Long polling as fallback in Node.js - Performance & Optimization

Performance: Long polling as fallback

MEDIUM IMPACT

This affects the responsiveness and load on the server and client during real-time data updates when WebSocket is unavailable.

Implementing real-time updates when WebSocket is not supported

Node.js

app.get('/updates', (req, res) => {
  const checkForUpdates = () => {
    if (dataAvailable()) {
      res.json({ data: getData() });
    } else {
      setTimeout(checkForUpdates, 1000); // retry after 1 second
    }
  };
  checkForUpdates();
});

This pattern checks for data availability before responding, reducing unnecessary open connections and improving responsiveness.

📈 Performance GainReduces server load by closing connections promptly when no data is available

Implementing real-time updates when WebSocket is not supported

Node.js

app.get('/updates', (req, res) => {
  setTimeout(() => {
    res.json({ data: 'new data' });
  }, 10000); // waits 10 seconds before responding
});

This naive long polling holds connections open for a fixed time regardless of data availability, causing unnecessary server resource use and delayed updates.

📉 Performance CostTriggers many open connections and high server memory usage under load

Performance Comparison

Pattern	DOM Operations	Reflows	Paint Cost	Verdict
Naive long polling with fixed delay	Minimal DOM changes	0 reflows triggered by polling itself	Delayed paint due to waiting on response	[X] Bad
Conditional long polling with data check	Minimal DOM changes	0 reflows triggered by polling itself	Faster paint when data is ready	[OK] Good

Rendering Pipeline

Long polling affects the interaction responsiveness by keeping HTTP connections open, which delays the browser's next paint until data arrives and the response completes.

→Network

→JavaScript Execution

→Paint

⚠️ BottleneckNetwork latency and server response time are the main bottlenecks.

Core Web Vital Affected

INP

This affects the responsiveness and load on the server and client during real-time data updates when WebSocket is unavailable.

Optimization Tips

1Avoid holding HTTP connections open longer than necessary.

2Check for data availability before responding to reduce latency.

3Use long polling only as a fallback when WebSocket is unavailable.

Performance Quiz - 3 Questions

What is the main performance drawback of naive long polling?

A It holds HTTP connections open unnecessarily long, increasing server load.

B It causes excessive DOM reflows on the client.

C It blocks the browser's main thread permanently.

D It increases the size of JavaScript bundles.

DevTools: Network

How to check: Open DevTools, go to Network tab, filter XHR requests, observe the duration of long polling requests and their frequency.

What to look for: Look for long-held HTTP connections and delayed responses indicating inefficient long polling.

Practice

1. What is the main purpose of using long polling in Node.js applications?

easy

A. To keep a client request open until new data is available

B. To immediately close the connection after sending data

C. To send multiple responses to a single request

D. To replace WebSocket connections entirely

Solution

1. Step 1: Understand long polling behavior

   Long polling keeps the client request open until the server has new data to send.

2. Step 2: Compare options to this behavior

   Only To keep a client request open until new data is available describes this behavior correctly; others describe different or incorrect behaviors.

3. Final Answer:

   To keep a client request open until new data is available -> Option A

4. Quick Check:

   Long polling = open request until data ready [OK]

Hint: Long polling holds request open until data arrives [OK]

Common Mistakes:

• Thinking long polling sends multiple responses per request
• Confusing long polling with WebSocket persistent connection
• Assuming long polling closes connection immediately

2. Which of the following is the correct way to implement a long polling endpoint in Node.js using Express?

easy

A. res.write('data'); // keep connection open without res.end()

B. res.send('data'); res.end();

C. res.json('data'); res.end();

D. res.send('data'); // then wait before closing connection

Solution

1. Step 1: Identify how to keep connection open

   In long polling, the server holds the connection open, so it should not call res.end() immediately.

2. Step 2: Check which option keeps connection open

   res.write('data'); // keep connection open without res.end() uses res.write() without res.end(), which keeps the connection open for long polling.

3. Final Answer:

   res.write('data'); // keep connection open without res.end() -> Option A

4. Quick Check:

   Keep connection open = no res.end() [OK]

Hint: Use res.write() without res.end() to keep connection open [OK]

Common Mistakes:

• Calling res.end() immediately closes connection
• Using res.send() closes connection right away
• Not sending any response causes timeout

3. Consider this Node.js long polling code snippet:

app.get('/poll', (req, res) => {
  setTimeout(() => {
    res.json({ message: 'Update available' });
  }, 5000);
});

What will the client experience when calling this endpoint?

medium

A. The client never receives a response

B. The client receives the update immediately

C. The client waits 5 seconds, then receives the update

D. The client receives multiple updates every 5 seconds

Solution

1. Step 1: Analyze the setTimeout delay

   The server waits 5 seconds before sending the JSON response.

2. Step 2: Understand client wait behavior

   The client request stays open during the 5 seconds, then receives the update once sent.

3. Final Answer:

   The client waits 5 seconds, then receives the update -> Option C

4. Quick Check:

   setTimeout delay = delayed response [OK]

Hint: setTimeout delays response, client waits that time [OK]

Common Mistakes:

• Assuming immediate response despite delay
• Expecting multiple responses from one request
• Thinking client never gets response

4. Given this long polling code snippet, what is the main issue?

app.get('/poll', (req, res) => {
  res.json({ message: 'Update' });
  setTimeout(() => {
    res.json({ message: 'Another update' });
  }, 3000);
});

medium

A. The first res.json() never sends data

B. Calling res.json() twice causes an error

C. setTimeout is not asynchronous

D. The server never responds to the client

Solution

1. Step 1: Understand res.json() behavior

   res.json() sends a response and closes the connection immediately.

2. Step 2: Analyze calling res.json() twice

   Calling res.json() a second time after the connection is closed causes an error.

3. Final Answer:

   Calling res.json() twice causes an error -> Option B

4. Quick Check:

   Multiple res.json() calls = error [OK]

Hint: Only one res.json() per request allowed [OK]

Common Mistakes:

• Thinking setTimeout blocks code
• Assuming multiple res.json() calls send multiple responses
• Ignoring connection close after first response

5. You want to implement long polling as a fallback in a Node.js app that first tries WebSocket. Which approach correctly handles fallback and client reconnection?

hard

A. Use WebSocket and long polling simultaneously, sending duplicate data to client

B. Use WebSocket only; if unavailable, do nothing and drop updates

C. Use long polling only; never try WebSocket to simplify code

D. Use WebSocket; if unavailable, client repeatedly sends GET requests that wait on server until data or timeout, then reconnect

Solution

1. Step 1: Understand fallback strategy

   WebSocket is preferred for real-time, but fallback uses long polling with repeated client requests.

2. Step 2: Identify correct client reconnection method

   The client must send repeated GET requests that wait on server until data or timeout, then reconnect to keep updates flowing.

3. Final Answer:

   Use WebSocket; if unavailable, client repeatedly sends GET requests that wait on server until data or timeout, then reconnect -> Option D

4. Quick Check:

   Fallback = repeated long polling requests [OK]

Hint: Fallback means repeated long polling requests after WebSocket fails [OK]

Common Mistakes:

• Ignoring fallback and dropping updates
• Sending duplicate data by using both methods simultaneously
• Not reconnecting client after long polling timeout