|
@asyncrs | |||||
|
Don't let anyone block you! We're introducing a new adaptive, @golang-inspired, @rustlang async runtime that automatically detects blocking behaviour and just continues to run and run. Don't worry about separate blocking and non-blocking tasks anymore!
async.rs/blog/stop-worr… pic.twitter.com/00gd1qaVtr
|
||||||
|
||||||
|
async-std
@asyncrs
|
16. pro |
|
The new runtime also allows you to use mature sync libraries like @dieselframework without fear! No need to manage a secondary threadpool for blocking tasks!
On top of that, the new runtime is also much better at scaling down, effectively becoming single-threaded if possible.
|
||
|
|
||
|
async-std
@asyncrs
|
16. pro |
|
So, how about speed? The new runtime is faster over 1.3.0, especially scaling much better.
How do you use it? Just try out our current PR branch, instructions are in the blog post.
async-std is fully compatible to futures-rs, for your ease! pic.twitter.com/6mgInNuK10
|
||
|
|
||
|
Leo Dutra
@leodutra
|
16. pro |
|
Wow, this is awesome. Thanks.
|
||
|
|
||
|
async-std
@asyncrs
|
16. pro |
|
🙇♂️ It's a pleasure!
|
||
|
|
||
|
Adil H
@le_didil
|
16. pro |
|
"The runtime measures the time it takes to perform the blocking operation and if it takes a while, a new thread is automatically spawned" ... would love a more detailed blog post explaining how this actually works
|
||
|
|
||
|
async-std
@asyncrs
|
16. pro |
|
Coming up down the line.
But essentially, its the `sysmon` from this: utcc.utoronto.ca/~cks/space/blo…
|
||
|
|
||
|
Rafał Cieślak
@Ravicious
|
16. pro |
|
Does it actually make the problem described in “What Color Is Your Function” disappear? 🤔 I pretty much don’t know Rust, but I see that you still have to use an async function.
|
||
|
|
||
|
async-std
@asyncrs
|
16. pro |
|
It makes the problem of calling sync functions within async functions. async function in Rust still have specific meaning (they are used to generate Futures, which you can execute). But that problem is rather inherent, but also not so painful.
|
||
|
|
||
|
Andy 🦀⚙️
@atolvstad
|
16. pro |
|
Not gonna lie, this reads kinda like an April Fools Day post
|
||
|
|
||
|
Markus A.
@brainafk
|
16. pro |
|
Interesting! Not just for blocking tasks, but also for potentially long running ones where it is hard to decide whether they are better offloaded to a thread pool or not 🧐. Seems like I finally have a reason to give @asyncrs a try (instead of using Tokio as I am used to).
|
||
|
|
||