[][src]Crate tange

tange-core

tange-core provides primitives for building and running task-based computations.

What is it?

Tange is a framework that makes it easy to write defered, data parallel computations that are executed concurrently across a local machine. It can scale up to millions of tasks per Graph and can be useful for a number of different applications:

  • Data processing.
  • All-Reduce operations.
  • Distributed machine learning algorithms.
  • General parallel computing.

How to Use It?

Tange defines a Deferred struct which represents a computation. Deferred objects are accessed with three simple functions:

  1. lift - Lift takes a concrete value and lifts it into a Deferred object
  2. apply - Apply applies a function to a Deferred, producing a new Deferred object.
  3. join - Join combines two Deferred objects with a joiner function, producing a new Deferred.

Example - Hello World!

use tange::deferred::Deferred;
use tange::scheduler::GreedyScheduler;
 
let hello = Deferred::lift("Hello".to_owned(), None);
let world = Deferred::lift("World".to_owned(), None);
let world_exclaim = world.apply(|w| format!("{}!", w));
let hello_world = hello.join(&world_exclaim, |h, w| format!("{} {}", h, w));
assert_eq!(hello_world.run(&GreedyScheduler::new()), Some("Hello World!".into()));

Modules

deferred

Contains Deferred primitive and function definitions Defines the Deferred primitive

scheduler

Contains Scheduler trait definition and implementations Contains all the runtimes scheduling Graphs for execution.