Get started with GraalVM and configure it to run programs written in Java (or other JVM-based languages), JavaScript and Node.js, Ruby, R, or Python. This page guides you through downloading and installing GraalVM and adding support for additional languages and/or plugins, and shows you how to run simple programs on GraalVM.

Install GraalVM

GraalVM is a standalone Java Development Kit to execute Java or JVM-based languages (e.g. Scala, Kotlin), dynamic languages (e.g. JavaScript, R, Ruby, R, Python), LLVM-based languages (e.g. C and C++) in one shared runtime, and supports Linux, macOS and Windows platforms on x86 64-bit systems. Please note, the support for Windows platform is experimental.

GraalVM 19.3.0 is available as Community Edition (CE) and Enterprise Edition (EE). GraalVM Community Editions are based on OpenJDK version 1.8.0_232 and on OpenJDK version 11.0.5. GraalVM Enterprise Editions are based on Oracle Java version 1.8.0_231 and on Oracle Java version 11.0.5. Proceed to the GraalVM distributions list to see a complete summary or to the Downloads page and make your choice depending on the operating system and the use case.

In this getting started guide we will focus on working with GraalVM Enterprise Edition based on JDK 8, obtained from the Oracle Technology Network. The base installation includes the JVM, the GraalVM compiler, the LLVM bitcode interpreter, and the JavaScript runtime with Node.js support – all in one package. GraalVM environment can be extended with:

  • GraalVM Native Image – a technology to compile an application ahead-of-time into a binary that runs natively on the system. Warning: GraalVM Native Image is available as an Early Adopter technology.
  • R, Ruby and Python languages interpreters. Warning: The support for Ruby, R and Python languages is experimental.
  • LLVM toolchain – a set of tools and APIs for compiling native programs to bitcode that can be executed with the GraalVM LLVM runtime.

Getting GraalVM installed and ready-to-go should only take a few minutes. Please note, unlike Java HotSpot VM or OpenJDK distributions for macOS that come as a .dmg file, GraalVM does not provide the installation wizard.

  1. Navigate to Oracle Technology Network Downloads page and accept the license agreement.
  2. Select and download Oracle GraalVM Enterprise Edition based on JDK8 for macOS (19.3.0).
  3. Extract the archive to your file system. To extract the file to the current directory from the console, type: $ tar -xvf archive.tar.gz
  4. There can be multiple JDKs installed on the machine and the next step is to configure the runtime environment.
    • Add the GraalVM bin folder to the PATH environment variable: $ export PATH=<path to GraalVM>/Contents/Home/bin:$PATH. Verify whether you are using GraalVM with the echo command: $ echo $PATH.
    • Set the JAVA_HOME environment variable to resolve to the GraalVM installation directory: $ export JAVA_HOME=<path to GraalVM>/Contents/Home.
  5. You can also specify GraalVM as the JRE or JDK installation in your Java IDE.

Optionally set the JAVA_HOME environment variable to resolve to the GraalVM installation directory. You can also specify GraalVM as the JRE or JDK installation in your Java IDE.

Note on macOS java_home command

The information property file, Info.plist, is in the top level Contents folder. This means that GraalVM participates in the macOS specific /usr/libexec/java_home mechanism. Depending on other JDK 8 installation(s) available, it is now possible that /usr/libexec/java_home -v1.8 returns /Library/Java/JavaVirtualMachines/graalvm-ee-19.3.0/Contents/Home. You can run /usr/libexec/java_home -v1.8 -V to see the complete list of 1.8 JVMs available to the java_home command. This command appears to sort the JVMs in decreasing version order and chooses the top one as the default for the specified version. Within a specific version, the sort order appears to be stable but is unspecified.

Since the executables of all language runtimes in GraalVM emulate the behavior of the languages’ default runtimes, setting GraalVM on your PATH should be enough to run an application with GraalVM.

GraalVM’s /bin directory is similar to that of a standard JDK, but includes a set of additional launchers:

  • js runs a JavaScript console with GraalVM.
  • node is a drop-in replacement for Node.js, using GraalVM’s JavaScript engine.
  • lli is a high-performance LLVM bitcode interpreter integrated with GraalVM.
  • gu (GraalVM Updater) can be used to install language packs for Python, R, and Ruby.

Notably, java runs the JVM with GraalVM’s default compiler. The Ruby, Python, R executables become available only if you install the corresponding languages engines. For more information on installing components please refer to GraalVM Updater documentation.

Once the PATH environment variable is set properly, check language versions with GraalVM launchers:

$ java -version
java version "1.8.0_231"
Java(TM) SE Runtime Environment (build 1.8.0_231-b11)
Java HotSpot(TM) 64-Bit GraalVM EE 19.3.0 (build 25.231-b11-jvmci-19.3-b05, mixed mode)

$ node -v
v12.10.0

$ lli --version
LLVM (GraalVM EE Native 19.3.0)

Docker Containers

The official Docker images for GraalVM CE are available from the Docker Hub: https://hub.docker.com/r/oracle/graalvm-ce/.

If you want to use the Docker container with GraalVM CE, use the docker pull command:

docker pull oracle/graalvm-ce:19.3.0

The image is based on Oracle Linux and has GraalVM CE downloaded, unzipped and made available. It means that Java, JavaScript, Node and the LLVM interpreter are available out of the box.

You can start a container and enter the bash session with the following run command:

docker run -it oracle/graalvm-ce:19.3.0 bash

Check that java, js and other commands work as expected.

→ docker run -it oracle/graalvm-ce:19.3.0 bash
bash-4.2# java -version
openjdk version "1.8.0_232"
OpenJDK Runtime Environment (build 1.8.0_232-20191009173705.graal.jdk8u-src-tar-gz-b07)
OpenJDK 64-Bit GraalVM CE 19.3.0 (build 25.232-b07-jvmci-19.3-b05, mixed mode)
bash-4.2# node
> 1 + 1
2
> process.exit()
bash-4.2# lli --version
LLVM (GraalVM CE Native 19.3.0)
bash-4.2#

Please note that the image contains only the components immediately available in the GraalVM CE distribution. However, the GraalVM Updater utility is on the PATH. You can install the support for additional languages like Ruby, R, or Python at will. For example, the following command installs the Ruby support (the output below is truncated for brevity):

docker run -it oracle/graalvm-ce:19.3.0 bash
bash-4.2# gu install ruby
Downloading: Component catalog
Processing component archive: Component ruby
Downloading: Component ruby
[######              ]
...

If you want to mount a directory from the host system to have it locally available in the container, use Docker volumes.

Here is a sample command that maps the /absolute/path/to/dir/no/trailing/slash directory from the host system to the /path/inside/container inside the container.

docker run -it -v /absolute/path/to/dir/no/trailing/slash:/path/inside/container oracle/graalvm-ce:19.3.0 bash

If you want to create docker images that contain GraalVM Ruby, R, or Python implementation, you can use dockerfiles like the example below, which uses oracle/graalvm-ce:19.3.0 as the base image, installs Ruby support using the gu utility, then creates and runs a sample Ruby program.

FROM oracle/graalvm-ce:19.3.0
RUN gu install ruby
WORKDIR /workdir
RUN echo 'puts "Hello from Truffleruby!\nVersion: #{RUBY_DESCRIPTION}"' > app.rb
CMD ruby app.rb

If you put the above snippet in the Dockerfile in the current directory, you can build and run it with the following commands.

docker build -t truffleruby-demo .
...
$ docker run -it --rm truffleruby-demo
Hello from Truffleruby!
Version: truffleruby 19.3.0, like ruby 2.6.2, GraalVM CE Native [x86_64-darwin]

Running Applications

Since the executables of all language runtimes in GraalVM emulate the behavior of the languages’ default runtimes, putting GraalVM on your PATH should be enough to run your applications with GraalVM.

Running Java

Take a look at this typical HelloWorld class:

public class HelloWorld {
  public static void main(String[] args) {
    System.out.println("Hello, World!");
  }
}

Run the following command to compile this class to bytecode and then run it on GraalVM:

$ javac HelloWorld.java
$ java HelloWorld
Hello World!

You can find a collection of larger examples of Java applications you can try running with GraalVM on the examples page.

Running JavaScript

GraalVM can execute plain JavaScript code, both in REPL mode and by executing script files directly.

$ js
> 1 + 2
3

GraalVM also supports running Node.js applications. More than 95,000 npm modules are tested and are fully compatible with GraalVM, including modules like express, react, async, request, browserify, grunt, mocha, and underscore. To install a Node.js module, use the npm executable in the /bin folder of the GraalVM package. The npm command is equivalent to the default Node.js command and supports all Node.js APIs.

1. Install the colors and ansispan modules using npm install:

npm install colors ansispan

After the modules are installed, you can use them from your application.

2. Add the following code snippet to a file named app.js and save it in the same directory where you installed the Node.js modules:

  // RUN-CMD: rm -rf node_modules
// RUN-CMD: npm install ansispan colors
// RUN-CMD: node {file}
// RUN-CMD: rm -r node_modules

// BEGIN-SNIPPET

const http = require("http");
const span = require("ansispan");
require("colors");

http.createServer(function (request, response) {
    response.writeHead(200, {"Content-Type": "text/html"});
    response.end(span("Hello Graal.js!".green));
}).listen(8000, function() { console.log("Graal.js server running at http://127.0.0.1:8000/".red); });
// END-SNIPPET

setTimeout(function() { console.log("DONE!"); process.exit(); }, 2000);
  

3. Execute it on GraalVM using the node command:

node app.js

More information on compatibility with the Node.js and configuring GraalVM read the reference manual on JavaScript in GraalVM.

Running LLVM Bitcode

The GraalVM LLVM runtime can execute C/C++, Rust, and other programming language that can be compiled to LLVM bitcode. A native program has to be compiled to LLVM bitcode using an LLVM frontend such as clang. C/C++ code can be compiled to LLVM bitcode using clang shipped with GraalVM via a pre-built LLVM toolchain.

To set up the LLVM toolchain support for GraalVM, execute the following commands:

$ gu install llvm-toolchain
$ export LLVM_TOOLCHAIN=$(lli --print-toolchain-path)

Put this C code example into a file named hello.c:

#include <stdio.h>

int main() {
    printf("Hello from GraalVM!\n");
    return 0;
}

Then compile hello.c to an executable hello with embedded LLVM bitcode and run it as follows:

$ $LLVM_TOOLCHAIN/clang hello.c -o hello
$ lli hello

More examples and information on running LLVM bitcode with GraalVM can be found in the reference manual for LLVM.

Running Ruby

The Ruby engine is not installed by default, but it can be added using GraalVM Updater:

gu install ruby

The above command will install a community version of a component from GitHub catalog. For GraalVM Enterprise Edition users, a manual component installation is required. Then the Ruby launchers like ruby, gem, irb, rake, rdoc and ri will become available:

$ ruby [options] program.rb

GraalVM Ruby implementation uses the same options as the standard implementation of Ruby, with some additions.

$ gem install chunky_png
$ ruby -r chunky_png -e "puts ChunkyPNG::Color.to_hex(ChunkyPNG::Color('mintcream @ 0.5'))"
#f5fffa80

Using Bundler

GraalVM implementation of Ruby ships with the Bundler environment. Therefore its installation, gem install bundler, is not needed.

$ bundle exec ...

More examples and additional information on Ruby support in GraalVM can be found in the reference manual for Ruby.

Running R

The R engine is not installed by default, and can be added using GraalVM Updater:

gu install R

Please note, the installation of the R language component is possible only from catalog. When the R engine is installed, you can execute R scripts and use the R REPL with GraalVM:

$ R
R version 3.6.1 (FastR)
...

> 1 + 1
[1] 2

More examples and additional information on R support in GraalVM can be found in the reference manual for R.

Running Python

GraalVM implementation of Python 3.7 has recently been started. The Python engine is not available by default, but it can be installed using GraalVM Updater:

gu install python

The above command will install a community version of a component from GitHub catalog. For GraalVM Enterprise Edition users, a manual component installation is required. Once the Python engine is installed, GraalVM can execute Python programs:

$ graalpython
...
>>> 1 + 2
3
>>> exit()

More examples and additional information on Python support in GraalVM can be found in the reference manual for Python.

Combine Languages

If enabled, using the --polyglot flag, scripts executed on GraalVM can use interoperability features to call into other languages and exchange data with them.

For example, running js --jvm --polyglot example.js executes example.js in a polyglot context. If the program calls any code in other supported languages, GraalVM executes that code in the same runtime as the example.js application. For more information on polyglot applications see the polyglot documentation.

Native Images

GraalVM can compile Java bytecode into native images to achieve faster startup and smaller footprint for your applications. The Native Image functionality is not available by default but can be easily installed.

The HelloWorld example from above is used here to demonstrate how to compile Java bytecode into a native image:

// HelloWorld.java
public class HelloWorld {
  public static void main(String[] args) {
    System.out.println("Hello, World!");
  }
}

Run the following to compile the class to bytecode and then build a native image:

$ javac HelloWorld.java
$ native-image HelloWorld

This builds an executable file named helloworld in the current working directory. Invoking it executes the natively compiled code of the HelloWorld class as follows:

$ ./helloworld
Hello, World!

Polyglot Capabilities of Native Images

GraalVM Native Image Generator also makes it easy to use polyglot capabilities. Take this example of a JSON pretty-printer using the GraalVM implementation of JavaScript:

// PrettyPrintJSON.java
import java.io.*;
import java.util.stream.*;
import org.graalvm.polyglot.*;

public class PrettyPrintJSON {
  public static void main(String[] args) throws java.io.IOException {
    BufferedReader reader = new BufferedReader(new InputStreamReader(System.in));
    String input = reader.lines().collect(Collectors.joining(System.lineSeparator()));
    try (Context context = Context.create("js")) {
      Value parse = context.eval("js", "JSON.parse");
      Value stringify = context.eval("js", "JSON.stringify");
      Value result = stringify.execute(parse.execute(input), null, 2);
      System.out.println(result.asString());
    }
  }
}

The --language:js argument ensures that the JavaScript engine is available in the generated image:

$ javac PrettyPrintJSON.java
$ native-image --language:js --initialize-at-build-time PrettyPrintJSON

The native image generation will take several minutes as it does not just build the PrettyPrintJSON class, but includes building a JavaScript engine (along with the support for partial evaluation). Additionally, the image building requires large amounts of physical memory, especially if you build an image with Truffle Language Implementation Framework included, which is exactly the case here. Memory requirements and some other limitations of ahead-of-time compilation with GraalVM are listed here.

The resulting executable can now perform JSON pretty printing:

$ ./prettyprintjson <<EOF
{"GraalVM":{"description":"Language Abstraction Platform","supports":["combining languages","embedding languages","creating native images"],"languages": ["Java","JavaScript","Node.js", "Python", "Ruby","R","LLVM"]}}
EOF

Here is the JSON output from the native executable:

{
  "GraalVM": {
    "description": "Language Abstraction Platform",
    "supports": [
      "combining languages",
      "embedding languages",
      "creating native images"
    ],
    "languages": [
      "Java",
      "JavaScript",
      "Node.js",
      "Python",
      "Ruby",
      "R",
      "LLVM"
    ]
  }
}

The native image is much faster than running the same code on the JVM directly:

$ time bin/java PrettyPrintJSON < test.json > /dev/null
real	0m1.101s
user	0m2.471s
sys	0m0.237s

$ time ./prettyprintjson < test.json > /dev/null
real	0m0.037s
user	0m0.015s
sys	0m0.016s

If you want to learn what GraalVM offers to different types of teams, read the Why GraalVM page. Some of the diverse features of GraalVM are disclosed and supported with examples in Top 10 Things To Do With GraalVM article. Or, you can examine different supported languages in action by looking at example applications. If you want to learn about the common tools GraalVM enables for the supported languages, proceed to the tools section of the reference manual. And if you are mostly interested in a specific language, more extensive documentation is available in the reference manual as well.