parallelism

The parallelism library empowers developers with precise control over parallel and concurrent task execution through its core functions: scheduled_task and task_scheduler. The scheduled_task function allows users to define, prioritize, and manage individual tasks, seamlessly harnessing the benefits of parallelism and concurrency. On the other hand, the task_scheduler function offers a high-level approach to orchestrating multiple tasks, efficiently distributing them across processes and threads while optimizing resource utilization. With these functions, the parallelism library simplifies the creation of efficient and responsive Python applications, facilitating both parallel and concurrent execution paradigms.

Information

Installation

pip install parallelism

API Reference