GraalVM是一个加速JVM语言运行的JDK工具，它同时提供了将JavaScript，Python等语言转化为JVM语言的能力，使得这些语言能集成到JVM语言体系中；
GraalVM采用两种工作方式，HotSpotJVM 上的 Graal just-in-time (JIT)，或者使用 ahead-of-time (AOT) 将JVM语言编译为原生二进制可执行文件！
官方网站：
系列文章：
GraalVM使用初体验
安装
GraalVM 的安装非常简单，和Java差不多：
下载压缩包、解压，然后将文件加入到 PATH 即可！
官方文档：
检查：
$ gu --version
GraalVM Updater 22.3.1和OracleJDK、OpenJDK类似，GraalVM也提供了两种：
- GraalVM-ce：社区版使用 OpenJDK，在 Github 下载：graalvm-ce-builds/releases
- GraalVM-ee：企业版使用 OracleJDK，在Oracle官网下载：install-graalvm-enterprise)
使用多语言
GraalVM 使用 gu 工具来管理其他工具链，并且在安装 GraalVM 时已经自带了 JDK；
可以直接使用：
javac HelloWorld.java
java HelloWorld
Hello World!如果想要运行其他语言，例如 JavaScript，需要先安装：
$ gu install js然后执行：
$ JAVA_HOME/bin/js
> 1 + 2
3此外，GraalVM 还提供了：
- nodejs
- llvm
- python
- ruby
- wasm
等等一系列语言；
使用方法也是类似；
Native原生编译
原生编译单个Java文件
编写 HelloWorld：
HelloWorld.java
public class HelloWorld {
public static void main(String[] args) {
System.out.println("Hello, Native World!");
}
} 编译并编译为二进制：
$ javac HelloWorld.java
$ native-image HelloWorld
=======================================================================================================
GraalVM Native Image: Generating 'helloworld' (executable)...
=======================================================================================================
[1/7] Initializing... (4.8s @ 0.21GB)
Version info: 'GraalVM 22.3.1 Java 17 EE'
Java version info: '17.0.6+9-LTS-jvmci-22.3-b11'
C compiler: cc (apple, arm64, 14.0.0)
Garbage collector: Serial GC
[2/7] Performing analysis... [*****] (3.7s @ 0.75GB)
1,751 (61.74%) of 2,836 classes reachable
1,567 (45.77%) of 3,424 fields reachable
6,874 (34.55%) of 19,898 methods reachable
18 classes, 0 fields, and 248 methods registered for reflection
49 classes, 33 fields, and 48 methods registered for JNI access
4 native libraries: -framework Foundation, dl, pthread, z
[3/7] Building universe... (0.5s @ 1.00GB)
[4/7] Parsing methods... [*] (0.4s @ 0.37GB)
[5/7] Inlining methods... [***] (0.2s @ 0.56GB)
[6/7] Compiling methods... [***] (6.7s @ 2.52GB)
[7/7] Creating image... (0.7s @ 0.29GB)
2.27MB (46.92%) for code area: 3,130 compilation units
2.36MB (48.74%) for image heap: 42,329 objects and 0 resources
215.15KB ( 4.34%) for other data
4.84MB in total
-------------------------------------------------------------------------------------------------------
Top 10 packages in code area: Top 10 object types in image heap:
316.28KB java.lang 424.41KB byte[] for code metadata
199.45KB java.util 339.53KB byte[] for java.lang.String
160.63KB com.oracle.svm.core.code 282.68KB java.lang.String
149.30KB com.oracle.svm.core.genscavenge 257.40KB java.lang.Class
116.85KB java.util.concurrent 226.30KB byte[] for general heap data
98.69KB java.math 111.71KB char[]
89.97KB com.oracle.svm.core.jni.functions 104.56KB java.util.HashMap$Node
83.14KB java.lang.invoke 78.59KB java.util.HashMap$Node[]
69.27KB com.oracle.svm.core 68.40KB c.o.svm.core.hub.DynamicHubCompanion
59.49KB com.oracle.svm.core.graal.snippets 66.45KB java.lang.Object[]
963.84KB for 92 more packages 418.79KB for 459 more object types
-------------------------------------------------------------------------------------------------------
0.3s (1.9% of total time) in 18 GCs | Peak RSS: 3.11GB | CPU load: 4.84
-------------------------------------------------------------------------------------------------------
Produced artifacts:
/Users/zk/workspace/java-all/cli/picocli/a-checksum/target/helloworld (executable)
/Users/zk/workspace/java-all/cli/picocli/a-checksum/target/helloworld.build_artifacts.txt (txt)
=======================================================================================================
Finished generating 'helloworld' in 17.5s.编译完成后就生成了可执行文件，运行：
$ ./helloworld
Hello, Native World!原生编译Jar包
这里使用了上一篇文章 《使用Java编写Cli命令行工具》 编译好的 Jar 包；
执行命令编译：
$ native-image -jar a-checksum-1.0-SNAPSHOT.jar
=======================================================================================================
GraalVM Native Image: Generating 'a-checksum-1.0-SNAPSHOT' (executable)...
=======================================================================================================
[1/7] Initializing... (5.2s @ 0.21GB)
Version info: 'GraalVM 22.3.1 Java 17 EE'
Java version info: '17.0.6+9-LTS-jvmci-22.3-b11'
C compiler: cc (apple, arm64, 14.0.0)
Garbage collector: Serial GC
[2/7] Performing analysis... [*****] (8.0s @ 1.60GB)
3,672 (76.88%) of 4,776 classes reachable
4,815 (54.41%) of 8,849 fields reachable
19,011 (49.51%) of 38,400 methods reachable
45 classes, 5 fields, and 551 methods registered for reflection
58 classes, 59 fields, and 52 methods registered for JNI access
4 native libraries: -framework Foundation, dl, pthread, z
[3/7] Building universe... (1.1s @ 0.81GB)
[4/7] Parsing methods... [*] (0.9s @ 1.95GB)
[5/7] Inlining methods... [***] (0.5s @ 2.48GB)
[6/7] Compiling methods... [****] (16.7s @ 3.28GB)
[7/7] Creating image... (1.7s @ 4.11GB)
9.03MB (49.09%) for code area: 9,751 compilation units
8.94MB (48.57%) for image heap: 149,730 objects and 5 resources
440.86KB ( 2.34%) for other data
18.40MB in total
-------------------------------------------------------------------------------------------------------
Top 10 packages in code area: Top 10 object types in image heap:
1017.12KB java.util 1.70MB byte[] for code metadata
930.98KB picocli 977.18KB java.lang.String
716.55KB java.lang 965.67KB byte[] for general heap data
515.41KB java.text 953.46KB byte[] for java.lang.String
423.18KB com.oracle.svm.core.code 611.61KB java.lang.Class
395.35KB jdk.internal.org.objectweb.asm 364.63KB java.util.HashMap$Node
367.71KB java.util.concurrent 350.19KB j.util.concurrent.ConcurrentHashMap$Node
261.72KB java.util.regex 172.13KB c.o.svm.core.hub.DynamicHubCompanion
252.84KB java.io 169.50KB byte[] for reflection metadata
195.60KB java.math 152.26KB java.util.HashMap$Node[]
4.02MB for 142 more packages 1.65MB for 929 more object types
-------------------------------------------------------------------------------------------------------
0.7s (1.9% of total time) in 24 GCs | Peak RSS: 4.83GB | CPU load: 5.61
-------------------------------------------------------------------------------------------------------
Produced artifacts:
/Users/zk/workspace/java-all/cli/picocli/a-checksum/target/a-checksum-1.0-SNAPSHOT (executable)
/Users/zk/workspace/java-all/cli/picocli/a-checksum/target/a-checksum-1.0-SNAPSHOT.build_artifacts.txt (txt)
=======================================================================================================
Finished generating 'a-checksum-1.0-SNAPSHOT' in 35.4s.运行编译好的文件：
$ ./a-checksum-1.0-SNAPSHOT
Missing required parameter: '<file>'
Usage: checksum [-hV] [-a=<algorithm>] <file>
Prints the checksum (SHA-256 by default) of a file to STDOUT.
<file> The file whose checksum to calculate.
-a, --algorithm=<algorithm>
MD5, SHA-1, SHA-256, ...
-h, --help Show this help message and exit.
-V, --version Print version information and exit.
$ ./a-checksum-1.0-SNAPSHOT ../hello.txt
5891b5b522d5df086d0ff0b110fbd9d21bb4fc7163af34d08286a2e846f6be03需要注意的是：
上面的 Jar 包是使用 JDK 8 编译产生的，但是使用的 GraalVM 工具链是：
GraalVM 22.3.1 Java 17 EE (Java Version 17.0.6+9-LTS-jvmci-22.3-b11)，即 JDK 17；
但是得益于 Java 良好的 Class 文件规范，生成的 Jar 包仍然能够被正确的编译！
总结
GraalVM 为 JVM 提供了更多的可能，例如：多语言（Polyglot Programming）、原生可执行文件（Native Image）；
Spring3 也已经支持了通过 GraalVM 将项目编译为原生可执行文件，相信以后这是一个发展的方向！
附录
系列文章：