java性能测试jmh

1 简介

jmh主要用于开发人员进行性能压测，代替手动编写相关程序。

官网：

http://openjdk.java.net/projects/code-tools/jmh/

JMH是一个Java工具，用于构建、运行和分析用Java和其他语言编写的针对JVM的nano/micro/milli/宏基准测试。

2 使用

2.1 maven依赖

<dependency>
    <groupId>org.openjdk.jmh</groupId>
    <artifactId>jmh-core</artifactId>
    <version>1.23</version>
</dependency>
<dependency>
    <groupId>org.openjdk.jmh</groupId>
    <artifactId>jmh-generator-annprocess</artifactId>
    <version>1.23</version>
    <scope>test</scope>
</dependency>

2.2 实例代码

import org.openjdk.jmh.runner.Runner;
import org.openjdk.jmh.runner.options.Options;
import org.openjdk.jmh.runner.options.OptionsBuilder;

import java.util.concurrent.TimeUnit;
//使用模式，默认是Mode.Throughput，表示吞吐量(AverageTime，表示每次执行时间，SampleTime表示采样时间，SingleShotTime表示只运行一次，用于测试冷启动消耗时间，All表示统计前面的所有指标)
@BenchmarkMode(Mode.Throughput)
//预热次数(预热2轮，每次1s)
@Warmup(iterations = 2, time = 1, timeUnit = TimeUnit.SECONDS)
//执行次数(运行3次，一次运行2秒)
@Measurement(iterations = 3, time = 2, timeUnit = TimeUnit.SECONDS)
//配置同时起多少个线程执行，默认值是Runtime.getRuntime().availableProcessors()
@Threads(1)
//启动多个单独的进程分别测试每个方法
@Fork(1)
//统计结果的时间单元
@OutputTimeUnit(TimeUnit.SECONDS)
public class MyBenchmark {
    
    //表示此方法需要进行测试
    @Benchmark
    public static void test01(){
        String s="1"+"2"+"3"+"4";
    }
    
    @Benchmark
    public static void test02(){
        String s=new StringBuffer("1").append("2").append("3").append("4").toString();
    }
    
    public static void main(String[] args) throws Exception{
        Options opt=new OptionsBuilder()
                //导入要测试的类
                .include(MyBenchmark.class.getSimpleName())
                .build();
        new Runner(opt).run();
    }
}

2.3 输出结果

# JMH version: 1.23
# VM version: JDK 1.8.0_121, Java HotSpot(TM) 64-Bit Server VM, 25.121-b13
# VM invoker: C:\Program Files\Java\jdk1.8.0_121\jre\bin\java.exe
# VM options: -javaagent:C:\work\software\IntelliJ IDEA 2019.2.3\lib\idea_rt.jar=53609:C:\work\software\IntelliJ IDEA 2019.2.3\bin -Dfile.encoding=UTF-8
# Warmup: 2 iterations, 1 s each
# Measurement: 3 iterations, 2 s each
# Timeout: 10 min per iteration
# Threads: 1 thread, will synchronize iterations
# Benchmark mode: Throughput, ops/time
# Benchmark: com.test.jmh.MyBenchmark.test01

# Run progress: 0.00% complete, ETA 00:00:16
# Fork: 1 of 1
# Warmup Iteration   1: 1668396652.659 ops/s
# Warmup Iteration   2: 1909507594.198 ops/s
Iteration   1: 1857680113.235 ops/s
Iteration   2: 1928208879.048 ops/s
Iteration   3: 1967828174.481 ops/s


Result "com.test.jmh.MyBenchmark.test01":
  1917905722.254 ±(99.9%) 1017856695.855 ops/s [Average]
  (min, avg, max) = (1857680113.235, 1917905722.254, 1967828174.481), stdev = 55792160.103
  CI (99.9%): [900049026.399, 2935762418.110] (assumes normal distribution)


# JMH version: 1.23
# VM version: JDK 1.8.0_121, Java HotSpot(TM) 64-Bit Server VM, 25.121-b13
# VM invoker: C:\Program Files\Java\jdk1.8.0_121\jre\bin\java.exe
# VM options: -javaagent:C:\work\software\IntelliJ IDEA 2019.2.3\lib\idea_rt.jar=53609:C:\work\software\IntelliJ IDEA 2019.2.3\bin -Dfile.encoding=UTF-8
# Warmup: 2 iterations, 1 s each
# Measurement: 3 iterations, 2 s each
# Timeout: 10 min per iteration
# Threads: 1 thread, will synchronize iterations
# Benchmark mode: Throughput, ops/time
# Benchmark: com.test.jmh.MyBenchmark.test02

# Run progress: 50.00% complete, ETA 00:00:09
# Fork: 1 of 1
# Warmup Iteration   1: 107194791.613 ops/s
# Warmup Iteration   2: 99651519.700 ops/s
Iteration   1: 170972299.861 ops/s
Iteration   2: 207053028.473 ops/s
Iteration   3: 203784523.495 ops/s


Result "com.test.jmh.MyBenchmark.test02":
  193936617.276 ±(99.9%) 364048302.894 ops/s [Average]
  (min, avg, max) = (170972299.861, 193936617.276, 207053028.473), stdev = 19954715.908
  CI (99.9%): [≈ 0, 557984920.171] (assumes normal distribution)


# Run complete. Total time: 00:00:18

REMEMBER: The numbers below are just data. To gain reusable insights, you need to follow up on
why the numbers are the way they are. Use profilers (see -prof, -lprof), design factorial
experiments, perform baseline and negative tests that provide experimental control, make sure
the benchmarking environment is safe on JVM/OS/HW level, ask for reviews from the domain experts.
Do not assume the numbers tell you what you want them to tell.

Benchmark            Mode  Cnt           Score            Error  Units
MyBenchmark.test01  thrpt    3  1917905722.254 ± 1017856695.855  ops/s
MyBenchmark.test02  thrpt    3   193936617.276 ±  364048302.894  ops/s

3 参数说明

3.1 测试模式

Mode表示JMH进行Benchmark时所使用的模式，代表测量的维度。

（1）Throughput：整体吞吐量，如1秒内可以执行多少次调用

（2）AverageTime：调用的平均时间，如每次调用平均耗时

（3）SampleTime：随机取样，最后输出取样结果的分布，如xx%的调用在xx毫秒以内，xx%的调用在xx毫秒以上

（4）SingleShotTime：只运行一次，用于测试冷启动的性能