Express.js-based load balancer that uses a weighted round-robin algorithm to distribute incoming requests across multiple servers.
Note: An article on this project is available on Medium, check it out.
A load balancer is a crucial component in distributing incoming network traffic across multiple servers to ensure optimal resource utilization, maximize throughput, and minimize response time. This document provides an overview of how a simple load balancer works, focusing on a scenario with ten servers and using a Weighted Round Robin algorithm.
In our example, we have ten servers, each assigned a specific weight. The weight represents the server's capacity to handle requests, and it influences the distribution of incoming traffic.
graph TD;
subgraph Servers
Server1
Server2
Server3
Server4
Server5
Server6
Server7
Server8
Server9
Server10
end
LoadBalancer -->|Weighted Distribution| Server1
LoadBalancer -->|Weighted Distribution| Server2
LoadBalancer -->|Weighted Distribution| Server3
LoadBalancer -->|Weighted Distribution| Server4
LoadBalancer -->|Weighted Distribution| Server5
LoadBalancer -->|Weighted Distribution| Server6
LoadBalancer -->|Weighted Distribution| Server7
LoadBalancer -->|Weighted Distribution| Server8
LoadBalancer -->|Weighted Distribution| Server9
LoadBalancer -->|Weighted Distribution| Server10
The load balancer employs a Weighted Round Robin algorithm to distribute incoming requests among the servers. This algorithm takes into account the weights assigned to each server and aims to maintain a balance in the workload.
The algorithm operates as follows:
- Servers are configured with specific weights reflecting their capacity.
- The load balancer maintains an index pointing to the current server in rotation.
- For each incoming request, the load balancer selects the next server in a round-robin fashion based on the server weights.
- The server index is updated, and the process repeats.
When a user makes a request to the load balancer, the Weighted Round Robin algorithm determines which server should handle the request. The algorithm considers the server weights, directing traffic to servers with higher capacities more frequently. Here is a visual representation of the load balancer distributing requests among the servers using the Weighted Round Robin algorithm:
sequenceDiagram
participant User1
participant User2
participant User3
participant LoadBalancer
participant Server1
participant Server2
participant Server3
participant Server4
participant Server5
participant Server6
participant Server7
participant Server8
participant Server9
participant Server10
User1->>LoadBalancer: Request to 8080
User2->>LoadBalancer: Request to 8080
User3->>LoadBalancer: Request to 8080
LoadBalancer->>Server1: Weighted Distribution
LoadBalancer->>Server2: Weighted Distribution
LoadBalancer->>Server3: Weighted Distribution
LoadBalancer->>Server4: Weighted Distribution
LoadBalancer->>Server5: Weighted Distribution
LoadBalancer->>Server6: Weighted Distribution
LoadBalancer->>Server7: Weighted Distribution
LoadBalancer->>Server8: Weighted Distribution
LoadBalancer->>Server9: Weighted Distribution
LoadBalancer->>Server10: Weighted Distribution
Server1-->>LoadBalancer: Response
Server2-->>LoadBalancer: Response
Server3-->>LoadBalancer: Response
Server4-->>LoadBalancer: Response
Server5-->>LoadBalancer: Response
Server6-->>LoadBalancer: Response
Server7-->>LoadBalancer: Response
Server8-->>LoadBalancer: Response
Server9-->>LoadBalancer: Response
Server10-->>LoadBalancer: Response
LoadBalancer-->>User1: Aggregated Response
LoadBalancer-->>User2: Aggregated Response
LoadBalancer-->>User3: Aggregated Response
This diagram visually represents the flow of requests from users through the load balancer to the servers, demonstrating the Weighted Round Robin algorithm in action. The load balancer ensures an even distribution of requests based on the specified weights, optimizing the performance of the entire system.
- Node.js
- npm
- Python 3
- npx
- pm2
- Clone the repository
git clone https://github.com/hendurhance/express-balancify.git- Change directory
cd express-balancify- Install dependencies
npm install- Start the servers
npm run start- Start the load balancer
npm run start:balancer- Test the load balancer
The available routes are
/,/api. You can access the load balancer athttp://localhost:8000health check route andhttp://localhost:8000/apifor the API route.
| Route | Description |
|---|---|
/ |
It is an health check route that returns a JSON object with the message |
/api |
It is a route that returns a JSON object of a sales sample data of up to 3000 records |
To test the load balancer, you can use the following command:
# Health check route
npm run test:load:health
# API route
npm run test:load:apiThis will send 1200 requests to the load balancer and 400 concurrent requests at a time. You can change the number of requests and concurrent requests by changing the values of in the package.json file.
Requests and their details, such as timestamp, IP address, server port, and original URL, are logged to a CSV file (request_logs.csv). The log file contains the following information:
# Log file content
Timestamp,IP Address,Server Port,Original URL
2023-12-17T17:39:46.250Z,::1,5673,/You can analyze the log file using the run the following command:
npm run analyzeThe script generates a report and visualizations, including bar charts for the distribution of requests between server ports, the highest amount of requests handled by each server, and a pie chart showing the percentage distribution of requests between server ports.
Feel free to contribute to this project by opening issues or submitting pull requests. Your feedback and contributions are highly appreciated.
This project is licensed under the MIT License. See the LICENSE file for details.
If you found this project useful, please consider starring it. You can also buy me a coffee ☕️ or become a sponsor using the links below ➡️ github sponsors. Thanks a bunch for your support!