Load Balancing

Overview

Load balancing is the process of distributing network traffic across multiple servers. This ensures no single server bears too much demand, which improves responsiveness and increases availability of applications. A load balancer sits between the client and the server farm and accepts incoming network and application traffic.

Key Concepts

1. Algorithms

  • Round Robin: Requests are distributed sequentially.
  • Least Connections: Sends requests to the server with the fewest active connections.
  • IP Hash: The IP address of the client is used to determine which server receives the request.
  • Weighted Round Robin: Servers with higher capacity receive more requests.

2. Layers

  • L4 (Transport Layer): Based on IP and port. Fast but doesn't look at the content of the packet.
  • L7 (Application Layer): Based on HTTP headers, cookies, or URL. Allows for "intelligent" routing.

3. Health Checks

Load balancers regularly check the health of backend servers. If a server is down, it is removed from the rotation until it recovers.

Trade-offs & Considerations

  • Single Point of Failure: If the load balancer goes down, the whole system is unreachable. This is solved by using redundant load balancers (Active-Passive or Active-Active).
  • Session Persistence (Sticky Sessions): Some applications require a user to stay on the same server. This can be handled by the load balancer but can lead to uneven distribution.
  • Hardware vs. Software: Hardware load balancers (F5) are powerful but expensive. Software load balancers (Nginx, HAProxy, AWS ELB) are more flexible and cost-effective.

Further Reading