why remove the python GIL? (intermediate - advanced) anthony explains

Подписаться 39 тыс.

Просмотров 9 тыс.

50% 1

today we talk about the "global interpreter lock" in python and why it currently limits multithreading to 100% usage (and what could be done if it were lifted)
- nogil project: docs.google.com/document/d/18... github.com/colesbury/nogil
playlist: • anthony explains
==========
twitch: / anthonywritescode
dicsord: / discord
twitter: / codewithanthony
github: github.com/asottile
stream github: github.com/anthonywritescode
I won't ask for subscriptions / likes / comments in videos but it really helps the channel. If you have any suggestions or things you'd like to see please comment below!

Наука

Опубликовано:

7 ноя 2021

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 25

@hemis.k3193 Год назад

Kickass explanation! What I am looking for is some good examples where we are stuck due to GIL but using the java implementation of python Jpython the issue got resolved

@stefanAH97 2 года назад

Interesting, high level content. Thnx for sharing

@latitzouri 2 года назад

Removing my grandma in law ? Never thought about it 🤔 could make my python's programming performance increase indeed

@shashishekhar---- Год назад

I am thankful for this, what an amazing way to explain this topic.

@bilbo1337 2 года назад

Since we’re on the topic I think it would be also good to explain the differences with async python

@anthonywritescode 2 года назад

planned at some point!

@rosmelylawliet3919 11 месяцев назад

We need an update/follow up video now :D

@AceofSpades5757 2 года назад

That was informative, thanks. I think asyncio would be a good complement to this topic.

@equus6460 10 месяцев назад

requirement for multy threaded applications in python is limited. This defect of single thread runing at a time even if we write multi threaded apps is overshadowed by having varity of modules in python. multithreading might not be necessary most of the time and if needed could be over come by writing a rust class and wrapping it with python for speed.

@user-kn4wt 11 месяцев назад

thank you!!

@mohamedsaif1928 2 года назад

Thank You

@skarfie123 Год назад

With the new pep getting a lot of buzz, I was looking for an explanation and this was great!

@davititchanturia 2 года назад

Nice One. is there any reason why you are not using autocompletion ?

@anthonywritescode 2 года назад

I haven't implemented it yet

@UriYakir 2 года назад

LOL you should really know by now that Anthony doesn’t need auto completion…

@SpeedingFlare 2 года назад

He's using an editor he made himself called babi

@UriYakir 2 года назад

Could you give a few real world examples of where multithreading would actually be useful in the current state of Python? Thanks for the video!

@anthonywritescode 2 года назад

depends... any place where you'd want to do any sort of parallel processing

@fiancheg 2 года назад

IO operations: file IO, http requests, etc.

@KSriram 2 года назад

@@fiancheg The GIL doesn't lock IO-bound threads. The main advantage of a nogil would be to CPU-bound parallel threads.

@DagarCoH 9 месяцев назад

Have that a bunch with software that connect different sensors and actors to run complex processes. The threads wait for sensor / system input and change the system or the actor state accordingly. In between there are short sleep cycles after which each thread polls for new data / state change

@stiffyBlicky 2 года назад

So using one thread in your pool, implies that your work is running serially right?

@anthonywritescode 2 года назад

a ThreadPool of size 1 would still allow work to happen in the main thread as well (main + 1 background thread) -- but also wouldn't be too terribly useful