class Fiber

Overview

A Fiber is a light-weight execution unit managed by the Crystal runtime.

It is conceptually similar to an operating system thread but with less overhead and completely internal to the Crystal process. The runtime includes a scheduler which schedules execution of fibers.

A Fiber has a stack size of 8 MiB which is usually also assigned to an operating system thread. But only 4KiB are actually allocated at first so the memory footprint is very small.

Communication between fibers is usually passed through Channel.

Cooperative

Fibers are cooperative. That means execution can only be drawn from a fiber when it offers it. It can't be interrupted in its execution at random. In order to make concurrency work, fibers must make sure to occasionally provide hooks for the scheduler to swap in other fibers. IO operations like reading from a file descriptor are natural implementations for this and the developer does not need to take further action on that. When IO access can't be served immediately by a buffer, the fiber will automatically wait and yield execution. When IO is ready it's going to be resumed through the event loop.

When a computation-intensive task has none or only rare IO operations, a fiber should explicitly offer to yield execution from time to time using Fiber.yield to break up tight loops. The frequency of this call depends on the application and concurrency model.

Event loop

The event loop is responsible for keeping track of sleeping fibers waiting for notifications that IO is ready or a timeout reached. When a fiber can be woken, the event loop enqueues it in the scheduler

Defined in:

stdlib/fiber.cr

Class Method Summary

Class Method Detail

def self.count #

[View source]
def self.list(&blk : Fiber -> Nil) #

[View source]