Course contentsShow
← System Design
Lesson 83 of 1,9193. Load BalancingFree lesson

Client-Side Load Balancing

Understand when clients can select backends directly without a centralized load balancer.

Client-Side Load Balancing

What you'll learn: How clients can distribute requests across backends without relying on a centralized load balancer.

The Core Idea

In client-side load balancing, the client application itself maintains a list of available backend servers and chooses which one to send each request to. There's no separate load balancer appliance or service in the middle—the load balancing logic runs directly in the client code.

Think of it like choosing which cashier line to join at a grocery store: you look at all the open registers, pick the one that seems best (maybe the shortest line), and walk directly to it. No store employee directs you—you make the choice yourself.

How It Works

The client needs three things:

  1. Service discovery: A way to learn which backend servers are available (often from a registry like we'll cover later, but for now imagine a configuration file or API that lists server addresses)
  2. Selection logic: An algorithm to pick one server from the list (round-robin, random, least-connections, etc.)
  3. Health tracking: Knowledge of which servers are healthy vs unavailable

Each time the client needs to make a request, it runs this logic and connects directly to the chosen backend.

When This Makes Sense

Client-side load balancing shines when:

  • You control both client and server code (like microservices talking to each other)
  • You want to eliminate the load balancer as a single point of failure
  • Network hops matter for latency (going direct is faster)
  • Clients are sophisticated enough to handle the complexity

Key Takeaway: Client-side load balancing moves the distribution logic into the client itself, enabling direct connections to backends without a centralized intermediary—useful when clients can be trusted with this responsibility.