Message 355395 - Python tracker

➜

This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author	vstinner
Recipients	aeros, asvetlov, vstinner, yselivanov
Date	2019-10-26.00:36:54
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1572050215.17.0.575706189641.issue38591@roundup.psfhosted.org>
In-reply-to

Content
> ThreadedChildWatcher starts a thread per process but has O(1) complexity. But it spawns a new Python thread per process which can be a blocker issue if a server memory is limited. What if you want to spawn 100 processes? Or 1000 processes? What is the memory usage? I like FastChildWatcher! > ... but working with asyncio subprocess API is still super complicated if asyncio code is running from multiple threads Well, I like the ability to choose the child watcher implementation depending on my use case. If asyncio is only run from the main thread, FastChildWatcher is safe, fast and has low memory footprint, no?

> ThreadedChildWatcher starts a thread per process but has O(1) complexity.

But it spawns a new Python thread per process which can be a blocker issue if a server memory is limited. What if you want to spawn 100 processes? Or 1000 processes? What is the memory usage?

I like FastChildWatcher!

> ... but working with asyncio subprocess API is still super complicated if asyncio code is running from multiple threads

Well, I like the ability to choose the child watcher implementation depending on my use case. If asyncio is only run from the main thread, FastChildWatcher is safe, fast and has low memory footprint, no?

History
Date	User	Action	Args
2019-10-26 00:36:55	vstinner	set	recipients: + vstinner, asvetlov, yselivanov, aeros
2019-10-26 00:36:55	vstinner	set	messageid: <1572050215.17.0.575706189641.issue38591@roundup.psfhosted.org>
2019-10-26 00:36:55	vstinner	link	issue38591 messages
2019-10-26 00:36:54	vstinner	create