Executing Cucumber Feature Files in Parallel

Introduction

In modern testing environments, parallel test execution can significantly improve the efficiency and speed of testing processes. Cucumber, a popular behavior-driven development (BDD) framework, allows for parallel execution of feature files.

Plugin-Based Approach

To achieve parallel execution in Cucumber, you can use the cucumber-jvm-parallel-plugin. This plugin dynamically creates test runner classes that can be executed in parallel.

Configuration

1. Add the Plugin to pom.xml:

  com.github.temyers
  cucumber-jvm-parallel-plugin
  2.1.0

2. Configure the Plugin in pom.xml:

  com.github.temyers
  cucumber-jvm-parallel-plugin
  2.1.0
  
    
      generateRunners
      generate-test-sources
      
        generateRunners
      
      
        foo, bar
        ${project.build.directory}/generated-test-sources/cucumber
        src/test/resources/features/
        target/cucumber-parallel
        json
      
    
  

Invoke Generated Runner Classes

Add a Maven Surefire plugin to invoke the generated runner classes in parallel:

  org.apache.maven.plugins
  maven-surefire-plugin
  2.19
   
    5
    true
    
      **/*IT.class
    
  

Shared WebDriver

To execute tests in parallel, the WebDriver instance must be shared and not explicitly closed within the tests. The SharedDriver class achieves this:

public class SharedDriver extends EventFiringWebDriver {
    private static WebDriver REAL_DRIVER = null;

    static {
        Runtime.getRuntime().addShutdownHook(CLOSE_THREAD);
    }

    public SharedDriver() {
        super(CreateDriver());
    }

    public static WebDriver CreateDriver() {
        WebDriver webDriver;
        if (REAL_DRIVER == null)
            webDriver = new FirefoxDriver();
        setWebDriver(webDriver);
        return webDriver;
    }
}

Additional Considerations

• Parallel Execution and Grid: For optimal performance, use a grid with sufficient nodes (browsers registered with the hub).
• Hub Memory: Increase the pool size (e.g., -DPOOL_MAX=512) for Hub installations with a high number of nodes (50 ).