what is the advantage of using FutureTask over Callable?

Question

There are two approaches to submitting and polling task for result

FutureTask futureTask = new FutureTask<String>(callable);

1. Use combination of Callable and Future and submit on ExecutorService. Retrieve result using future.get().

   Future future = service.submit(callable);
   

2. Use FutureTask. This will wrap Callable and then retrieve result using FutureTask.

   service.execute(task);
   

What is the advantage of using FutureTask over Callable + Future combination ?

Scribe

The FutureTask implements the Future interface, so what exactly is the difference between approaches in concrete terms? (i.e. show an example) – Nez

FutureTask task = new FutureTask<String>(callable); //Implementation 1 Future fut = service.submit(callable); //Implementation 2. service.execute(task); why should we prefer Implementation 1 over Implementation 2. Is there any advantage ? – Scribe

Don't post your code in comments, update the answer. – Nez

@AmrishPandey Advantage? DougSchmidt and brianGoetz talk about Memoizer class(section 5.6 - JCIP) using FutureTask. DougSchmidt says, briangoetz's version of Memoizer class has some bug, in this presentation. Take a look. – Roundhouse

@AmrishPandey Why Memoizer? A caching technique – Roundhouse

Answer 1

Almost certainly none at all. A quick browse on GrepCode of the AbstractExecutorService shows each of these methods are simply helper methods that ultimately wrap the Callable/Runnable in a Future for you.

protected <T> RunnableFuture<T> newTaskFor(Runnable runnable, T value) {
    return new FutureTask<T>(runnable, value);
}

protected <T> RunnableFuture<T> newTaskFor(Callable<T> callable) {
    return new FutureTask<T>(callable);
}

public Future<?> submit(Runnable task) {
    // ...
    RunnableFuture<Object> ftask = newTaskFor(task, null);
    execute(ftask);
    return ftask;
}

public <T> Future<T> submit(Runnable task, T result) {
    // ...
    RunnableFuture<T> ftask = newTaskFor(task, result);
    execute(ftask);
    return ftask;
}

public <T> Future<T> submit(Callable<T> task) {
    // ...
    RunnableFuture<T> ftask = newTaskFor(task);
    execute(ftask);
    return ftask;
}

Ctenidium

Answer 2

Using Future we can find out the status of the Callable task and get the returned Object. It provides get() method that can wait for the Callable to finish and then return the result.

Future provides cancel() method to cancel the associated Callable task. There is an overloaded version of get() method where we can specify the time to wait for the result, it’s useful to avoid current thread getting blocked for longer time. There are isDone() and isCancelled() methods to find out the current status of associated Callable task.

Here is a simple example of Callable task that returns the name of thread executing the task after one second. We are using Executor framework to execute 100 tasks in parallel and use Future to get the result of the submitted tasks.

    import java.util.ArrayList;
    import java.util.Date;
    import java.util.List;
    import java.util.concurrent.Callable;
    import java.util.concurrent.ExecutionException;
    import java.util.concurrent.ExecutorService;
    import java.util.concurrent.Executors;
    import java.util.concurrent.Future;

    public class MyCallable implements Callable<String> {

        @Override
        public String call() throws Exception {
            Thread.sleep(1000);
            //return the thread name executing this callable task
            return Thread.currentThread().getName();
        }

        public static void main(String args[]){
            //Get ExecutorService from Executors utility class, thread pool size is 10
            ExecutorService executor = Executors.newFixedThreadPool(10);
            //create a list to hold the Future object associated with Callable
            List<Future<String>> list = new ArrayList<Future<String>>();
            //Create MyCallable instance
            Callable<String> callable = new MyCallable();
            for(int i=0; i< 100; i++){
                //submit Callable tasks to be executed by thread pool
                Future<String> future = executor.submit(callable);
                //add Future to the list, we can get return value using Future
                list.add(future);
            }
            for(Future<String> fut : list){
                try {
                    //print the return value of Future, notice the output delay in console
                    // because Future.get() waits for task to get completed
                    System.out.println(new Date()+ "::"+fut.get());
                } catch (InterruptedException | ExecutionException e) {
                    e.printStackTrace();
                }
            }
            //shut down the executor service now
            executor.shutdown();
        }
    }

Where as FutureTask is base concrete implementation of Future interface and provides asynchronous processing. It contains the methods to start and cancel a task and also methods that can return the state of the FutureTask as whether it’s completed or cancelled. We need a callable object to create a future task and then we can use Java Thread Pool Executor to process these asynchronously.

Let’s see the example of FutureTask with a simple program.

Since FutureTask requires a callable object, we will create a simple Callable implementation.

    public class MyCallable implements Callable<String> {

    private long waitTime;

    public MyCallable(int timeInMillis){
        this.waitTime=timeInMillis;
    }
    @Override
    public String call() throws Exception {
        Thread.sleep(waitTime);
        //return the thread name executing this callable task
        return Thread.currentThread().getName();
    }

}

    import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.FutureTask;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.TimeoutException;

public class FutureTaskExample {

    public static void main(String[] args) {
        MyCallable callable1 = new MyCallable(1000);
        MyCallable callable2 = new MyCallable(2000);

        FutureTask<String> futureTask1 = new FutureTask<String>(callable1);
        FutureTask<String> futureTask2 = new FutureTask<String>(callable2);

        ExecutorService executor = Executors.newFixedThreadPool(2);
        executor.execute(futureTask1);
        executor.execute(futureTask2);

        while (true) {
            try {
                if(futureTask1.isDone() && futureTask2.isDone()){
                    System.out.println("Done");
                    //shut down executor service
                    executor.shutdown();
                    return;
                }

                if(!futureTask1.isDone()){
                //wait indefinitely for future task to complete
                System.out.println("FutureTask1 output="+futureTask1.get());
                }

                System.out.println("Waiting for FutureTask2 to complete");
                String s = futureTask2.get(200L, TimeUnit.MILLISECONDS);
                if(s !=null){
                    System.out.println("FutureTask2 output="+s);
                }
            } catch (InterruptedException | ExecutionException e) {
                e.printStackTrace();
            }catch(TimeoutException e){
                //do nothing
            }
        }

    }

}

Miles

These are just API differences. Functionally , FutureTask does not seem to offer any advantage . – Scribe

Rather than advantage it's more to do it with utility and the situation requirement. You would only need to use FutureTask if you want to change its behaviour or access its Callable later but a good reason to let the Executor construct the FutureTask for you is to ensure that there is no possible way more than one reference exists to the FutureTask instance. That is, the Executor owns this instance. Majority of times, we use Callable and Future. – Miles

Answer 3

FutureTask<T> class contains an additional " done()" method so we can override the done() method, then add the FutureTask object to the ExecutorService, so the done() method will invoke when the FutureTask completed immediately.

Dogtooth