定时线程分配统计事件:jdk.ThreadAllocationStatistics

引入版本:Java 11

相关 ISSUES

  1. Test jdk/jfr/event/runtime/TestThreadAllocationEvent.java fails with null thread:在某些情况下,jdk.ThreadAllocationStatistics 没有采集到 JFR 相关线程,导致空指针。这个 Bug 和 jdk/jfr/event/compiler/TestCompilerCompile.java failed due to "RuntimeException: No thread in event" 重复.该 BUG 对于使用没有大影响,于 Java 16 修复。
  2. Per thread IO statistics in JFR,这是一个新特性,期望和 ThreadAllocationStatistics 事件类似,提供每个线程的 IO 数据统计,这个特性还没有通过讨论。

各版本配置:

从 Java 11 引入之后没有改变过:

默认配置default.jfc of Java 11default.jfc of Java 12default.jfc of Java 13default.jfc of Java 14default.jfc of Java 15default.jfc of Java 16default.jfc of Java 17):

配置 描述
enabled false 默认不启用
period everyChunk 在生成每个新 Chunk 文件的时候采集一次

采样配置profile.jfc of Java 11profile.jfc of Java 12profile.jfc of Java 13profile.jfc of Java 14profile.jfc of Java 15profile.jfc of Java 16profile.jfc of Java 17):

配置 描述
enabled true 默认启用
period everyChunk 在生成每个新 Chunk 文件的时候采集一次

为何需要这个事件?

Java 中业务线程都会分配对象,对于以下关键的业务,对象的分配可能更加频繁。有时候我们可能会遇到以下两个情况:

  1. 线上应用频繁 GC,可能是先开始 Young GC(New GC,Light GC,不同 GC 叫法不一样),之后伴随着 Old GC 或者 Full GC,可以观察到线上应用占用 CPU 高的线程也是 GC 线程。这个很可能是某个业务 BUG 拉取了大量的数据,例如查询数据库没加条件导致查询整个表,三方接口没限制 limit 返回了超级多的数据。想要直观快速的定位这个问题,可以通过观察那些线程突然分配了很多内存,然后查看这些线程的堆栈进一步确认。jdk.ThreadAllocationStatistics 就是用来查看这个的
  2. 线上 GC 变多,想要减少 GC,光看 jmap 对象统计太多抽象,并不直观知道是那些代码创建的这些对象。堆 dump 太重,对于大内存进程分析成本也很大。可以通过 jdk.ThreadAllocationStatistics 查看是那些线程分配对象比较多,采集这些线程的堆栈可以定位相关代码。

事件包含属性

属性 说明 举例
startTime 事件开始时间 10:16:27.718
allocated 分配过的空间大小 10.0 MB
thread 线程信息 "Thread-0" (javaThreadId = 27)

使用代码测试这个事件

package com.github.hashjang.jfr.test;

import com.sun.management.ThreadMXBean;
import jdk.jfr.Recording;
import jdk.jfr.consumer.RecordedEvent;
import jdk.jfr.consumer.RecordedThread;
import jdk.jfr.consumer.RecordingFile;
import sun.hotspot.WhiteBox; import java.io.File;
import java.io.IOException;
import java.lang.management.ManagementFactory;
import java.nio.file.Path;
import java.time.Duration;
import java.util.Arrays;
import java.util.HashMap;
import java.util.Map;
import java.util.Random;
import java.util.concurrent.CountDownLatch;
import java.util.concurrent.TimeUnit;
import java.util.stream.Collectors; public class TestThreadAllocationStatistics {
private static String EVENT_NAME = "jdk.ThreadAllocationStatistics";
//50ms采集一次
private static long eventPeriodMillis = 50;
//对于字节数组对象头占用16字节
private static final int BYTE_ARRAY_OVERHEAD = 16;
//我们要测试的对象大小是1kb
private static final int OBJECT_SIZE = 1024; public static void main(String[] args) throws IOException { Recording recording = new Recording();
recording.enable(EVENT_NAME).withPeriod(Duration.ofMillis(eventPeriodMillis));
recording.start();
//使用 WhiteBox 执行 FullGC,清除干扰
WhiteBox whiteBox = WhiteBox.getWhiteBox();
whiteBox.fullGC(); Allocator allocators[] = new Allocator[4];
CountDownLatch allocationsDoneLatch = new CountDownLatch(allocators.length);
for (int i = 0; i < allocators.length; i++) {
allocators[i] = new Allocator(allocationsDoneLatch, OBJECT_SIZE * (i + 1) - BYTE_ARRAY_OVERHEAD);
allocators[i].setDaemon(true);
allocators[i].start();
}
Map<Long, Allocator> map = Arrays.stream(allocators).collect(Collectors.toMap(Thread::getId, v -> v));
Map<Long, Long> threadAllocatedMap = new HashMap<>();
try {
allocationsDoneLatch.await();
//再等待一段时间让定时采集 jdk.ThreadAllocationStatistics 采集到最新的
TimeUnit.SECONDS.sleep(1);
} catch (InterruptedException e) {
e.printStackTrace();
} recording.stop();
Path path = new File(new File(".").getAbsolutePath(), "recording-" + recording.getId() + "-pid" + ProcessHandle.current().pid() + ".jfr").toPath();
recording.dump(path);
for (RecordedEvent event : RecordingFile.readAllEvents(path)) {
if (!EVENT_NAME.equals(event.getEventType().getName())) {
continue;
}
long javaThreadId = ((RecordedThread) event.getValue("thread")).getJavaThreadId();
if (map.containsKey(javaThreadId)) {
//事件是时间有序的(采集时间比较短),所以放入的最后一个就是采集到的线程最终的分配大小
threadAllocatedMap.put(javaThreadId, event.getLong("allocated"));
System.out.println(event);
}
} map.forEach((id, thread) -> {
System.out.println("Thread " + id + " allocated(from JMX): " + thread.totalAllocated + "; allocated(from jdk.ThreadAllocationStatistics): " + threadAllocatedMap.get(id));
});
} public static class Allocator extends Thread {
private volatile long totalAllocated = -1;
private final int allocationSize;
public byte[] buffer;
private final CountDownLatch allocationsDoneLatch; public Allocator(CountDownLatch allocationsDoneLatch, int allocationSize) {
this.allocationsDoneLatch = allocationsDoneLatch;
this.allocationSize = allocationSize;
} @Override
public void run() {
for (int batches = 0; batches < 100; batches++) {
for (int i = 0; i < 1024; i++) {
buffer = new byte[allocationSize];
}
try {
//期望每个采集周期之间的分配次数为 5
Thread.sleep(eventPeriodMillis / 5);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
//获取当前线程分配的对象大小
totalAllocated = getThreadAllocatedBytes();
allocationsDoneLatch.countDown();
//设置线程为守护线程,等待主线程结束之后会自动结束
//这里进入死循环是因为防止定时采集 jdk.ThreadAllocationStatistics 事件的时候采集不到
while (true) {
Thread.yield();
}
} private long getThreadAllocatedBytes() {
ThreadMXBean bean = (ThreadMXBean) ManagementFactory.getThreadMXBean();
return bean.getThreadAllocatedBytes(Thread.currentThread().getId());
}
}
}

使用以下 JVM 参数启动:

-Xbootclasspath/a:D:\github\jfr-spring-all\jdk-white-box\target\jdk-white-box-17.0-SNAPSHOT.jar  -Xms512m -Xmx512m  -XX:+UnlockDiagnosticVMOptions -XX:+WhiteBoxAPI -Xlog:gc

输出结果:

[0.016s][info][gc] Using G1
[0.679s][info][gc] GC(0) Pause Full (WhiteBox Initiated Full GC) 19M->2M(512M) 8.804ms
[0.714s][info][gc] GC(1) Pause Young (Normal) (G1 Evacuation Pause) 27M->2M(512M) 1.309ms
[1.099s][info][gc] GC(2) Pause Young (Normal) (G1 Evacuation Pause) 308M->2M(512M) 0.943ms
[1.445s][info][gc] GC(3) Pause Young (Normal) (G1 Evacuation Pause) 308M->2M(512M) 1.141ms
[1.788s][info][gc] GC(4) Pause Young (Normal) (G1 Evacuation Pause) 308M->2M(512M) 1.117ms
jdk.ThreadAllocationStatistics {
startTime = 12:41:29.915
allocated = 1.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.915
allocated = 2.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.915
allocated = 3.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.915
allocated = 4.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.965
allocated = 6.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.965
allocated = 12.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.965
allocated = 15.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:29.965
allocated = 20.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.015
allocated = 10.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.015
allocated = 20.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.015
allocated = 27.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.015
allocated = 36.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.066
allocated = 15.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.066
allocated = 28.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.066
allocated = 39.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.066
allocated = 50.3 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.116
allocated = 19.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.116
allocated = 36.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.116
allocated = 51.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.116
allocated = 64.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.167
allocated = 24.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.167
allocated = 44.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.167
allocated = 63.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.167
allocated = 80.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.217
allocated = 28.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.217
allocated = 52.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.217
allocated = 75.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.217
allocated = 96.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.268
allocated = 32.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.268
allocated = 61.3 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.268
allocated = 87.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.268
allocated = 109.5 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.319
allocated = 37.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.319
allocated = 68.3 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.319
allocated = 99.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.319
allocated = 124.3 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.370
allocated = 41.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.370
allocated = 78.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.370
allocated = 111.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.370
allocated = 144.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.420
allocated = 45.4 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.420
allocated = 86.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.420
allocated = 126.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.420
allocated = 160.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.470
allocated = 50.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.470
allocated = 96.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.470
allocated = 138.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.470
allocated = 180.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.520
allocated = 54.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.520
allocated = 104.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.520
allocated = 150.4 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.520
allocated = 196.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.571
allocated = 58.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.571
allocated = 112.3 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.571
allocated = 165.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.571
allocated = 212.5 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.621
allocated = 63.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.621
allocated = 122.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.621
allocated = 180.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.621
allocated = 232.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.671
allocated = 67.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.671
allocated = 130.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.671
allocated = 192.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.671
allocated = 248.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.722
allocated = 72.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.722
allocated = 140.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.722
allocated = 207.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.722
allocated = 268.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.772
allocated = 76.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.772
allocated = 148.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.772
allocated = 219.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.772
allocated = 284.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.581
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.581
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.581
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.581
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.631
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.631
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.631
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.631
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.682
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.682
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.682
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.682
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.732
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.732
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.732
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.732
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.823
allocated = 81.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.823
allocated = 158.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.823
allocated = 234.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.823
allocated = 304.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.874
allocated = 85.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.874
allocated = 166.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.874
allocated = 246.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.874
allocated = 320.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.925
allocated = 90.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.925
allocated = 176.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.925
allocated = 258.5 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.925
allocated = 340.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.975
allocated = 94.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.975
allocated = 184.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.975
allocated = 273.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:30.975
allocated = 356.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.024
allocated = 99.0 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.024
allocated = 194.0 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.024
allocated = 285.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.024
allocated = 372.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.075
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.075
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.075
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.075
allocated = 392.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.126
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.126
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.126
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.126
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.177
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.177
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.177
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.177
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.228
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.228
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.228
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.228
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.278
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.278
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.278
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.278
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.329
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.329
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.329
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.329
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.379
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.379
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.379
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.379
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.429
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.429
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.429
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.429
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.479
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.479
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.479
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.479
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.530
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.530
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.530
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.530
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.791
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.791
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.791
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.791
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.842
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.842
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.842
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.842
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.892
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.892
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.892
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.892
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.943
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.943
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.943
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.943
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.993
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.993
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.993
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:31.993
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.043
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.043
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.043
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.043
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.093
allocated = 100.3 MB
thread = "Thread-0" (javaThreadId = 27)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.093
allocated = 200.1 MB
thread = "Thread-1" (javaThreadId = 28)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.093
allocated = 300.0 MB
thread = "Thread-2" (javaThreadId = 29)
} jdk.ThreadAllocationStatistics {
startTime = 12:41:32.093
allocated = 400.0 MB
thread = "Thread-3" (javaThreadId = 30)
} Thread 27 allocated(from JMX): 105179160; allocated(from jdk.ThreadAllocationStatistics): 105179160
Thread 28 allocated(from JMX): 209780040; allocated(from jdk.ThreadAllocationStatistics): 209780040
Thread 29 allocated(from JMX): 314573616; allocated(from jdk.ThreadAllocationStatistics): 314573616
Thread 30 allocated(from JMX): 419431216; allocated(from jdk.ThreadAllocationStatistics): 419431216

底层原理以及相关 JVM 源码

每隔配置的间隔,就会调用 requestThreadAllocationStatistics(void) 方法采集每一个线程的分配信息:

jfrPeriodic.cpp

#define TRACE_REQUEST_FUNC(id)    void JfrPeriodicEventSet::request##id(void)

//使用了宏定义函数,实际函数就是 requestThreadAllocationStatistics(void)
TRACE_REQUEST_FUNC(ThreadAllocationStatistics) {
ResourceMark rm;
//获取线程数量,虽然后面线程数量在采集过程中可能会改变,利用这个值初始化数组
int initial_size = Threads::number_of_threads();
//建立一个数组记录每个线程的分配大小
GrowableArray<jlong> allocated(initial_size);
//建立一个数组记录线程号,和上面那个数组一一对应
GrowableArray<traceid> thread_ids(initial_size);
//记录当前时间
JfrTicks time_stamp = JfrTicks::now();
//新建一个 JfrJavaThreadIterator 来遍历每个线程
JfrJavaThreadIterator iter;
while (iter.has_next()) {
JavaThread* const jt = iter.next();
assert(jt != NULL, "invariant");
//读取每个线程的 cooked_allocated_bytes()
allocated.append(jt->cooked_allocated_bytes());
//记录线程号(包括系统线程号以及 Java 线程号)
thread_ids.append(JFR_THREAD_ID(jt));
} //遍历数组,生成 JFR 事件并采集
for(int i = 0; i < thread_ids.length(); i++) {
EventThreadAllocationStatistics event(UNTIMED);
//设置当前线程已分配大小
event.set_allocated(allocated.at(i));
//设置线程信息
event.set_thread(thread_ids.at(i));
//设置结束时间
event.set_endtime(time_stamp);
event.commit();
}
}

统计线程的分配内存的大小方法即 cooked_allocated_bytes() 方法的源码如下:

thread.inline.hpp

inline jlong Thread::cooked_allocated_bytes() {
//原子读取 _allocated_bytes
jlong allocated_bytes = Atomic::load_acquire(&_allocated_bytes);
//如果开启了 TLAB(默认开启),则需要查看下当前线程的 TLAB 已经分配的对象大小
if (UseTLAB) {
//统计当前线程的 TLAB 已经分配的大小
size_t used_bytes = tlab().used_bytes();
//如果当前 TLAB 在初始化,或者全局 TLAB 在初始化,会发生 used_bytes > max_size 的情况,忽略这种情况。
if (used_bytes <= ThreadLocalAllocBuffer::max_size_in_bytes()) {
//这样统计其实有可能有点问题,即发生 TLAB retire 的时候,_allocated_bytes 会加上 used_bytes 之后申请一个新的 TLAB,这时候调用这个方法可能会把这个 used_bytes 加了两遍
return allocated_bytes + used_bytes;
}
}
return allocated_bytes;
}

那么哪里会修改 _allocated_bytes 这个变量呢?有两个地方,一个是发生 TLAB 外分配的时候,另一个是发生 TLAB retire 的时候。 TLAB retire 是在发生 GC 以及 TLAB 满了并且剩余空间小于当前最大浪费空间限制的时候会重新申请一个新的 TLAB 进行分配,申请新的之前需要将当前 TLAB retire。

发生 TLAB 外分配修改 _allocated_bytes 对应源码:

memAllocator.cpp

//如果是 TLAB 外分配,会调用这个方法
HeapWord* MemAllocator::allocate_outside_tlab(Allocation& allocation) const {
//标记当前分配为 TLAB 外分配
allocation._allocated_outside_tlab = true;
//执行 TLAB 外分配,不同 GC 不同
HeapWord* mem = Universe::heap()->mem_allocate(_word_size, &allocation._overhead_limit_exceeded);
if (mem == NULL) {
return mem;
} NOT_PRODUCT(Universe::heap()->check_for_non_bad_heap_word_value(mem, _word_size));
//计算当前分配的字节大小
size_t size_in_bytes = _word_size * HeapWordSize;
//增加 _allocated_bytes 当前分配的字节大小
_thread->incr_allocated_bytes(size_in_bytes); return mem;
}

TLAB retire 对应源码:

threadLocalAllocBuffer.cpp

void ThreadLocalAllocBuffer::retire(ThreadLocalAllocStats* stats) {
if (stats != NULL) {
accumulate_and_reset_statistics(stats);
} if (end() != NULL) {
invariants();
//将当前 TLAB 使用的字节数加到 _allocated_bytes
//从这里可以看出,如果这个方法和 cooked_allocated_bytes() 同时被调用,可能 used_bytes 被加了两遍。
thread()->incr_allocated_bytes(used_bytes());
insert_filler();
initialize(NULL, NULL, NULL);
}
}

我们再来看看 JMX 中查看线程分配内存大小的具体实现。首先对应的 JMX 的代码是:

//我们示例中底层调用的实际就是这个方法获取线程分配的内存大小
protected long getThreadAllocatedBytes(long id) {
boolean verified = verifyThreadAllocatedMemory(id); if (verified) {
//调用 getThreadAllocatedMemory0 获取分配的内存大小
return getThreadAllocatedMemory0(
Thread.currentThread().getId() == id ? 0 : id);
}
return -1;
} //getThreadAllocatedMemory0 是一个 native 方法
private static native long getThreadAllocatedMemory0(long id);

这个 native 方法对应的 JVM 源码是:

ThreadImpl.c

JNIEXPORT jlong JNICALL
Java_sun_management_ThreadImpl_getThreadAllocatedMemory0
(JNIEnv *env, jclass cls, jlong tid)
{
//实际实现方法是 GetOneThreadAllocatedMemory
return jmm_interface->GetOneThreadAllocatedMemory(env, tid);
}

management.cpp

JVM_ENTRY(jlong, jmm_GetOneThreadAllocatedMemory(JNIEnv *env, jlong thread_id))
if (thread_id < 0) {
THROW_MSG_(vmSymbols::java_lang_IllegalArgumentException(),
"Invalid thread ID", -1);
}
//获取当前线程
if (thread_id == 0) { // current thread
//调用 cooked_allocated_bytes,和采集 jdk.ThreadAllocationStatistics 调用的底层方法一样
return thread->cooked_allocated_bytes();
} //根据线程号获取线程
ThreadsListHandle tlh;
JavaThread* java_thread = tlh.list()->find_JavaThread_from_java_tid(thread_id); if (java_thread != NULL) {
//调用 cooked_allocated_bytes,和采集 jdk.ThreadAllocationStatistics 调用的底层方法一样
return java_thread->cooked_allocated_bytes();
}
return -1;
JVM_END

可以看出,其实底层调用的和 JFR 采集 jdk.ThreadAllocationStatistics 事件一样,都是调用 Thread 的 cooked_allocated_bytes() 方法。

针对这个 JFR 事件的一些思考

首先,提出一个观点,jdk.ThreadAllocationStatistics 这个事件并不太消耗性能。原因有二:

  1. 统计线程分配大小是近似统计,并不需要进入全局安全点统计。同时也代表,jdk.ThreadAllocationStatistics 同一时间的不同线程事件的分配大小实际并不是同一时间点的,因为没有进入安全点暂停所有线程
  2. 统计仅仅是原子读取每个线程的分配对象大小这个变量,之后加上每个线程当前 TLAB 分配对象大小(这个大小是通过读取两个指针地址获取的),可以看出操作是很轻量级的。即使有很多线程,也不会增加多少性能负担

然后,默认配置的采集周期,并不能满足我们的需求。默认的采集周期是 everyChunk,默认的 chunk 大小(maxchunksize)是 12M,也就是每采集 12M 的 JFR 事件之后,采集一次 jdk.ThreadAllocationStatistics。这是不太可控的,我一般配置为每过 5s 采集一次。这样对于我们上面提到的那两个需要这个事件的场景也是很适合的。

微信搜索“我的编程喵”关注公众号,加作者微信,每日一刷,轻松提升技术,斩获各种offer

Java JFR 民间指南 - 事件详解 - jdk.ThreadAllocationStatistics的更多相关文章

  1. Java JFR 民间指南 - 事件详解 - jdk.ObjectAllocationInNewTLAB

    重新申请 TLAB 分配对象事件:jdk.ObjectAllocationInNewTLAB 引入版本:Java 11 相关 ISSUES: JFR: RecordingStream leaks me ...

  2. Java JFR 民间指南 - 事件详解 - jdk.ObjectAllocationOutsideTLAB

    重新申请 TLAB 分配对象事件:jdk.ObjectAllocationOutsideTLAB 引入版本:Java 11 相关 ISSUES: JFR: RecordingStream leaks ...

  3. Java JFR 民间指南 - 事件详解 - jdk.ObjectAllocationSample

    对象分配采样:jdk.ObjectAllocationSample 引入版本:Java 16 相关 ISSUE:Introduce JFR Event Throttling and new jdk.O ...

  4. Java虚拟机之垃圾回收详解一

    Java虚拟机之垃圾回收详解一 Java技术和JVM(Java虚拟机) 一.Java技术概述: Java是一门编程语言,是一种计算平台,是SUN公司于1995年首次发布.它是Java程序的技术基础,这 ...

  5. Java网络编程和NIO详解5:Java 非阻塞 IO 和异步 IO

    Java网络编程和NIO详解5:Java 非阻塞 IO 和异步 IO Java 非阻塞 IO 和异步 IO 转自https://www.javadoop.com/post/nio-and-aio 本系 ...

  6. Java网络编程和NIO详解2:JAVA NIO一步步构建IO多路复用的请求模型

    Java网络编程与NIO详解2:JAVA NIO一步步构建IO多路复用的请求模型 知识点 nio 下 I/O 阻塞与非阻塞实现 SocketChannel 介绍 I/O 多路复用的原理 事件选择器与 ...

  7. JavaScript事件详解-jQuery的事件实现(三)

    正文 本文所涉及到的jQuery版本是3.1.1,可以在压缩包中找到event模块.该篇算是阅读笔记,jQuery代码太长.... Dean Edward的addEvent.js 相对于zepto的e ...

  8. java中的io系统详解 - ilibaba的专栏 - 博客频道 - CSDN.NET

    java中的io系统详解 - ilibaba的专栏 - 博客频道 - CSDN.NET 亲,“社区之星”已经一周岁了!      社区福利快来领取免费参加MDCC大会机会哦    Tag功能介绍—我们 ...

  9. Java Spring cron表达式使用详解

    Java Spring cron表达式使用详解   By:授客 QQ:1033553122 语法格式 Seconds Minutes Hours DayofMonth Month DayofWeek ...

随机推荐

  1. 【转】理解Serverless

    理解Serverless No silver bullet. - The Mythical Man-Month 许多年前,我们开发的软件还是C/S(客户端/服务器)和MVC(模型-试图-控制器)的形式 ...

  2. 《C++ Primer》笔记 第10章 泛型算法

    迭代器令算法不依赖于容器,但算法依赖于元素类型的操作. 算法永远不会执行容器的操作.算法永远不会改变底层容器的大小. accumulate定义在头文件numeric中,接受三个参数,前两个指出需要求和 ...

  3. C++数组的存储与初始化

    下面随笔给出C++数组的存储与初始化的细节内容. 数组的存储与初始化 一维数组的存储 数组元素在内存中顺次存放,它们的地址是连续的.元素间物理地址上的相邻,对应着逻辑次序上的相邻. 例如:

  4. MySQL时间戳unix_timestamp

    函数:FROM_UNIXTIME作用:将MYSQL中以INT(11)存储的时间以"YYYY-MM-DD"格式来显示.语法:FROM_UNIXTIME(unix_timestamp, ...

  5. 如何获取下载 FreeBSD

    『如何获取下载 FreeBSD 』 『如何获取下载 FreeBSD 』 FreeBSD 是免费获取的. [下载地址] O网页链接 版本选择,尽量选择较新版本,桌面用户可选择 current 版本.st ...

  6. Codeforces 976C Nested Segments

    题面: 传送门 C. Nested Segments Input file: standard input Output file: standard output Time limit: 2 secon ...

  7. 数字转人民币读法-python3

    """ 2 把一个浮点数分解成证书备份和小数部分 3 """ 4 def divide(num): 5 intnum = int(num) ...

  8. 学习一下 SpringCloud (六)-- 注册中心与配置中心 Nacos、网关 Gateway

    (1) 相关博文地址: 学习一下 SpringCloud (一)-- 从单体架构到微服务架构.代码拆分(maven 聚合): https://www.cnblogs.com/l-y-h/p/14105 ...

  9. CMS前世今生

    CMS一直是面试中的常考点,今天我们用通俗易懂的语言简单介绍下. 垃圾回收器为什么要分区分代? 如上图:JVM虚拟机将堆内存区域分代了,先生代是朝生夕死的区域,老年代是老不死的区域,不同的年代对象有不 ...

  10. 基于Hive进行数仓建设的资源元数据信息统计:Hive篇

    在数据仓库建设中,元数据管理是非常重要的环节之一.根据Kimball的数据仓库理论,可以将元数据分为这三类: 技术元数据,如表的存储结构结构.文件的路径 业务元数据,如血缘关系.业务的归属 过程元数据 ...