VPS Load Balancing: Distribute Traffic Across Servers

Load balancing distributes traffic across multiple backend servers, preventing any single server from becoming a bottleneck. With load balancing, you can handle 10x more traffic, maintain uptime during maintenance, and scale horizontally. Essential for high-traffic applications.

Load Balancing Overview

Why load balancing matters:**

• Scalability: Handle more traffic by adding servers
• Redundancy: If one server dies, traffic routes to others
• Maintenance: Update servers without downtime
• Performance: Distribute requests evenly
• Cost: Multiple small servers cheaper than one huge server

Typical architecture:**

Internet → Load Balancer (HAProxy/Nginx) → Web Server 1
                                        → Web Server 2
                                        → Web Server 3

Load Balancing Algorithms

HAProxy Setup

Install HAProxy:**

apt update && apt install haproxy -y
systemctl enable haproxy
systemctl start haproxy

Configure /etc/haproxy/haproxy.cfg:**

global
    log stdout local0
    maxconn 4096

frontend web_in
    bind *:80
    mode http
    default_backend web_servers

backend web_servers
    mode http
    balance roundrobin

    server web1 10.0.0.1:80 check
    server web2 10.0.0.2:80 check
    server web3 10.0.0.3:80 check

    # Health check (every 2 seconds, fail after 3 failures)
    default-server inter 2000 fall 3 rise 2

Restart HAProxy:**

systemctl restart haproxy
systemctl status haproxy

Nginx Load Balancer

Nginx for load balancing (upstream):**

upstream backend {
    least_conn;  # Least connections algorithm

    server 10.0.0.1:8000 weight=1;
    server 10.0.0.2:8000 weight=1;
    server 10.0.0.3:8000 weight=2;  # Handle 2x more traffic

    # Health check
    server 10.0.0.4:8000 backup;  # Backup if others fail
}

server {
    listen 80;
    server_name yourdomain.com;

    location / {
        proxy_pass http://backend;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
    }
}

Health Checks

HAProxy health checks:**

# HTTP health check (GET /health endpoint)
server web1 10.0.0.1:80 check http \
  option httpchk GET /health \
  http-check expect status 200

Implement health check endpoint in your app:**

# Express.js example
app.get('/health', (req, res) => {
    res.json({ status: 'ok', uptime: process.uptime() });
});

# Flask example
@app.route('/health')
def health():
    return {'status': 'ok'}

Session Persistence

Sticky sessions (route same user to same server):**

# HAProxy: use JSESSIONID cookie
backend web_servers
    cookie JSESSIONID prefix nocache
    server web1 10.0.0.1:80 check cookie web1
    server web2 10.0.0.2:80 check cookie web2

# Or use source IP-based persistence
backend web_servers
    balance source  # IP hash algorithm

Horizontal Scaling

Add new server to load balancer:**

# In HAProxy config, add line:
server web4 10.0.0.4:80 check

# Reload without restarting
haproxy -f /etc/haproxy/haproxy.cfg -sf $(pgrep haproxy)

Remove server for maintenance:**

# Mark as backup (stops receiving traffic)
server web1 10.0.0.1:80 check backup

# Or set to drain (complete active connections, stop new ones)
set server backend/web1 state drain

Monitor Load Balancer

HAProxy stats page:**

# In haproxy.cfg, enable stats
listen stats
    bind *:8404
    stats enable
    stats uri /stats
    stats refresh 5s

# Access at: http://server:8404/stats

Monitor backend server health:**

tail -f /var/log/haproxy.log | grep "backend"
# Shows server up/down status

Related: VPS clustering | High availability | Scaling strategy | Nginx reverse proxy