RecursiveTask is an abstract class encapsulates a task that returns a result. It is a subclass of ForkJoinTask. The RecursiveTask class is extended to create a task that has a particular return type. The code that represents the computational portion of the task is kept within the compute() method of RecursiveTask.
RecursiveTask class is mostly used in the context of parallel programming. Tasks that can be divided into independent subtasks and the final outcome of the task can be obtained from the outcomes of the subtask, can be implemented more efficiently using RecursiveTask. For example, searching for an element in a large array.
Class Hierarchy
java.lang.Object ↳ java.util.concurrent.ForkJoinTask ↳ java.util.concurrent.RecursiveTask<V>
Constructor
- RecursiveTask()– Creates an object of RecursiveTask with default settings.
public RecursiveTask()
Methods
- compute()– The method that defines the task.
protected abstract void compute()
- exec()– This method implements the basic rules necessary for the execution of a task.
protected final boolean exec()
- getRawResult()– The function returns the value obtained after the completion of the task, even if the task is completed abnormally. It returns null, if the task is not yet completed.
public final Void getRawResult()
- setRawResult()– The function sets the return value of the task to the value passed in the argument.
protected final void setRawResult(Void mustBeNull)
Example to demonstrate RecursiveTask
import java.util.concurrent.ForkJoinPool;import java.util.concurrent.RecursiveTask;  public class RecursiveTaskDemo {    public static void main(String[] args)    {        ForkJoinPool fjp = new ForkJoinPool();          double[] nums = new double[5000];        for (int i = 0; i < nums.length; i++) {            nums[i] = (double)(((i % 2) == 0) ? i : -1);        }        Sum task = new Sum(nums, 0, nums.length);        double summation = fjp.invoke(task);        System.out.println("Summation " + summation);    }}  class Sum extends RecursiveTask<Double> {    final int seqThreshold = 500;    double[] data;    int start, end;      Sum(double[] data, int start, int end)    {        this.data = data;        this.start = start;        this.end = end;    }      @Override    protected Double compute()    {        double sum = 0;        if ((end - start) < seqThreshold) {            for (int i = start; i < end; i++)                sum += data[i];        }        else {            int middle = (start + end) / 2;              Sum subtaskA = new Sum(data, start, middle);            Sum subtaskB = new Sum(data, middle, end);              subtaskA.fork();            subtaskB.fork();              sum += subtaskA.join() + subtaskB.join();        }        return sum;    }} |
Summation 6245000.0
Reference: https://docs.oracle.com/javase/8/docs/api/java/util/concurrent/RecursiveTask.html
