都是通过main方法调用示例代码的test方法,输出结果到控制台。
版权声明:本文内容由网络用户投稿,版权归原作者所有,本站不拥有其著作权,亦不承担相应法律责任。如果您发现本站中有涉嫌抄袭或描述失实的内容,请联系我们jiasou666@gmail.com 处理,核实后本网站将在24小时内删除侵权内容。
发表评论
暂时没有评论,来抢沙发吧~