Course contentsShow
← JavaScript
Lesson 2164 of 2,72345. Workers and ConcurrencyFree lesson

Sharing Data with SharedArrayBuffer

Use SharedArrayBuffer to share memory between the main thread and workers for low-latency communication.

Sharing Data with SharedArrayBuffer

You'll learn how to use SharedArrayBuffer for ultra-fast, shared memory communication between threads without copying data.

The Core Idea

Normally, when you send data between the main thread and a worker using postMessage, JavaScript copies the data. For large datasets or frequent updates, this copying creates overhead. SharedArrayBuffer creates a chunk of memory that both threads can access directly—no copying needed.

Think of it like a whiteboard in a shared room: instead of passing notes back and forth (copying), both people can read and write on the same board simultaneously.

How It Works

Create a SharedArrayBuffer, then access it via typed arrays in both threads:

// Main thread
const shared = new SharedArrayBuffer(16); // 16 bytes of shared memory
const view = new Int32Array(shared);
view[0] = 42;

const worker = new Worker('worker.js');
worker.postMessage({ shared });
// worker.js
self.addEventListener('message', (e) => {
  const view = new Int32Array(e.data.shared);
  console.log(view[0]); // 42 — reading shared memory
  view[0] = 100; // Both threads see this change
});

Notice the SharedArrayBuffer itself is transferred, but the memory is shared—changes in one thread are immediately visible in the other.

When to Use It

Use SharedArrayBuffer when you need low-latency, high-frequency communication: game physics engines, audio processing, or real-time data analysis. Avoid it for simple tasks where message-passing suffices—shared memory introduces race conditions you'll need to manage carefully (typically with Atomics).

Key Takeaway:
SharedArrayBuffer enables zero-copy memory sharing between threads, delivering the lowest-latency communication for performance-critical concurrent tasks.