Understanding Load Balancing in Modern Infrastructure

load-balancing devops

Load balancing is a fundamental technique for achieving scalability and reliability in web applications. A load balancer acts as a traffic manager sitting in front of your servers.

Why Load Balancing?

Without a load balancer, a single server handles all client requests. This quickly becomes a bottleneck. By distributing requests across multiple backend servers, you can handle more traffic and maintain availability even if a server fails.

Common Strategies

Round-Robin

Round-robin sends each incoming request to the next server in line. It’s simple and works well when servers are equally capable.

upstream backend {
    server server1.example.com;
    server server2.example.com;
    server server3.example.com;
}

server {
    listen 80;
    location / {
        proxy_pass http://backend;
    }
}

Least Connections

This algorithm forwards requests to the server with the fewest active connections. It’s helpful when your servers handle workloads of varying duration.

IP Hash

With IP hash, each client’s IP address maps to a specific server. This ensures a client consistently reaches the same backend, which can be useful for session persistence.

Application-Layer Load Balancing

At Layer 7 of the OSI model, load balancers make routing decisions based on HTTP headers, cookies, or application data. This allows advanced traffic management such as routing by URL path.

Diagram

Below is a simplified view of traffic distribution to three backend servers:

Load balancers can be hardware appliances or software-based solutions. Cloud providers also offer managed options. Choosing the right strategy depends on your application’s requirements, the type of traffic, and scaling goals.