You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 23, 2020. It is now read-only.
The greenlet pool is used for consuming implicit asynchronous function in a pool of greenlets.
The performance of consuming in the pool is worse than writing explicit asynchronous code.
The reason for the worse performance is the creation of tasks to run in the greenlet pool.
To improve things we should have a task always running which consume tasks from a queue.
Steps to reproduce
python setup.py bench -a green --sequential --io uv
with results
TestGreenIo.test_green_io: repeated 10(x2000) times, average 0.29259 secs, stdev 2.81 %
TestGreenIo.test_green_pool: repeated 10(x2000) times, average 0.38892 secs, stdev 2.64 %
TestGreenIo.test_yield_io: repeated 10(x2000) times, average 0.27353 secs, stdev 1.9 %
Ran 3 tests in 9.699s
The text was updated successfully, but these errors were encountered:
Description
The greenlet pool is used for consuming implicit asynchronous function in a pool of greenlets.
The performance of consuming in the pool is worse than writing explicit asynchronous code.
The reason for the worse performance is the creation of tasks to run in the greenlet pool.
To improve things we should have a task always running which consume tasks from a queue.
Steps to reproduce
with results
The text was updated successfully, but these errors were encountered: