Load Balancing — Animate | AI Wisdom

← Back to topic
🎞 ANIMATE · Mode 3 of 5 · +100 XP— Round-Robin Load Balancer
📋Health Check
⚙️Route Request
🔄Round-Robin
✅Failover
🎯 Next: Test →
Step 1 / 4
Speed
Health check servers
Every 5s: ping /health on each server. Mark up or down.💡 Analogy
Like a manager checking which employees are at their desks.
Regularly check each server's health endpoint. Mark healthy or unhealthy based on response.
Console Output
> health check:
> s1:3000/health → 200 ✅
> s2:3000/health → 200 ✅
> s3:3000/health → timeout ❌ → mark unhealthy
●load-balancer.ts
Health check serversL21–30
1class LoadBalancer {
2  private servers: string[];
3  private currentIndex = 0;
4  private health: Map<string, boolean>;
5 
6  constructor(servers: string[]) {
7    this.servers = servers;
8    this.health = new Map(servers.map(s => [s, true]));
9    this.startHealthChecks();
10  }
11 
12  getNextServer(): string {
13    const healthy = this.servers.filter(s => this.health.get(s));
14    if (healthy.length === 0) throw new Error('No healthy servers');
15 
16    const server = healthy[this.currentIndex % healthy.length];
17    this.currentIndex++;
18    return server;
19  }
20 
21  private async startHealthChecks() {
22    setInterval(async () => {
23      for (const server of this.servers) {
24        try {
25          await fetch(`${server}/health`);
26          this.health.set(server, true);
27        } catch {
28          this.health.set(server, false);   // Mark unhealthy
29        }
30      }
31    }, 5000); // Every 5 seconds
32  }
33}
34 
35const lb = new LoadBalancer(['http://s1:3000','http://s2:3000','http://s3:3000']);
36lb.getNextServer(); // s1, s2, s3, s1, s2...
36 lines
Timeline — 4 Steps
Health check servers
Every 5s: ping /health on each server. Mark up or down.💡 Analogy
Like a manager checking which employees are at their desks.
Regularly check each server's health endpoint. Mark healthy or unhealthy based on response.
Console Output
> health check:
> s1:3000/health → 200 ✅
> s2:3000/health → 200 ✅
> s3:3000/health → timeout ❌ → mark unhealthy

Was this page helpful?

Discussion

Every 5s: ping /health on each server. Mark up or down.

💡 Analogy

Like a manager checking which employees are at their desks.

Regularly check each server's health endpoint. Mark healthy or unhealthy based on response.

Console Output

> health check:
> s1:3000/health → 200 ✅
> s2:3000/health → 200 ✅
> s3:3000/health → timeout ❌ → mark unhealthy

Every 5s: ping /health on each server. Mark up or down.

💡 Analogy

Like a manager checking which employees are at their desks.

Regularly check each server's health endpoint. Mark healthy or unhealthy based on response.

Console Output

> health check:
> s1:3000/health → 200 ✅
> s2:3000/health → 200 ✅
> s3:3000/health → timeout ❌ → mark unhealthy