我有一个简单的类，它使用ExecutorService并使用蒙特卡罗方法计算PI。
public class MonteCarlo {
public static void main(String[] args) throws ExecutionException, InterruptedException {
long start = System.currentTimeMillis();
ExecutorService executorService = Executors.newFixedThreadPool(2);
long numberOfWorkers = 1000000;
double x = 0.0;
double y = 0.0;
double radius = 5.0;
long numberOfAllShoots = 0;
Set<Future<Integer>> futures = new HashSet<>();
for(int i=0;i<numberOfWorkers;i++) {
long shootNumber = 500;
numberOfAllShoots+=shootNumber;
Callable callable = new Shoot(x,y,radius,shootNumber);
Future<Integer> future = executorService.submit(callable);
futures.add(future);
}
int numberOfCorrectShoots = 0;
for (Future<Integer> future : futures) {
numberOfCorrectShoots += future.get();
}
executorService.shutdown();
double pi = 4 * ((double)numberOfCorrectShoots / numberOfAllShoots );
System.out.printf("PI is approximated to " + pi);
long end = System.currentTimeMillis();
System.out.println("\n"+(end - start) + " miliseconds ");
}
}
但问题是当我将newFixedThreadPool参数设置为100或2时，我的运行时间几乎相同。使用newSingleThreadExecutor的结果相同。 这段代码有什么问题？我期望使用newFixedThreadPool会比单线程
快得多修改
拍摄课程
public class Shoot implements Callable {
private final double x;
private final double y;
private final double radius;
private final long numberOfShoots;
public Shoot(double x, double y, double radius, long numberOfShoots) {
this.x = x;
this.y = y;
this.radius = radius;
this.numberOfShoots = numberOfShoots;
}
@Override
public Integer call() {
int numberOfCorrectShoot = 0;
for(int i=0;i<numberOfShoots;i++) {
double t = randomDoubleFromRange(x-radius,x+radius);
double u = randomDoubleFromRange(y-radius,y+radius);
if((x-t)*(x-t) + (y-u)*(y-u) <= radius * radius) {
numberOfCorrectShoot++;
}
}
return numberOfCorrectShoot;
}
private double randomDoubleFromRange(double rangeMin, double rangeMax) {
Random random = new Random();
return rangeMin + (rangeMax - rangeMin) * random.nextDouble();
}
}
nproc命令返回2
我尝试过numberOfWorkers = 400和shootNumber = 1000000，newSingleThreadExecutor和newSingleThreadExecutor（2）或更多都给出了大约51000毫秒的结果