WebSockets & Realtime Systems Programming · 课时

双向流式传输与流量控制

了解如何管理双向的连续数据流，并实现基本的流量控制。

双向流式传输与流量控制 是 CoddyKit 上的免费 WebSockets & Realtime Systems Programming 课时。这是第 3 节课，共 4 节。你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。这是 WebSockets & Realtime Systems Programming 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 WebSockets & Realtime Systems Programming 课程共包含 4 节课。

本课时的部分内容尚未翻译，以英文显示。

Beyond Simple Messages

So far, we've mostly thought about WebSockets as a way to send discrete messages back and forth. But what if you need to transfer a continuous stream of data?

This is where bidirectional streaming comes in. It's about maintaining a steady, ongoing flow of data simultaneously in both directions, not just isolated messages.

Why Bidirectional Streaming Matters

Imagine scenarios like:

Live Audio/Video: Sending and receiving real-time media streams.
Large File Transfers: Uploading or downloading big files in chunks.
Real-time Analytics: Continuous updates for dashboards with high data volume.

For these, a constant 'river' of data is more efficient than many small, separate 'droplets'.

The Challenge of Data Flow

What happens if one side sends data much faster than the other can process it? Think of a firehose pouring water into a small cup.

The receiver's temporary storage (called a buffer) will quickly fill up. This can lead to:

Data loss
System slowdowns
Memory exhaustion

This is a critical problem for continuous data flow.

Introducing Flow Control

To prevent overwhelming a receiver, we use Flow Control. It's a mechanism that manages the rate of data transmission between a sender and a receiver.

Its main goal is to ensure the sender doesn't send data faster than the receiver can handle, making the communication smooth and reliable.

Key Flow Control Concepts

Flow control relies on a few core ideas:

Buffering: Temporary storage for data that's been sent but not yet processed by the receiver.
Backpressure: A signal from the receiver to the sender, indicating it needs to slow down or pause.
Pause/Resume: Explicit commands or implicit behaviors to halt and restart data flow.

Think of it like a traffic light for your data stream.

WebSocket Buffering in Node.js

When you use ws.send(data) in Node.js, the ws library manages an internal buffer for outgoing messages.

The ws.send() method returns a boolean:

true: Data was sent immediately or buffered successfully.
false: The internal buffer is full. You should pause sending new data.

You can also check ws.bufferedAmount, which tells you how many bytes are currently in the outgoing buffer.

const WebSocket = require('ws');
const wss = new WebSocket.Server({ port: 8080 });

wss.on('connection', ws => {
  console.log('Client connected');

  // Check buffer size before sending
  const currentBuffer = ws.bufferedAmount;
  console.log(`Current buffer size: ${currentBuffer} bytes`);

  const ok = ws.send('Hello, streaming!');
  if (!ok) {
    console.log('Buffer full right away!');
  } else {
    console.log(`Message sent. New buffer size: ${ws.bufferedAmount} bytes`);
  }

  ws.on('message', message => {
    console.log(`Received: ${message}`);
  });

  ws.on('close', () => console.log('Client disconnected'));
});

console.log('Server running on port 8080');

Server-Side Backpressure (Part 1)

To implement flow control, your server needs to react when its outgoing buffer is full. If ws.send() returns false, you must stop sending data until the buffer clears.

This prevents the server from consuming too much memory or overloading the client's connection.

Here's the core idea of pausing:

const WebSocket = require('ws');
const wss = new WebSocket.Server({ port: 8080 });

wss.on('connection', ws => {
  let messageCounter = 0;
  let isPaused = false;

  function sendLotsOfData() {
    if (isPaused) return; // Don't send if paused

    while (messageCounter < 1000) { // Send 1000 messages
      const data = `Data chunk ${messageCounter++}`;
      const ok = ws.send(data);

      if (!ok) {
        console.log('Buffer full! Pausing send...');
        isPaused = true; // Set flag to pause
        break; // Stop sending for now
      }
    }
    if (messageCounter >= 1000) {
      console.log('All data chunks sent!');
    }
  }

  sendLotsOfData(); // Start sending

  ws.on('message', msg => {}); // Placeholder
  ws.on('close', () => {}); // Placeholder
});

Server-Side Backpressure (Part 2)

When the buffer clears enough for more data, the ws library emits a 'drain' event. This is your cue to resume sending!

Combining ws.send()'s return value with the 'drain' event creates robust server-side flow control.

Try running this complete example. You'll see 'Buffer full!' and 'Buffer drained!' messages as flow control kicks in.

const WebSocket = require('ws');
const wss = new WebSocket.Server({ port: 8080 });

wss.on('connection', ws => {
  console.log('Client connected. Starting data stream...');

  let messageCounter = 0;
  let isSending = false; // Flag to manage sending state

  function streamData() {
    if (isSending) return; // Already sending, wait for drain or completion

    isSending = true;
    while (messageCounter < 500) { // Simulate sending 500 messages
      const data = `Stream chunk ${messageCounter++} from server.`;
      const ok = ws.send(data);

      if (!ok) {
        console.log(`Buffer full (${ws.bufferedAmount} bytes), pausing send...`);
        isSending = false; // Stop sending until drain
        break; // Exit loop, wait for drain
      }
    }
    if (messageCounter >= 500) {
      console.log('Finished sending all stream chunks.');
      isSending = false;
    }
  }

  streamData(); // Start sending data after connection

  ws.on('drain', () => {
    console.log('Buffer drained, resuming send.');
    streamData(); // Resume sending
  });

  ws.on('message', message => {
    console.log(`Received from client: ${message}`);
    // In a real app, client might also stream data here,
    // requiring similar flow control logic on the client side.
  });

  ws.on('close', () => console.log('Client disconnected'));
  ws.on('error', error => console.error('WebSocket error:', error));
});

console.log('WebSocket server started on port 8080. Connect a client to see streaming.');

Client-Side Flow Control

While the server-side example focuses on outgoing data, clients also need to manage incoming streams and potentially their own outgoing streams.

On the client (browser JavaScript):

For receiving data, you might buffer incoming messages if processing is slow.
For sending large data (e.g., file uploads), you'd send in chunks and might need server-sent acknowledgments or explicit 'pause' signals to implement client-side backpressure.

The core principles remain the same: don't send faster than the receiver can handle.

Quick Check: Flow Control Logic

Consider a Node.js WebSocket server trying to send a large amount of data to a client. Which of the following statements about implementing server-side flow control are TRUE?

Recap: Mastering Realtime Streams

In this lesson, we explored bidirectional streaming and the crucial concept of flow control in WebSockets.

Bidirectional streaming allows for continuous, simultaneous data flow, ideal for applications like live video or large data transfers.
Flow control, using mechanisms like buffering, backpressure, and the 'drain' event, prevents overwhelming either the sender or receiver.
Implementing proper flow control is essential for building robust, high-performance, and reliable real-time applications.

Keep these principles in mind as you build your next streaming WebSocket application!

常见问题解答

「双向流式传输与流量控制」课时是免费的吗？

是的 — 「双向流式传输与流量控制」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 WebSockets & Realtime Systems Programming 课程的其余内容，请升级到 CoddyKit PRO。 WebSockets & Realtime Systems Programming 课程共包含 4 节课。

「双向流式传输与流量控制」这节课中我会学到什么？

了解如何管理双向的连续数据流，并实现基本的流量控制。你通过在浏览器中直接运行的动手代码来练习 WebSockets & Realtime Systems Programming，全天候 AI 导师会在你学习这节课的过程中回答你的问题。

学习 WebSockets & Realtime Systems Programming 需要有经验吗？

无需任何先前经验。CoddyKit 上的 WebSockets & Realtime Systems Programming 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。这是第 3 节课，共 4 节。

「双向流式传输与流量控制」课时需要多长时间？

大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。

我能在这节 WebSockets & Realtime Systems Programming 课中编写并运行代码吗？

能。每节 WebSockets & Realtime Systems Programming 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。

此课程中的所有课时

← 返回 WebSockets & Realtime Systems Programming