What is the difference between ExecutorService and ForkJoinPool?

The 30 second answer

ExecutorService is the general interface for running tasks on a pool; the classic version, ThreadPoolExecutor, has workers that take tasks from one shared queue. ForkJoinPool is one implementation built for work stealing: each worker has its own queue, and idle workers steal tasks from busy ones. It fits divide-and-conquer and many small tasks. Parallel streams and default CompletableFuture async calls use its common pool.

ExecutorService io = Executors.newFixedThreadPool(32);            // ThreadPoolExecutor
ExecutorService vt = Executors.newVirtualThreadPerTaskExecutor(); // Java 21

class Sum extends RecursiveTask<Long> {
    final long[] a; final int lo, hi;
    Sum(long[] a, int lo, int hi) { this.a = a; this.lo = lo; this.hi = hi; }
    protected Long compute() {
        if (hi - lo < 10_000) { long s = 0; for (int i = lo; i < hi; i++) s += a[i]; return s; }
        int mid = (lo + hi) >>> 1;
        Sum left = new Sum(a, lo, mid); left.fork();
        return new Sum(a, mid, hi).compute() + left.join();
    }
}
long total = ForkJoinPool.commonPool().invoke(new Sum(arr, 0, arr.length));

ThreadPoolExecutor: core/max pool size, a work queue and a rejection policy. Good for independent, coarse tasks, including blocking ones.
ForkJoinPool: per-worker deques; a worker processes its own tasks and steals from others when idle. Good for recursive splitting. It also implements ExecutorService.
The common pool's default parallelism is one less than the number of available processors. Blocking inside it slows every parallel stream in the JVM. [Opinion] Use separate executors for blocking I/O.

Sources: ForkJoinPool Javadoc, ExecutorService Javadoc, ThreadPoolExecutor Javadoc

Sources (0)

See inline citations in the answer above.