nodeswarm
v0.0.1
Published
Lightweight library for managing worker threads and parallel execution of tasks in Node.js.
Downloads
2
Maintainers
Readme
:warning: Warning: This is an experimental package and may undergo significant changes. Please use with caution and feel free to report any issues or suggestions.
NodeSwarm
Efficiently manage and utilize worker threads in Node.js.
NodeSwarm
is a lightweight, easy-to-use library designed to simplify the process of creating, managing, and utilizing
worker threads in Node.js, enabling developers to easily leverage multicore processing power for CPU-bound tasks.
Motivation and Features
NodeSwarm
was developed to simplify multithreading in Node.js, reducing the hassle and distraction from extensive boilerplate required by worker_threads
. The library enables direct execution of any function using threads, removing the need for separate files, especially beneficial for TypeScript users, all within a clean and intuitive API.
Features
- Efficiency: Offers optimal performance and scalability through simplified multithreading and efficient management of CPU-bound tasks.
- Virtual Threads & Queue: Ensures orderly and resource-efficient task execution.
- Simple API: Guarantees easy integration and straightforward task execution.
- Error Handling: Facilitates smooth debugging with concise error feedback.
- Flexibility: Provides adjustable configurations to suit various needs.
Installation
Install NodeSwarm using npm:
npm install nodeswarm
Or with yarn:
yarn add nodeswarm
Quick Start
Here's a simple example of how to use NodeSwarm to run a task in a separate thread:
const { ThreadPool } = require('nodeswarm');
const pool = new ThreadPool();
async function main() {
try {
const result = await pool.thread((a, b) => a + b, 2, 3);
console.log(result); // Outputs: 5
} catch (error) {
console.error(error);
} finally {
await pool.close();
}
}
main();
Fibonacci Example
Let's use the famous Fibonacci function to compare running tasks with normal Promises and with the NodeSwarm ThreadPool. Fibonacci is a CPU-bound task which makes it a good candidate to showcase the benefits of multithreading.
// Import your thread pool
import { ThreadPool } from 'nodeswarm'; // Adjust the import path as necessary
const pool = new ThreadPool();
// Define a fib function
function fib(n) {
if (n < 2) return n;
return fib(n - 1) + fib(n - 2);
}
// Run fib normally
console.time('fib block');
const promises1 = Array(10).fill(0).map((_, i) => fib(40));
await Promise.all(promises1);
console.timeEnd('fib block');
// Run fib with threads
console.time('fib thread');
const promises2 = Array(10).fill(0).map((_, i) => pool.thread(fib, 40));
await Promise.all(promises2);
console.timeEnd('fib thread');
Limitations
NodeSwarm
offers a streamlined approach to multithreading in Node.js but has its constraints:
- Independent Execution: Threads run separately, and despite parallel execution, each context remains single-threaded.
- Data Sharing: Data between threads is copied, not shared.
SharedArrayBuffer
support is planned for the future. - Optimized for CPU-bound Tasks: For I/O-bound tasks, Node.js’s asynchronous I/O is typically more efficient.
- Overhead: Multithreading introduces added complexity and might be excessive for simple applications.
- Node.js Dependency: Requires Node.js supporting
worker_threads
(experimental pre-11.7.0, stable from v12). - Incompatibility with Classes: The library may not handle classes as expected.
- Experimental Status: Potential for bugs or API changes. Use cautiously in production.
Ensure NodeSwarm
aligns with your requirements before integrating it into projects.
Security Consideration: Use of Eval
NodeSwarm
utilizes eval()
internally to dynamically execute functions in worker threads. While this offers flexibility and ease of use, it also necessitates caution:
- Do not use
NodeSwarm
to run untrusted or user-provided code. Doing so can expose your application to significant security risks, including arbitrary code execution. - Use exclusively with your own, well-reviewed code. Ensure that the code being run with
NodeSwarm
is secure and has been thoroughly reviewed to avoid potential security vulnerabilities.
By adhering strictly to these guidelines, you can mitigate risks associated with the use of eval()
and safely enjoy the benefits of NodeSwarm
.
API
ThreadPool
new ThreadPool([config])
Create a new thread pool. Optionally, you can provide a config object to specify the pool size.
thread(fn: Function, ...args: any[]): Promise<any>
Executes the provided function fn in a separate thread with the provided arguments args and returns a promise that resolves with the result.
close(): Promise<void>
Gracefully closes all worker threads after completing all ongoing jobs.
terminate(): void
Immediately terminates all worker threads.
License
NodeSwarm is MIT licensed.