Module scheduling 

Work-stealing schedulers for adaptive algorithms

This module provides work-stealing task schedulers optimized for adaptive numerical algorithms. These schedulers dynamically balance workload across threads, which is particularly important for algorithms with irregular computational patterns.

Work-Stealing Concepts

Work-stealing is a scheduling technique where idle threads “steal” work from busy threads’ task queues. This is especially effective for:

• Adaptive algorithms with unpredictable work distribution
• Recursive divide-and-conquer algorithms
• Dynamic load balancing scenarios

Examples

use scirs2_integrate::scheduling::{WorkStealingPool, Task};

// Create work-stealing pool with 4 threads
let pool = WorkStealingPool::new(4);

// Submit a simple task
let task = Task::new(|| 0.5 * 0.5); // Simple computation
pool.submit(task);

Structs

AdaptiveIntegrationTask

Adaptive integration task for work-stealing scheduler

PoolStatistics

Statistics about pool performance

Task

Simple boxed task for closures

WorkStealingPool

Work-stealing thread pool for adaptive algorithms

Traits

WorkStealingTask

Generic task that can be executed by the work-stealing scheduler