8. RP vs LB vs GWs

---

📌 Reverse Proxy vs Load Balancer vs API Gateway

Core Idea:
All 3 sit between Client ↔ Server, but each solves a different infrastructure problem.

---

⚡ Quick Responsibility (Interview Revision)

Reverse Proxy

Feature	Purpose
SSL Termination	Remove encryption overhead
Caching	Serve repeated responses
Compression	Reduce payload size
Security	Hide backend + filter attacks
Routing	Forward requests

Load Balancer

Feature	Purpose
Traffic Distribution	Spread requests
Health Checks	Detect failures
Failover	Skip unhealthy servers
Horizontal Scaling (ASG)	Add more machines

API Gateway

Feature	Purpose
Authentication	JWT / API keys
Authorization	Access control
Rate Limiting	Protect APIs
Versioning	/v1, /v2
Transformation	JSON ↔ XML
Analytics	API metrics

---

🏗️ How They Work Together (Real Architecture)

User
 ↓
CDN  → Global delivery
 ↓
API Gateway → API policies
 ↓
Load Balancer  → Traffic distribution
 ↓
Service Instances
 ↓
Reverse Proxy → Optimization & protection
 ↓
Application

---

🎯 Decision Rule

Need SSL / Cache / Security?
→ Reverse Proxy
 
Need Multiple Servers?
→ Load Balancer
 
Need Auth / Rate Limiting / Public APIs?
→ API Gateway

---

1️⃣ Reverse Proxy → Protect + Optimize Servers

What is it?

A server-side proxy that sits in front of backend servers.

Client
 ↓
Reverse Proxy
 ↓
Backend

Users never directly access backend servers.

Why Do We Need It?

Backend servers should focus only on business logic.

Move infrastructure concerns outside:

• SSL termination
• Caching
• Compression
• Security filtering
• Static file serving

Responsibilities

Feature	Purpose
SSL Termination	Remove encryption overhead
Caching	Serve repeated responses
Compression	Reduce payload size
Security	Hide backend + filter attacks
Routing	Forward requests

Example Flow

User
 ↓
Reverse Proxy (NGINX)
 ↓
Application Server

Popular Tools

• NGINX
• HAProxy
• Caddy
• Envoy

---

2️⃣ Load Balancer → Scale + High Availability

What is it?

A specialized reverse proxy that distributes traffic across multiple servers.

Client
 ↓
Load Balancer
 ↓
Multiple Servers

Why Do We Need It?

A single server eventually hits limits:

• CPU
• Memory
• Network connections Solution:

Add more servers
↓
Distribute traffic

Responsibilities

Feature	Purpose
Traffic Distribution	Spread requests
Health Checks	Detect failures
Failover	Skip unhealthy servers
Horizontal Scaling	Add more machines

Common Algorithms

Round Robin

Req1 → A
Req2 → B
Req3 → C

Least Connections

Send traffic to least busy server

Weighted Routing

Powerful servers → Receive more traffic

IP Hash

Same user → Same server

Layer 4 vs Layer 7

Layer 4 (NLB)	Layer 7 (ALB)
TCP/IP	HTTP
Faster	Smarter
No URL awareness	Reads URL & Headers

Examples:

• Layer 4 → TCP balancing
• Layer 7 → /users → Cluster A

Popular Tools

• HAProxy
• NGINX
• AWS Application Load Balancer
• AWS Network Load Balancer

---

3️⃣ API Gateway → Manage APIs

What is it?

An API-aware reverse proxy.

Client
 ↓
API Gateway
 ↓
Microservices

Why Do We Need It?

Microservices duplicate infrastructure logic:

• Authentication
• Rate limiting
• Monitoring
• Transformation

Centralize everything.

Responsibilities

Feature	Purpose
Authentication	JWT / API keys
Authorization	Access control
Rate Limiting	Protect APIs
Versioning	/v1, /v2
Transformation	JSON ↔ XML
Analytics	API metrics

Example

/api/users
     ↓
API Gateway
     ↓
User Service

Popular Tools

• Kong Gateway
• AWS API Gateway
• Google Apigee
• Tyk

---

📊 Comparison Table

Feature	Reverse Proxy	Load Balancer	API Gateway
Main Goal	Protect	Scale	Manage APIs
Multiple Servers	Optional	Yes	Usually
SSL	✅	✅	✅
Caching	✅	Sometimes	Sometimes
Health Check	❌	✅	✅
Traffic Distribution	Basic	Advanced	API-based
Authentication	❌	❌	✅
Rate Limiting	❌	❌	✅
Versioning	❌	❌	✅

---