Go Load Balancer

A simple load balancer implemented in Go with round-robin request forwarding, health checks and connection pooling. The load balancer forwards requests to backend servers, performs periodic health checks and reuses HTTP connections to optimize performance.

10f8b66fc64dc6b0fcfb487e0cb82001.mp4

Features

• Round-Robin Load Balancing: Distributes incoming requests across multiple backend servers in a round-robin fashion
• Health Checks: Periodically checks backend servers for availability and skips unhealthy servers
• Connection Pooling: Reuses existing HTTP connections to reduce the overhead of creating new connections
• Concurrent Request Handling: Uses Go's http.ListenAndServe to handle requests concurrently

Requirements

• Go 1.23+ is required to build and run this project.

Getting Started

Clone the Repository

git clone git@github.com:rjp2525/load-balancer.git && cd load-balancer

Install Dependencies

go mod tidy

Running the Load Balancer

1. You'll need at least two backend servers listening on different ports to act as targets for the load balancer. The balancer is setup to check /up for Laravel projects, but you can create a fake backend server to listen on for testing purposes:

package main

import (
    "fmt"
    "net/http"
)

func healthCheckHandler(w http.ResponseWriter, r *http.Request) {
    fmt.Fprintln(w, "OK")
}

func main() {
    http.HandleFunc("/up", healthCheckHandler)

    fmt.Println("Starting server on :8000")
    if err := http.ListenAndServe(":8000", nil); err != nil {
        fmt.Printf("Error starting server: %s\n", err)
    }
}

Run this server on different ports (e.g. 8000, 8001) to simulate the backend servers.

2. Once your backend servers are running, run the load balancer:

go run main.go

Example

Once the load balancer is running, you can forward requests to the backend servers via the load balancer:

curl http://127.0.0.1:8080/

The load balancer will forward the request to one of the healthy backend servers in a round-robin fashion, and you will see the response from the backend server.

Configuration (Optional)

You can configure the backend servers and port for the load balancer using a config.json file, for example:

{
    "port": 8080,
    "servers": [
        "http://127.0.0.1:8000",
        "http://127.0.0.1:8001"
    ]
}

To use this configuration, ensure the file is located in the root of the project and the load balancer will read from it automatically.

Testing

Unit tests are available for the load balancer including health checks, connection pooling and round-robin request distribution.

Run the tests using:

go test ./balancer

You should see output indicating whether the tests passed or failed.

License

This project is licensed under the MIT License.