SharedArrayBuffer for shared memory in Node.js

Introduction

SharedArrayBuffer lets different parts of a program share the same memory space. This helps them work together faster without copying data.

When you want multiple threads or workers to read and write the same data quickly.

When you need to share large data like images or audio between workers without copying.

When you want to build fast communication between parts of your program.

When you want to avoid delays caused by sending messages with copies of data.

When you want to coordinate tasks by sharing flags or counters in memory.

Syntax

Node.js

const sharedBuffer = new SharedArrayBuffer(byteLength);
const sharedArray = new Uint8Array(sharedBuffer);

byteLength is the size in bytes of the shared memory you want.

You use typed arrays like Uint8Array or Int32Array to read and write the shared memory.

Examples

Create a 16-byte shared buffer and set the first byte to 42.

Node.js

const sharedBuffer = new SharedArrayBuffer(16);
const sharedArray = new Uint8Array(sharedBuffer);
sharedArray[0] = 42;

Edge case: zero-length buffer means no shared memory.

Node.js

const sharedBuffer = new SharedArrayBuffer(0);
// This creates an empty shared buffer with no bytes.

Using Int32Array to store a 32-bit integer in shared memory.

Node.js

const sharedBuffer = new SharedArrayBuffer(4);
const sharedArray = new Int32Array(sharedBuffer);
sharedArray[0] = 123456;

Sample Program

This program shows how main thread and worker share a counter using SharedArrayBuffer. The worker increments the counter and main thread prints the updated value.

Node.js

import { Worker, isMainThread, parentPort } from 'node:worker_threads';

if (isMainThread) {
  // Main thread creates shared memory
  const sharedBuffer = new SharedArrayBuffer(4); // 4 bytes for one Int32
  const sharedArray = new Int32Array(sharedBuffer);
  sharedArray[0] = 0; // Initialize counter to 0

  // Create a worker and pass the shared buffer
  const worker = new Worker(new URL(import.meta.url));
  worker.postMessage(sharedBuffer);

  // Listen for messages from worker
  worker.on('message', () => {
    console.log('Main thread sees counter:', sharedArray[0]);
  });
} else {
  // Worker thread
  parentPort.once('message', (sharedBuffer) => {
    const sharedArray = new Int32Array(sharedBuffer);
    // Increment the shared counter
    Atomics.add(sharedArray, 0, 1);
    // Notify main thread
    parentPort.postMessage('done');
  });
}

Important Notes

Using Atomics methods like Atomics.add is important to avoid race conditions when multiple threads change shared memory.

SharedArrayBuffer size must be a nonnegative integer; zero size means no shared memory.

SharedArrayBuffer is useful when you want fast, low-level communication between threads without copying data.

Summary

SharedArrayBuffer creates memory that multiple threads can access together.

Use typed arrays to read and write shared memory.

Use Atomics to safely change shared data without conflicts.

Practice

1. What is the main purpose of SharedArrayBuffer in Node.js?

easy

A. To create a memory area that multiple threads can access simultaneously

B. To store large strings efficiently

C. To replace regular arrays with faster versions

D. To handle file system operations asynchronously

Solution

1. Step 1: Understand Shared Memory Concept

   SharedArrayBuffer is designed to create a block of memory that can be shared between multiple threads or workers.

2. Step 2: Compare with Other Options

   Options B, C, and D describe unrelated features: string storage, array speed, and file system operations, which are not the purpose of SharedArrayBuffer.

3. Final Answer:

   To create a memory area that multiple threads can access simultaneously -> Option A

4. Quick Check:

   Shared memory = multiple threads access [OK]

Hint: SharedArrayBuffer is about sharing memory across threads [OK]

Common Mistakes:

• Thinking it stores strings or files
• Confusing with normal arrays
• Assuming it handles async file tasks

2. Which of the following is the correct way to create a SharedArrayBuffer of 1024 bytes in Node.js?

easy

A. const sab = SharedArrayBuffer(1024);

B. const sab = SharedArrayBuffer.new(1024);

C. const sab = new SharedArrayBuffer();

D. const sab = new SharedArrayBuffer(1024);

Solution

1. Step 1: Check the correct constructor usage

   The SharedArrayBuffer must be created with the new keyword and a size in bytes as argument.

2. Step 2: Validate each option

   const sab = new SharedArrayBuffer(1024); uses new SharedArrayBuffer(1024), which is correct. const sab = SharedArrayBuffer(1024); misses new. const sab = new SharedArrayBuffer(); misses size argument. const sab = SharedArrayBuffer.new(1024); uses invalid syntax.

3. Final Answer:

   const sab = new SharedArrayBuffer(1024); -> Option D

4. Quick Check:

   Use new with size in bytes [OK]

Hint: Always use 'new' with SharedArrayBuffer and specify size [OK]

Common Mistakes:

• Omitting 'new' keyword
• Not providing size argument
• Using incorrect constructor syntax

3. Given the code below, what will be the output?

const sab = new SharedArrayBuffer(4);
const int32 = new Int32Array(sab);
int32[0] = 10;
Atomics.add(int32, 0, 5);
console.log(int32[0]);

medium

A. 10

B. 5

C. 15

D. NaN

Solution

1. Step 1: Understand initial value and Atomics.add

   The int32[0] is set to 10. Then Atomics.add adds 5 to this value atomically.

2. Step 2: Calculate the new value

   10 + 5 = 15, so int32[0] becomes 15.

3. Final Answer:

   15 -> Option C

4. Quick Check:

   Atomics.add adds value safely = 15 [OK]

Hint: Atomics.add adds value and returns old value, array updates [OK]

Common Mistakes:

• Expecting Atomics.add to return new value
• Ignoring atomic operation effect
• Confusing initial and updated values

4. What is wrong with the following code snippet?

const sab = new SharedArrayBuffer(8);
const uint8 = new Uint8Array(sab);
uint8[0] = 255;
Atomics.store(uint8, 0, 256);
console.log(uint8[0]);

medium

A. Atomics.store cannot be used with Uint8Array

B. 256 is out of range for Uint8Array element

C. SharedArrayBuffer size is too small

D. Uint8Array cannot be created from SharedArrayBuffer

Solution

1. Step 1: Check Uint8Array element range

   Uint8Array elements can only hold values from 0 to 255. The value 256 is out of this range.

2. Step 2: Understand effect of storing 256

   Storing 256 wraps around to 0 because 256 mod 256 = 0, so uint8[0] becomes 0, not 256.

3. Final Answer:

   256 is out of range for Uint8Array element -> Option B

4. Quick Check:

   Uint8 max value 255, 256 wraps to 0 [OK]

Hint: Uint8Array values must be 0-255; higher values wrap [OK]

Common Mistakes:

• Assuming Atomics.store rejects Uint8Array
• Ignoring value wrapping behavior
• Thinking SharedArrayBuffer size is insufficient

5. You want to safely increment a shared counter in a Node.js worker thread using SharedArrayBuffer. Which code snippet correctly increments the counter without race conditions? // Shared buffer and typed array const sab = new SharedArrayBuffer(4); const counter = new Int32Array(sab); counter[0] = 0; // Increment function function increment() { // Which line correctly increments? }

hard

A. Atomics.add(counter, 0, 1);

B. counter[0] = counter[0] + 1;

C. counter[0] += 1;

D. counter[0]++;

Solution

1. Step 1: Understand race conditions in shared memory

   Directly modifying counter[0] with normal operators can cause race conditions when multiple threads run concurrently.

2. Step 2: Use atomic operations for safe increments

   Atomics.add(counter, 0, 1) safely increments the value at index 0 without conflicts.

3. Final Answer:

   Atomics.add(counter, 0, 1); -> Option A

4. Quick Check:

   Use Atomics for safe shared memory updates [OK]

Hint: Always use Atomics methods to update shared memory safely [OK]

Common Mistakes:

• Using normal arithmetic on shared memory
• Ignoring atomicity causing race conditions
• Assuming ++ or += are thread-safe