Java后端进阶-JVM参数调优

package com.study.performance;

import org.springframework.boot.SpringApplication;

import org.springframework.boot.autoconfigure.SpringBootApplication;

import java.util.Random;

import java.util.concurrent.Executors;

import java.util.concurrent.TimeUnit;

// 启动程序，模拟用户请求

// 每100毫秒钟创建1000线程，每个线程创建一个512kb的对象，最多100毫秒内同时存在1000线程，并发量1000/s，吞吐量6000/s，查看GC的情况

@SpringBootApplication

public class PerformanceApplication {

public static void main(String[] args) {

SpringApplication.run(PerformanceApplication.class, args);

Executors.newScheduledThreadPool(1).scheduleAtFixedRate(() -> {

new Thread(() -> {

for (int i = 0; i < 1000; i++) {

try {

//  不干活，专门512kb的小对象

byte[] temp = new byte[1024 * 512];

Thread.sleep(new Random().nextInt(100)); // 随机睡眠200毫秒秒以内

} catch (InterruptedException e) {

e.printStackTrace();

}

}

}).start();

}, 100, 100, TimeUnit.MILLISECONDS);

}

}



// 打包 mvn clean package

// 服务器上运行 performance-1.0.0.jar

//对象存活1秒的场景，远远超过平时接口的响应时间要求，场景应该是吞吐量优先

ivan@ivan-virtual-machine:~/Desktop$ jmap
Usage:
    jmap [option] <pid>
        (to connect to running process)
    jmap [option] <executable <core>
        (to connect to a core file)
    jmap [option] [server_id@]<remote server IP or hostname>
        (to connect to remote debug server)

where <option> is one of:

<none>               to print same info as Solaris pmap

-heap                to print java heap summary

-histo[:live]        to print histogram of java object heap; if the "live"

suboption is specified, only count live objects

-clstats             to print class loader statistics

-finalizerinfo       to print information on objects awaiting finalization

-dump:<dump-options> to dump java heap in hprof binary format

dump-options:

live         dump only live objects; if not specified,

all objects in the heap are dumped.

format=b     binary format

file=<file>  dump heap to <file>

Example: jmap -dump:live,format=b,file=heap.bin <pid>

-F                   force. Use with -dump:<dump-options> <pid> or -histo

to force a heap dump or histogram when <pid> does not

respond. The "live" suboption is not supported

in this mode.

-h | -help           to print this help message

-J<flag>             to pass <flag> directly to the runtime system

ivan@ivan-virtual-machine:~/Desktop$ jcmd

9476 sun.tools.jcmd.JCmd

8622 performance-1.1.0.jar

root@ivan-virtual-machine:/home/ivan/Desktop# jmap -heap 8622

Attaching to process ID 8622, please wait...

Debugger attached successfully.

Server compiler detected.

JVM version is 25.242-b08

using thread-local object allocation.

Mark Sweep Compact GC

Heap Configuration:

MinHeapFreeRatio         = 40

MaxHeapFreeRatio         = 70

MaxHeapSize              = 1073741824 (1024.0MB)

NewSize                  = 11141120 (10.625MB)

MaxNewSize               = 357892096 (341.3125MB)

OldSize                  = 22413312 (21.375MB)

NewRatio                 = 2

SurvivorRatio            = 8

MetaspaceSize            = 21807104 (20.796875MB)

CompressedClassSpaceSize = 1073741824 (1024.0MB)

MaxMetaspaceSize         = 17592186044415 MB

G1HeapRegionSize         = 0 (0.0MB)



Heap Usage:

Exception in thread "main" java.lang.reflect.InvocationTargetException

at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)

at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)

at java.lang.reflect.Method.invoke(Method.java:498)

at sun.tools.jmap.JMap.runTool(JMap.java:201)

at sun.tools.jmap.JMap.main(JMap.java:130)

Caused by: java.lang.RuntimeException: unknown CollectedHeap type : class sun.jvm.hotspot.gc_interface.CollectedHeap

at sun.jvm.hotspot.tools.HeapSummary.run(HeapSummary.java:144)

at sun.jvm.hotspot.tools.Tool.startInternal(Tool.java:260)

at sun.jvm.hotspot.tools.Tool.start(Tool.java:223)

at sun.jvm.hotspot.tools.Tool.execute(Tool.java:118)

at sun.jvm.hotspot.tools.HeapSummary.main(HeapSummary.java:49)

... 6 more

java版本问题报错

最大停顿时间是尽力保持，调整新生代和老年代的大小

GCTimeRatio也没啥用

SurvivorRatio也没啥用，自动会调整

案例

测试环境：JVM配置为2核1G，JAVA8，固定设置堆大小 1G

awk这个就是转换为字符串并且用空格进行分割，然后取得第一位

jinfo查看一下不是默认的参数

1、 调大-XX:ParallelGCThreads=4

java -Xmx1024m -Xloggc:/netease/gc2.log -XX:ParallelGCThreads=4  -jar performance-1.0.0.jar

jstat -gc -h10 $(jcmd | grep "performance-1.0.0.jar" | awk '{print $1}') 1000

线程还是要和核数相同最优

2、 降低耗时

设置-XX:MaxGCPauseMills=10

java -Xmx1024m -Xloggc:/netease/gc3.log -XX:MaxGCPauseMillis=10  -jar performance-1.0.0.jar

3、 改用CMS回收器

java -Xmx1024m -Xloggc:/netease/gc4.log -XX:+UseConcMarkSweepGC  -jar performance-1.0.0.jar

这个垃圾回收器不适用于吞吐量优先的情况。

https://docs.oracle.com/javase/8/

4、 增加线程

java -Xmx1024m -Xloggc:/netease/gc4.log -XX:+UseConcMarkSweepGC  -XX:ConcGCThreads=3 -jar performance-1.0.0.jar

用户线程和GC线程，会和用户线程争抢cpu

5、改用G1

java -Xmx1024m -Xloggc:/netease/gc10.log  -XX:+UseG1GC -jar performance-1.0.0.jar

6、增加分区大小

java -Xmx1024m -Xloggc:/netease/gc11.log  -XX:+UseG1GC -XX:G1HeapRegionSize=64m -jar performance-1.0.0.jar

示例代码 -2

2.1  GC调优

3、 结语

主要是演示切换的过程和思路，实际还是要结合系统情况、系统需要来调整。

1、 GC调优就是逐步调试的过程，对每个参数的含义了解后，再根据官方手册，一个个调试，找到符合应用的最佳配置点。

2、 再重复一句，性能问题，98.75%上是业务代码上面。

3、 无监控，不调优。