Fix monitor.py bottleneck by removing excess Redis queries. #1786
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ericl
approved these changes
Mar 27, 2018
| redis_port = get_port(args.redis_address) | ||
|
|
||
| # Initialize the global state. | ||
| ray.global_state._initialize_global_state(redis_ip_address, redis_port) |
Contributor
There was a problem hiding this comment.
Is this since we don't need to get the client table any more?
Collaborator
Author
There was a problem hiding this comment.
I'm not 100% sure, but I think that was always redundant because the monitor creates its own global state at
Lines 85 to 86 in 51fdbe3
python/ray/monitor.py
Outdated
| for ls in local_schedulers: | ||
| if ls["DBClientID"] == client_id: | ||
| ip = ls["AuxAddress"].split(":")[0] | ||
| ip = self.local_scheduler_id_to_ip_map.get(client_id, None) |
Contributor
There was a problem hiding this comment.
nit: None is already the default for get
Collaborator
Author
There was a problem hiding this comment.
Oh, thanks!
|
Test PASSed. |
|
Test PASSed. |
royf
added a commit
to royf/ray
that referenced
this pull request
Apr 22, 2018
* commit 'f69cbd35d4e86f2a3c2ace875aaf8166edb69f5d': (64 commits) Bump version to 0.4.0. (ray-project#1745) Fix monitor.py bottleneck by removing excess Redis queries. (ray-project#1786) Convert the ObjectTable implementation to a Log (ray-project#1779) Acquire worker lock when importing actor. (ray-project#1783) Introduce a log interface for the new GCS (ray-project#1771) [tune] Fix linting error (ray-project#1777) [tune] Added pbt with keras on cifar10 dataset example (ray-project#1729) Add a GCS table for the xray task flatbuffer (ray-project#1775) [tune] Change tune resource request syntax to be less confusing (ray-project#1764) Remove from X import Y convention in RLlib ES. (ray-project#1774) Check if the provider is external before getting the config. (ray-project#1743) Request and cancel notifications in the new GCS API (ray-project#1758) Fix resource bookkeeping for blocked actor methods. (ray-project#1766) Fix bug when connecting another driver in local case. (ray-project#1760) Define string prefixes for all tables in the new GCS API (ray-project#1755) [rllib] Update RLlib to work with new actor scheduling behavior (ray-project#1754) Redirect output of all processes by default. (ray-project#1752) Add API for getting total cluster resources. (ray-project#1736) Always send actor creation tasks to the global scheduler. (ray-project#1757) Print error when actor takes too long to start, and refactor error me… (ray-project#1747) ... # Conflicts: # python/ray/rllib/__init__.py # python/ray/rllib/dqn/dqn.py # python/ray/rllib/dqn/dqn_evaluator.py # python/ray/rllib/dqn/dqn_replay_evaluator.py # python/ray/rllib/optimizers/__init__.py # python/ray/rllib/tuned_examples/pong-dqn.yaml
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
With large clusters,
monitor.pyis at very high utilization. Using cProfile, I see that it's spending all of it's time in the lineray/python/ray/monitor.py
Line 272 in 51fdbe3
This call does N Redis queries, where N is the number of nodes in the cluster. It is currently called once per heartbeat. There are 10N heartbeats per second, for a total of 10N^2 Redis queries per second (400K for a 200 node cluster). This will prevent the monitor from being able to keep up with all of the heartbeats.
As a separate note, the main loop in
monitor.pyisray/python/ray/monitor.py
Lines 517 to 526 in 51fdbe3
This code assumes that the call to
self.process_messages()will return. However,proess_messagesruns its own infinite loop which will never return for large clusters (i.e., it only returns when there are no more messages left to process).To address this, I force
process_messagesto return after processing a maximum of 10K messages.cc @ericl