Async Python is still confusing af - when do I need it, what happens under the hood, does it actually help with performance, sometimes the GIL comes into play and sometimes it doesn't, why do we ever use threads at all if there's a GIL, why is it called asyncio if we can use it for anything. My mind is kind of scattered and people seems to be using a lot of async Python for some reason.
I agree that especially within the standard context of python and its syntax, async seems weird (because it's sprinkled into an existing paradigm).
The best mental model for me always was to think: Here's an await, that means "interpreter, go and do something else that's currently waiting while this is not done yet."
And that's all about IO, because what you can wait on is essentially IO.
By the way, I really wish there was a better story to executors in async python. To think we still have the same queue/pickle-based multiprocessing to send async tasks to another core is kind of sad. Hoping for 3.12 and beyond there.
[edit] one really neat example that helped me get asyncio was Guido van Rossum's crawler in 500 lines of python [1]. A lot of the syntax is deprecated now, but it's still a great walk-through
It's great for when you can do concurrent I/O tasks. For example running a web backend, web scraping, or many API calls. FastAPI (and Starlette that it's built off of) is an async web framework and in my experience performs well.
Basically, your program will normally halt when doing I/O, and won't proceed until that I/O is done. During that halt, your program is doing nothing (no cpu being used). With asyncio, you can schedule multiple tasks to run, so if one is halted doing I/O, another can run.
Edit: And AFAIK, the GIL does not come into play at all with async. Only when multithreading.
> the GIL does not come into play at all with async.
In the sense that the GIL is still held and you can have at most one path of Python code executing at a time regardless of whether you use async or threads, sure. Most blocking I/O was already releasing the GIL so the difference is purely in how you can design your modules; for any reasoning about performance the GIL behaves the same way whether you use asyncio or not.
It makes concurrent programming much simpler than using threads.
Very few locking and care is needed with asyncio, as opposed to using threads. Race conditions are basically not a thing if you write reasonably idiomatic code.
It might be (or not) faster than using threads, but that's not the main benefit in my view - this easiness of use is.
It's basically a fancy way to do epoll() around file descriptors, but hide the need to keep a main loop and state, and keep it hidden as functions that run in fake concurrency, stopping whenever they block, and being executed again when their file descriptor has activity.
It doesn't necessarily improve performance. It's just a much easier way to do non-blocking I/O (note that blocking and threaded I/O is easier to do but much much heavier).
In personal experience, when you write anything more complicated using threads in python, I end up writing lots of boilerplate code that I wished I could just shove somewhere - Events, 1 element Queues, ContextVar contexts, all of which don't benefit for being named and make appearance only in two spots in code. Async removed a lot of this for me while also unifying the future ecosystem and making my code more composable and easy to integrate.
When working in a fully async context, it becomes very logical and natural. A great example of this is the FastAPI web framework. You never write any top-level code, only functions that the framework calls, so you never have to deal with the event loop directly. You basically just sprinkle some async and await keywords around your IO bottlenecks and things suddenly run smoother.
> Async Python is still confusing af - when do I need it,
It's needed when you're spending a lot of time waiting for an I/O request to complete (network/HTTP requests, disk reads/writes, database reads/writes)
(Please read the entire blog post - It goes through the necessary concepts like generators, event loops, & coroutines)
> does it actually help with performance,
Refer to the first answer: You'll see improved performances if your workloads are mainly comprised of waiting for other stuff to complete. If you're compute-heavy, it'll be better to use the 'multiprocessing' library instead.
> sometimes the GIL comes into play and sometimes it doesn't,
The GIL comes into play when you have a lot of compute-heavy tasks: Otherwise, you'll rarely encounter it.
It's only when you have that many compute-heavy tasks that you start to use the 'multiprocessing' library.
> why do we ever use threads at all if there's a GIL,
Threads exist because it was there before asyncio & event loops came into Python.
> why is it called asyncio if we can use it for anything.
Its name came from PEP 3156, proposing the asyncio library back in 2012.
As for why, asynchronous I/O stands in contrast to synchronous I/O, where the program/thread had to wait for the I/O request to complete before it can do anything else. Making tasks asynchronous allows it to do other stuff while it waits for a task's request to complete, increasing CPU & I/O utilization.
> My mind is kind of scattered and people seems to be using a lot of async Python for some reason. Any good resources to clear things up?
Highly recommend this video from mcoding: It's fairly simple & goes through a sample implementation.
Async looks like parallelism, but its just smart scheduling. The core concept of async is saying "hey, im waiting for something to complete thats not under my control (like waiting for data on a socket to be able to be read), go ahead and do other things in the mean time".
If you ever coded in sockets in C (and its a good exercise to do so), you probably have at some point ran across `select` which is essentially a non blocking way to check which sockets have data available to read, and then sequentially read the data. This gives the ability for a program to appear parallelized in the sense that it can handle multiple client connections, but its not truly parallel. Different clients can be handled at different time depending on which order they connect, which is asynchronous in nature (versus processing each client in sequence and waiting on each one to connect and disconnect before moving on to the next one)
Async in Python is basically this concept, with a core fundamental feature of time limited execution. Functions can say that they are pausing for x seconds, allowing other functions to run, or functions can say that they give a certain function x seconds to run before resuming execution. If you async code (along with any library you may use) doesn't contain any sleeps or timeouts, its exactly equivalent to synchronous code (since the event loop never really recieves a message that it can suspend a routine or cancel it). With sleeps and timeouts, you gain control over things that can potentially block, both from a caller perspective of not having a function call block your own, and from a callee perspective of not making your function blocking.
The use case is for it is that it is good for I/O bound operations like Threading is, but with the addition that you don't have to worry about synchronization or race conditions, since by design your code will have predictable access patterns. The downside is that your code and any libraries that you use within your code has to be implemented as async libraries, and any library that is async has to have async wrappers around the calls to its methods, which in turn means that your entire code has to be async.
Threading with Python is generally not useful, as its not true parallelism because of GIL. GIL allows only one thread in Python to run. Threading is safer in Python because of this, however it obviously has drawbacks. In general its best used if you want asyncio like performance with a library that is not written with async, since GIL is smart enough to detect when a thread is waiting for input and switch context.
True parallelism in Python is achieved with multiprocessing, however the use case is a little different. Rather than spinning off processes, you generally launch a bunch of worker processes up front (to avoid the larger overhead), then use smart scheduling to distribute work between these processes. Here though you do have to worry about race conditions and synchronization, and use things like locks and mutexes.
> Clearly create_task is as close as you get to free in the Python world, and I would need to look elsewhere for optimizations. Turns out Textual spends far more time processing CSS rules than creating tasks (obvious in retrospect).
Takeaways:
1. Creating async tasks is cheap.
2. It is important to confirm intuitions, before acting on them.
Save a reference to the result of this function, to avoid a task disappearing mid-execution. The event loop only keeps weak references to tasks. A task that isn’t referenced elsewhere may get garbage collected at any time, even before it’s done."
If you can, best is to always spawn them in a task group (either using anyio or Python 3.11's task groups).
This prevents tasks from being garbage collected, but also prevents situations where components can create tasks that outlive their own lifetime. Plus, it's a saner approach when dealing with exception handling and cancellation.
Perhaps I just don't get them, but task groups never really made sense to me.
The whole beauty of async tasks is that you can spawn, retry, and consume them lazily. When you create a task group, you again end up waiting on a single long-running last task, desperately trying to fix individual failures and retries that hold up the entire group.
Python is my strongest language. If you ask me to write some asynchronous code I will try my best not to write it in Python. Usually it's just not worth it.
Yeah, I did the exact same thing and got very similar results. Just to save people clicking through, the Go/goroutine version is 25x as fast as Python. :-)
I preferred gevent (it's been probably 10 years since I've used it.) Yes, you need a ton of monkey patching, etc... but it was less intrusive once you had everything set up. Sprinkling await and async everywhere always struck me as inelegant.
Readit News
Any good resources to clear things up?
The best mental model for me always was to think: Here's an await, that means "interpreter, go and do something else that's currently waiting while this is not done yet."
And that's all about IO, because what you can wait on is essentially IO.
By the way, I really wish there was a better story to executors in async python. To think we still have the same queue/pickle-based multiprocessing to send async tasks to another core is kind of sad. Hoping for 3.12 and beyond there.
[edit] one really neat example that helped me get asyncio was Guido van Rossum's crawler in 500 lines of python [1]. A lot of the syntax is deprecated now, but it's still a great walk-through
[1] http://aosabook.org/en/500L/a-web-crawler-with-asyncio-corou...
Basically, your program will normally halt when doing I/O, and won't proceed until that I/O is done. During that halt, your program is doing nothing (no cpu being used). With asyncio, you can schedule multiple tasks to run, so if one is halted doing I/O, another can run.
Edit: And AFAIK, the GIL does not come into play at all with async. Only when multithreading.
In the sense that the GIL is still held and you can have at most one path of Python code executing at a time regardless of whether you use async or threads, sure. Most blocking I/O was already releasing the GIL so the difference is purely in how you can design your modules; for any reasoning about performance the GIL behaves the same way whether you use asyncio or not.
Very few locking and care is needed with asyncio, as opposed to using threads. Race conditions are basically not a thing if you write reasonably idiomatic code.
It might be (or not) faster than using threads, but that's not the main benefit in my view - this easiness of use is.
It doesn't necessarily improve performance. It's just a much easier way to do non-blocking I/O (note that blocking and threaded I/O is easier to do but much much heavier).
It's needed when you're spending a lot of time waiting for an I/O request to complete (network/HTTP requests, disk reads/writes, database reads/writes)
> what happens under the hood,
https://tenthousandmeters.com/blog/python-behind-the-scenes-...
(Please read the entire blog post - It goes through the necessary concepts like generators, event loops, & coroutines)
> does it actually help with performance,
Refer to the first answer: You'll see improved performances if your workloads are mainly comprised of waiting for other stuff to complete. If you're compute-heavy, it'll be better to use the 'multiprocessing' library instead.
> sometimes the GIL comes into play and sometimes it doesn't,
The GIL comes into play when you have a lot of compute-heavy tasks: Otherwise, you'll rarely encounter it.
It's only when you have that many compute-heavy tasks that you start to use the 'multiprocessing' library.
> why do we ever use threads at all if there's a GIL,
Threads exist because it was there before asyncio & event loops came into Python.
> why is it called asyncio if we can use it for anything.
Its name came from PEP 3156, proposing the asyncio library back in 2012.
https://peps.python.org/pep-3156/
As for why, asynchronous I/O stands in contrast to synchronous I/O, where the program/thread had to wait for the I/O request to complete before it can do anything else. Making tasks asynchronous allows it to do other stuff while it waits for a task's request to complete, increasing CPU & I/O utilization.
> My mind is kind of scattered and people seems to be using a lot of async Python for some reason. Any good resources to clear things up?
Highly recommend this video from mcoding: It's fairly simple & goes through a sample implementation.
https://www.youtube.com/watch?v=ftmdDlwMwwQ
Also, this article:
https://realpython.com/async-io-python/
If you ever coded in sockets in C (and its a good exercise to do so), you probably have at some point ran across `select` which is essentially a non blocking way to check which sockets have data available to read, and then sequentially read the data. This gives the ability for a program to appear parallelized in the sense that it can handle multiple client connections, but its not truly parallel. Different clients can be handled at different time depending on which order they connect, which is asynchronous in nature (versus processing each client in sequence and waiting on each one to connect and disconnect before moving on to the next one)
Async in Python is basically this concept, with a core fundamental feature of time limited execution. Functions can say that they are pausing for x seconds, allowing other functions to run, or functions can say that they give a certain function x seconds to run before resuming execution. If you async code (along with any library you may use) doesn't contain any sleeps or timeouts, its exactly equivalent to synchronous code (since the event loop never really recieves a message that it can suspend a routine or cancel it). With sleeps and timeouts, you gain control over things that can potentially block, both from a caller perspective of not having a function call block your own, and from a callee perspective of not making your function blocking.
The use case is for it is that it is good for I/O bound operations like Threading is, but with the addition that you don't have to worry about synchronization or race conditions, since by design your code will have predictable access patterns. The downside is that your code and any libraries that you use within your code has to be implemented as async libraries, and any library that is async has to have async wrappers around the calls to its methods, which in turn means that your entire code has to be async.
Threading with Python is generally not useful, as its not true parallelism because of GIL. GIL allows only one thread in Python to run. Threading is safer in Python because of this, however it obviously has drawbacks. In general its best used if you want asyncio like performance with a library that is not written with async, since GIL is smart enough to detect when a thread is waiting for input and switch context.
True parallelism in Python is achieved with multiprocessing, however the use case is a little different. Rather than spinning off processes, you generally launch a bunch of worker processes up front (to avoid the larger overhead), then use smart scheduling to distribute work between these processes. Here though you do have to worry about race conditions and synchronization, and use things like locks and mutexes.
Takeaways:
1. Creating async tasks is cheap. 2. It is important to confirm intuitions, before acting on them.
Be careful to hold your references, because async tasks without active references will be garbage collected. I've been bitten by that in the past.
Long discussion here: https://bugs.python.org/issue21163
Docs: https://docs.python.org/3/library/asyncio-task.html#asyncio....
"Important
Save a reference to the result of this function, to avoid a task disappearing mid-execution. The event loop only keeps weak references to tasks. A task that isn’t referenced elsewhere may get garbage collected at any time, even before it’s done."
This prevents tasks from being garbage collected, but also prevents situations where components can create tasks that outlive their own lifetime. Plus, it's a saner approach when dealing with exception handling and cancellation.
The whole beauty of async tasks is that you can spawn, retry, and consume them lazily. When you create a task group, you again end up waiting on a single long-running last task, desperately trying to fix individual failures and retries that hold up the entire group.
I found this article: https://blog.dalibo.com/2022/09/12/monitoring-python-subproc... and while async/await syntax is the same, it's not entirely clear for me, why there's some event loop and what exactly happens, when I pass function to asyncio.run(), like here: https://github.com/pallets/click/issues/85#issuecomment-5034...
So, you can use it and it's not that hard, but there are some parts that are vague for me, no matter which language implements async support.
Edit: Updating per request below.
Tested on a TR2950x. I wonder if NUMA issues in my case or bad python version (3.9.4).
Python
.NET 6EDIT. My results on a 5950x (undervolted)
python3.8.exe test.py
100,000 tasks 139,130 tasks per/s
200,000 tasks 121,905 tasks per/s
300,000 tasks 120,000 tasks per/s
400,000 tasks 114,286 tasks per/s
500,000 tasks 119,403 tasks per/s
600,000 tasks 117,073 tasks per/s
700,000 tasks 130,612 tasks per/s
800,000 tasks 122,488 tasks per/s
900,000 tasks 120,000 tasks per/s
1,000,000 tasks 110,155 tasks per/s
python3.11.exe .\test.py
100,000 tasks 206,452 tasks per/s
200,000 tasks 185,507 tasks per/s
300,000 tasks 186,408 tasks per/s
400,000 tasks 179,021 tasks per/s
500,000 tasks 167,539 tasks per/s
600,000 tasks 177,778 tasks per/s
700,000 tasks 188,235 tasks per/s
800,000 tasks 180,919 tasks per/s
900,000 tasks 168,421 tasks per/s
.\test.exe (go 1.20 compiled)
100000 tasks 2710563.336378 tasks per/s
200000 tasks 3076885.207567 tasks per/s
300000 tasks 3332292.917434 tasks per/s
400000 tasks 3040479.422795 tasks per/s
500000 tasks 2810232.844653 tasks per/s
600000 tasks 3004138.200371 tasks per/s
700000 tasks 2738877.029117 tasks per/s
800000 tasks 2893730.985022 tasks per/s
900000 tasks 3043877.494077 tasks per/s
1000000 tasks 2857992.089078 tasks per/s
Dead Comment
See: https://gist.github.com/jimmy-lt/4a3c6ad9cab1545692e5a3fe971...