Fix async clash with dask #12
Merged
Conversation
This file contains 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
|
I ran a quick smoke test locally, and things seem to be moving in the right direction: With @lsetiawan changes, I am noticing some asynchronicity + parallelism on the server side. I am all for merging this for now, and I will work on getting #10 done so that we can run some benchmarks.
|
andersy005
approved these changes
Mar 3, 2020
|
Thank you for the clarification, too! I am amazed by how easy of a fix this is. I am learning quite a lot about Python's async features :) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Overview
This PR fixes #7.
Problem
It appears that there was a clash/confusion between the uvicorn event loop and the dask distributed event loop. When using
asyncin a FastAPI app, this method will be sent to the uvicorn event loop. So the request for individual chunks were processed concurrently. Each dask compute becomes blocking and is only processed by one worker which I think is part of the uvicron event loop.Solution
To allow dask to actually compute in parallel the async and await for
get_keyshould be remove, which frees up this method to use the compute within dask distributed event loop rather than the uvicorn event loop.