rayon-rs/rayon

Features

• Parallel Data Iterators - Converts sequential data structures into parallel iterators to execute operations across multiple threads.
• Parallel Iterators - Transforms sequential data structures into parallel iterators to partition work across multiple processor cores.
• Parallel Processing - Provides parallel iterators and work-stealing algorithms to distribute data processing workloads across multiple CPU cores.
• Data Parallelism Frameworks - Provides a framework for transforming sequential data structures into parallel iterators for concurrent processing.
• Thread Pools - Provides a global thread pool for managing concurrent tasks and minimizing thread creation overhead.
• Parallel Loop Transformers - Transforms standard sequential loops into parallel versions to process collection elements concurrently.
• Scoped Synchronization - Ensures memory safety by guaranteeing all spawned parallel operations complete before the parent scope exits.
• Work-Stealing Schedulers - Implements work-stealing scheduling to dynamically balance computational load across available processor cores.
• Scoped Task Synchronization - Ensures memory safety by guaranteeing that all spawned parallel tasks complete before the parent scope finishes execution.
• Parallel Processing Utilities - Offers a comprehensive library for converting sequential Rust iterators into parallel versions for multi-core execution.
• Execution Scopes - Establishes controlled execution environments where all spawned operations must finish before the scope exits.
• Task Queues - Uses lock-free atomic structures to exchange work items between threads for high-throughput task distribution.
• Task Schedulers - Implements a work-stealing scheduler that dynamically balances tasks across a thread pool to maximize throughput.
• Concurrent Task Runners - Provides primitives for orchestrating concurrent tasks within a managed scope to improve application performance.
• Parallel Task Spawning - Spawns multiple operations within a shared scope to perform concurrent calculations and merge results.
• Recursive Decomposers - Orchestrates complex, recursive task decomposition to process sub-tasks in parallel until reaching a sequential base case.
• Custom Thread Pool Configurations - Supports the creation of isolated, custom thread pools to prevent resource contention during heavy workloads.
• Parallel Execution - Executes multiple operations concurrently within a shared scope to improve overall processing speed.
• Custom Parallel Task Execution - Divides complex workloads into smaller independent units of work that run concurrently within a managed scope.
• Broadcast Operations - Enables the distribution of operations to every thread within a pool for uniform updates or data collection.
• Asynchronous Task Queues - Submits independent operations to a global thread pool for asynchronous background execution.

Star history