Mini Load Balancing Demo with NGINX + Node.js

This repo is a tiny practical project that demonstrates some core distributed systems concepts:

• Latency vs Throughput
• Horizontal vs Vertical Scaling
• How Load Balancers work (Reverse Proxy, Forward Proxy)
• A hands-on demo using NGINX as a load balancer across two simple Node servers

---

Key Concepts (Beginner-Friendly)

Latency

• Meaning: The time taken for one request to travel from client → server → back to client.
• Think: How fast does one flashcard appear after I click “next”?
• Low latency = snappy, instant responses.
• High latency = laggy, delayed responses.

---

Throughput

• Meaning: The number of requests handled per second (capacity).
• Think: How many students can click “next card” at the same time without overloading the system?
• High throughput = can serve many users simultaneously.
• Low throughput = requests pile up, users wait.

---

Horizontal vs Vertical Scaling

• Vertical Scaling (Scale Up)

  • Add more power (CPU, RAM) to a single machine.
  • Easy to do, but limited — one machine can only get so big.
  • Analogy: Hire one super-teacher for the entire school.

• Horizontal Scaling (Scale Out)

  • Add more machines with the same code and spread load across them.
  • Virtually unlimited scaling.
  • Needs load balancing.
  • Analogy: Hire 10 normal teachers and share the students among them.

Benefits & Use Cases

• Vertical → good for small apps or quick fixes.
• Horizontal → essential for large systems like Google, Amazon, Netflix.

---

Load Balancers

A Load Balancer (LB) is like a traffic cop 🚦:

• Clients send requests to the LB (one public address).
• LB forwards each request to one of many backend servers.
• Spreads the load, improves reliability, and hides server details.

Reverse Proxy

• Sits in front of backend servers.
• Clients don’t know about the actual servers, only the proxy.
• Can do load balancing, SSL termination, caching, etc.
• Example: NGINX in this repo.

Forward Proxy

• Sits in front of clients.
• Client sends request to proxy → proxy fetches data on their behalf.
• Often used for privacy, filtering, or caching.
• Example: A school proxy that filters websites for students.

---

Common Load Balancing Algorithms

Load balancers use different strategies to decide which server should handle each request. The most popular ones are:

• Round Robin (default in NGINX)

  • Requests are sent to servers one by one in a cycle (A → B → C → A → B …).
  • Simple and effective when servers are similar in capacity.

• Least Connections

  • Sends traffic to the server with the fewest active connections.
  • Useful when requests vary in length (some short, some long).

• IP Hash (Sticky Sessions)

  • The client’s IP determines which server they always connect to.
  • Ensures the same user keeps hitting the same server (important if server stores session in memory).

• Weighted Round Robin

  • Each server is assigned a “weight” (e.g., Server A = 2, Server B = 1).
  • Higher-weighted servers get more requests.
  • Useful if some servers are more powerful than others.

---

🛠️ What This Repo Does

This is a mini demo of horizontal scaling with load balancing:

1. Created two identical Node.js servers running on different ports (3000 and 3001).

   • Each server replies with its port number so we know which one handled the request.

2. Configured NGINX as a reverse proxy / load balancer on port 8080.

   • Requests to http://localhost:8080/ are forwarded to Node server 3000 or 3001.
   • NGINX uses round robin (default) to alternate between servers.

3. Tested with Postman and curl.

   • Repeated requests show alternating responses:
   
   • This simulates horizontal scaling at a tiny level — multiple servers sharing load with a load balancer in front.

---

Why This Matters

• Even though this demo runs only on your laptop, it mirrors how big companies (Google, Amazon, Netflix) scale to millions of users.
• The same principles (load balancing, horizontal scaling, reducing latency, increasing throughput) apply whether you’re running 2 servers or 20,000.

---

How to Run Locally

1. Start two Node servers:

   PORT=3000 node backend/server.js
   PORT=3001 node backend/server.js