Utility classes#

This page describes ActorPool, which load-balances method calls across a pool of actors, and Ray Queue, which passes messages between tasks and actors.

Actor pool#

Python

The ray.util module contains ActorPool, a utility class similar to multiprocessing.Pool. Use it to schedule Ray tasks over a fixed pool of actors.

import ray
from ray.util import ActorPool


@ray.remote
class Actor:
    def double(self, n):
        return n * 2


a1, a2 = Actor.remote(), Actor.remote()
pool = ActorPool([a1, a2])

# pool.map(..) returns a Python generator object ActorPool.map
gen = pool.map(lambda a, v: a.double.remote(v), [1, 2, 3, 4])
print(list(gen))
# [2, 4, 6, 8]

See the package reference for more information.

Java

Actor pool isn’t available in Java yet.

C++

Actor pool isn’t available in C++ yet.

Message passing using Ray Queue#

One signal isn’t always enough for synchronization. To send data among many tasks or actors, use ray.util.queue.Queue.

import ray
from ray.util.queue import Queue, Empty

ray.init()
# You can pass this object around to different tasks/actors
queue = Queue(maxsize=100)


@ray.remote
def consumer(id, queue):
    try:
        while True:
            next_item = queue.get(block=True, timeout=1)
            print(f"consumer {id} got work {next_item}")
    except Empty:
        pass


[queue.put(i) for i in range(10)]
print("Put work 1 - 10 to queue...")

consumers = [consumer.remote(id, queue) for id in range(2)]
ray.get(consumers)

The Ray Queue API is similar to Python’s asyncio.Queue and queue.Queue.