hiro/ray
1
0
Fork 0
mirror of https://github.com/vale981/ray synced 2025-03-06 10:31:39 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[docs] Don't recommend SignalActor from test_utils

Browse source
This commit is contained in:
Edward Oakes 2021-09-27 14:08:37 -05:00 • committed by GitHub
parent cbd7dc749c
commit 627e475cc1
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 11 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

17
doc/source/advanced.rst
View file

@ -42,17 +42,23 @@ This often occurs for data loading and preprocessing.
# hi there!
# hi there!
Multi-node synchronization using ``SignalActor``
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Multi-node synchronization using an Actor
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When you have multiple tasks that need to wait on some condition, you can use a ``SignalActor`` to coordinate.
When you have multiple tasks that need to wait on some condition or otherwise
need to synchronize across tasks & actors on a cluster, you can use a central
actor to coordinate among them. Below is an example of using a ``SignalActor``
that wraps an ``asyncio.Event`` for basic synchronization.
.. code-block:: python
# Also available via `from ray._private.test_utils import SignalActor`
import ray
import asyncio
import ray
ray.init()
# We set num_cpus to zero because this actor will mostly just block on I/O.
@ray.remote(num_cpus=0)
class SignalActor:
def __init__(self):
@ -73,7 +79,6 @@ When you have multiple tasks that need to wait on some condition, you can use a
print("go!")
ray.init()
signal = SignalActor.remote()
tasks = [wait_and_go.remote(signal) for _ in range(4)]
print("ready...")