Load Balancer Project

Overview

This project implements a customizable load balancer that distributes client requests among several server replicas using consistent hashing. The primary goal is distributing the load evenly and efficiently among the servers using consistent hashing. The load balancer can handle server failures by spawning new replicas and ensuring services' high availability and scalability.

System Architecture

The load balancer routes client requests to server replicas, maintaining a consistent and balanced load distribution. It uses a consistent hashing mechanism with virtual servers to ensure minimal load shifts when scaling up or down.

Coding Environment

OS: Ubuntu 20.04 LTS or above Docker: Version 20.10.23 or above Languages: Python 3.8 or above

Installation

1. Install Docker

   sudo apt-get update
   sudo apt-get install ca-certificates curl gnupg lsb-release
   sudo mkdir -p /etc/apt/keyrings
   curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg
   echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
   sudo apt-get update
   sudo apt-get install docker-ce docker-ce-cli containerd.io
   

2. Install Docker Compose

   sudo curl -SL https://github.com/docker/compose/releases/download/v2.15.1/docker-compose-linux-x86_64 -o /usr/local/bin/docker-compose
    sudo chmod +x /usr/local/bin/docker-compose
    sudo ln -s /usr/local/bin/docker-compose /usr/bin/docker-compose
   

3. Install Python Dependencies

   sudo apt-get install python3-pip
   pip3 install -r requirements.txt
   

4. Clone the repository:

   git clone <repo-url>
   cd load_balancer_project
   
   

5. Build and run the services:

    make up
   
   
   

Usage

• Access the load balancer at http://localhost:5000

Testing

Run the tests using:

make test

Performance Analysis

Experiment 1: Load Distribution

1. Launch 10,000 asynchronous requests on 3 server containers.
2. Record the number of requests each server handles and plot a bar chart.

Observation

• The MD5 hash function resulted in an imbalanced load distribution among the servers

Experiment 2: Scalability

1. Increment the number of server containers from 2 to 6 (launching 10,000 requests each time).
2. Plot a line chart showing the average load of the servers at each run.

Expected Outcome

• Efficient scaling with even load distribution as server instances increase

Experiment 3: Failure Recovery

1. Test load balancer endpoints and simulate server failures.
2. Ensure the load balancer spawns new instances to handle the load and maintain the specified number of replicas.

Observations:

• The load balancer quickly detected the removal of 2 servers and redistributed the load.
• Response times remained stable, indicating efficient handling of server failures.
• The load balancer effectively scaled down when 2 servers were removed.
• Load distribution post-scaling was balanced, maintaining system performance.

Experiment 4: Hash Function Modification

1. Modify the hash function: i % 512 (number) of slots.
2. Repeat Experiments 1 and 2, analyzing the impact on load distribution and scalability.

• Experiment 1 Results

- Experiment 2 results