Struct thruster::Server

pub struct Server<T: 'static + Context<Response = Response> + Clone + Send + Sync, S: 'static + Send> { /* fields omitted */ }

Implementations

impl<T: 'static + Context<Response = Response> + Clone + Send + Sync, S: 'static + Send + Sync> Server<T, S>

pub fn start_work_stealing_optimized(self, host: &str, port: u16)

Starts the app with the default tokio runtime execution model

pub fn start_small_load_optimized(self, host: &str, port: u16)

Starts the app with a thread pool optimized for small requests and quick timeouts. This is done internally by spawning a separate thread for each reactor core. This is valuable if all server endpoints are similar in their load, as work is divided evenly among threads. As seanmonstar points out though, this is a very specific use case and might not be useful for everyday work loads.alloc

See the discussion here for more information:

https://users.rust-lang.org/t/getting-tokio-to-match-actix-web-performance/18659/7

Trait Implementations

impl<T: Context<Response = Response> + Clone + Send + Sync, S: 'static + Send + Sync> ThrusterServer for Server<T, S>

type Context = T

type Response = Response

type Request = Request

type State = S

fn new(app: App<Self::Request, T, S>) -> Self

fn build<'life0, 'async_trait>(
    self,
    host: &'life0 str,
    port: u16
) -> Pin<Box<dyn Future<Output = ()> + Send + 'async_trait>> where
    'life0: 'async_trait,
    Self: 'async_trait, 

fn start(self, host: &str, port: u16) where
    Self: Sized,