我需要监视应用程序产生的线程消耗的内存量。如果贪婪的线程消耗太多内存，则想法是采取纠正措施。我已提到Java线程占用多少内存？。关于该链接的建议之一是getThreadAllocatedBytes在ThreadMXBean.我尝试getThreadAllocatedBytes以下工作时使用。

List<Long> primes = new ArrayList<Long>();
long i = 0;
while (true) {
            primes.add(++i);
            if ((i % 10) == 0) {
                primes.clear();
                System.runFinalization();
                System.gc();
            }
        }

我在四个线程上运行了很长时间。尽管作业不会连续地累积内存，但是所返回的值getThreadAllocatedBytes会不断增加，甚至不会下降。这意味着getThreadAllocatedBytes不会返回线程使用的堆上的实际内存量。它返回自线程启动以来在堆上为线程分配的内存总量。我的平台详细信息如下：

Linux PG85213.egi.ericsson.com 3.5.0-030500-generic＃201207211835 SMP Sat Jul
21 22:35:55 UTC 2012 x86_64 x86_64 x86_64 GNU / Linux Java版本“ 1.7.0_45”
Java（TM）SE运行时环境（build 1.7.0_45-b18）Java
HotSpot（TM）64位服务器VM（内部版本24.45-b08，混合模式）

以上行为是否是预期的行为getThreadAllocatedBytes？如果是这样，则无法在线程使用的堆上找到有效的内存。

列出完整的程序以供参考：

package workbench;

import java.lang.management.ManagementFactory;
import com.sun.management.ThreadMXBean;
import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.Executors;
import java.util.logging.Level;
import java.util.logging.Logger;

public class AnotherWorkBench {

private static final CountDownLatch latch = new CountDownLatch(4);
static final List<Long> threadIds = Collections.synchronizedList(new ArrayList<Long>());

private void dummyJob() {
    List<Long> primes = new ArrayList<Long>();
    long i = 0;
    while (true) {
        primes.add(++i);
        if ((i % 10) == 0) {
            primes.clear();
            //introduce sleep to prevent process hogging 
            try {
                Thread.currentThread().sleep(2000);
            } catch (InterruptedException ex) {
                Logger.getLogger(AnotherWorkBench.class.getName()).log(Level.SEVERE, null, ex);
            }
            System.runFinalization();
            System.gc();
        }
    }
}

private void runDummyJobs() {

    Runnable dummyJob = new Runnable() {
        @Override
        public void run() {
            threadIds.add(Thread.currentThread().getId());
            latch.countDown();
            dummyJob();
        }
    };

    Runnable memoryMonitorJob = new Runnable() {
        @Override
        public void run() {

            System.out.println(Thread.currentThread().getName() + " : Monitor thread started");
            ThreadMXBean threadMxBean = (ThreadMXBean) ManagementFactory.getThreadMXBean();
            threadMxBean.setThreadAllocatedMemoryEnabled(true);

            while (true) {
                for (Long threadId : threadIds) {
                    System.out.println(Thread.currentThread().getName() + " : Thread ID : " + threadId + " : memory = " + threadMxBean.getThreadAllocatedBytes(threadId) + " bytes");
                }

                //wait between subsequent scans
                try {
                    System.out.println(Thread.currentThread().getName() + " : secondary sleep");
                    Thread.currentThread().sleep(5000);
                    System.out.println(Thread.currentThread().getName() + " : out of secondary sleep");
                } catch (InterruptedException ex) {
                    Logger.getLogger(WorkBench.class.getName()).log(Level.SEVERE, null, ex);
                }
            }


        }
    };

    Executors.newSingleThreadExecutor().submit(dummyJob);
    Executors.newSingleThreadExecutor().submit(dummyJob);
    Executors.newSingleThreadExecutor().submit(dummyJob);
    Executors.newSingleThreadExecutor().submit(dummyJob);

    try {
        latch.await();
    } catch (InterruptedException ex) {
        Logger.getLogger(AnotherWorkBench.class.getName()).log(Level.SEVERE, null, ex);
    }
    Executors.newSingleThreadExecutor().submit(memoryMonitorJob);
}

/**
 * @param args the command line arguments
 */
public static void main(String[] args) {
    new AnotherWorkBench().runDummyJobs();
}
}