DEV Community: Subham Khandual | MERN Stack Developer

Vertical scaling vs Horizontal scaling

Subham Khandual | MERN Stack Developer — Mon, 05 Oct 2026 06:14:58 +0000

“Vertical vs Horizontal Scaling: Scale Up or Scale Out?”

Use the uploaded infographic as the main visual reference and structure the article around the concepts shown in it.

AUTHOR DETAILS:
Name: Subham Khandual
Role: Full Stack Developer | MERN Stack Developer | AI/ML Engineer | DevOps Engineer
Location: Bhubaneswar, Odisha, India

ARTICLE GOAL:
Explain Vertical Scaling and Horizontal Scaling in a beginner-friendly but professional System Design style. The article should be useful for students, developers preparing for system design interviews, backend developers, and DevOps beginners.

IMPORTANT TECHNICAL ACCURACY:
Do NOT falsely claim that WhatsApp's production architecture is purely vertically scaled or that BookMyShow's exact internal architecture is known.

Instead:

Use WhatsApp as a CONCEPTUAL example to explain the idea of Vertical Scaling / Scale Up.
Explain that if a messaging backend running on one server needed more capacity, upgrading CPU, RAM, or SSD would represent vertical scaling.
Clearly mention that real-world WhatsApp operates at massive scale and uses distributed/horizontally scaled infrastructure.
Use BookMyShow as a CONCEPTUAL example of Horizontal Scaling / Scale Out.
Explain how a high-traffic ticket-booking platform can use multiple application servers behind a Load Balancer.
Clearly state that this is a simplified educational architecture and not a claim about BookMyShow's private production infrastructure.

ARTICLE STRUCTURE:

Strong opening hook Start with a real-world problem: “What happens when your application suddenly gets 10x more users?”

Introduce scaling and why it matters in modern system design.

What is Scaling?
Explain scaling using a simple example:
4 GB RAM + 2 CPU cores → growing traffic → server becomes a bottleneck.
Vertical Scaling — Scale Up
Explain:
What it means
How it works
CPU upgrade
RAM upgrade
SSD/storage upgrade
Advantages
Disadvantages
When it makes sense

Include a simple ASCII architecture diagram.

Example:

Users
↓
Application Server
↓
Upgrade CPU + RAM + Storage
↓
More powerful server

WhatsApp CONCEPTUAL EXAMPLE Explain a simplified messaging-server scenario.

Before:
Users → One Server

After Scale Up:
Users → More Powerful Server

Use examples such as:
2 CPU → 8 CPU
4 GB RAM → 32 GB RAM

Then add an important note:
“This is a conceptual example for understanding vertical scaling. WhatsApp's actual production infrastructure is distributed and operates across many systems and servers.”

Horizontal Scaling — Scale Out Explain:
What it means
Adding multiple servers
Load Balancer
Traffic distribution
Fault tolerance
High availability
Elastic growth

Show:

            Users
              ↓
        Load Balancer
              ↓
    ┌─────────┼─────────┐
    ↓         ↓         ↓
 Server 1  Server 2  Server 3

BookMyShow CONCEPTUAL EXAMPLE Explain a movie-ticket booking traffic spike.

Example:
Normal traffic:
3 application servers

During a major movie release:
3 → 6 → 10 application servers

Architecture:

                Users
                  ↓
           Load Balancer
                  ↓
      ┌───────────┼───────────┐
      ↓           ↓           ↓
   Server 1    Server 2    Server 3
      │           │           │
      └───────────┼───────────┘
                  ↓
               Database

Explain how horizontal scaling helps distribute requests.

Clearly mention:
“This is a simplified educational model, not BookMyShow's actual private architecture.”

Vertical vs Horizontal Scaling Create a clean comparison table covering:

Meaning
Approach
Servers
Hardware limits
Complexity
Load Balancer
Availability
Fault tolerance
Cost
Downtime
Best use cases

Advantages and Disadvantages

Vertical Scaling:

Advantages:

Simple architecture
Easy to implement
Less operational complexity
Good for small applications

Disadvantages:

Hardware limits
Expensive high-end machines
Potential downtime during upgrades
Single-server failure can affect availability

Horizontal Scaling:

Advantages:

Better scalability
Higher availability potential
Fault tolerance
Easier capacity expansion
Supports large traffic volumes

Disadvantages:

More complex architecture
Requires load balancing
Distributed state becomes a challenge
Database scaling becomes important
Monitoring and deployment become more complex

What Problems Appear After Horizontal Scaling?

Explain:

Load balancing
Database bottlenecks
Session management
Stateless architecture
Redis/cache
Database replication
Database sharding
Message queues
CDN

Show how the architecture evolves:

Users
↓
Load Balancer
↓
Multiple App Servers
↓
Redis / Cache
↓
Database Cluster

When Should You Scale Up?

Explain that vertical scaling is useful for:

MVPs
Small applications
Simple architectures
Low/moderate traffic
Systems where operational simplicity matters

When Should You Scale Out?

Explain that horizontal scaling becomes useful for:

High traffic
Rapid growth
High availability
Fault tolerance
24/7 systems
Large distributed applications

Real-World Systems Usually Use BOTH

Explain that vertical and horizontal scaling are not competitors.

A modern architecture may use:

Horizontal scaling for application servers
Vertical scaling for individual database/cache nodes
Read replicas
Sharding
Redis
CDN
Load balancers
Auto scaling

Show a final architecture diagram.

Simple Memory Trick

Make this visually memorable:

VERTICAL = SCALE UP
“Make one machine stronger.”

HORIZONTAL = SCALE OUT
“Add more machines.”

Use:
🖥️ → 🖥️💪

versus

🖥️ + 🖥️ + 🖥️ + 🖥️

Final Takeaway

End with a strong conclusion explaining:

Scaling is not simply about adding CPU or RAM.

The real system-design challenge is:
“How can we handle increasing traffic while maintaining performance, availability, reliability and manageable complexity?”

Mention that scaling naturally leads to concepts such as:
Load Balancers, Redis, Database Replication, Sharding, CDN, Message Queues and Distributed Systems.

PERSONAL LEARNING SECTION

Add a section:

“What I Learned”

Write it in first person as Subham Khandual.

Mention that while learning System Design, I realized scaling is not just about making a server more powerful. The bigger challenge is distributing workload efficiently while keeping the system reliable, available and maintainable.

Keep it authentic and not overly promotional.

WRITING STYLE:

Professional but beginner-friendly
Clear explanations
Practical examples
Short paragraphs
Strong headings
Use emojis sparingly
Use Markdown
Use code blocks for architecture diagrams
Avoid unnecessary jargon
Explain technical terms when first introduced
Do not sound like AI-generated content
Make it feel like a developer sharing what they learned
Include practical System Design insights
Do not overclaim knowledge of private company architectures

DEV.TO SEO:
Create:

SEO-friendly title
Short subtitle/excerpt
Article body
Suggested canonical title
5–8 relevant DEV.to tags

Recommended tags:

systemdesign

scalability

backend

devops

softwareengineering

architecture

webdevelopment

distributedsystems

TITLE SHOULD BE ENGAGING:
Prefer something similar to:
“Scale Up or Scale Out? Understanding Vertical vs Horizontal Scaling 🚀”

Also provide 3 alternative titles.

VISUAL:
Use the uploaded infographic naturally in the article as the introductory/reference image. Explain that the visual summarizes the fundamental difference between Scale Up and Scale Out.

FINAL ARTICLE REQUIREMENTS:

Make the article publication-ready for DEV.to.
Do not include instructions to the writer.
Do not mention “as an AI”.
Do not fabricate internal architecture details about WhatsApp or BookMyShow.
Keep the technical explanation accurate.
Make the article engaging enough that a developer would want to read until the end.
Include useful architecture diagrams.
End with a concise takeaway and relevant hashtags.
Author name: Subham Khandual.

System design(Load Balancer)

Subham Khandual | MERN Stack Developer — Sun, 04 Oct 2026 19:50:22 +0000

🚀 SYSTEM DESIGN — LOAD BALANCER

Write the post as if it is being shared by a passionate Full Stack Developer, AI/ML Engineer, and DevOps Engineer who is continuously learning System Design and sharing useful technical knowledge with the developer community.

The post should feel HUMAN, confident, educational, and authentic — NOT robotic or AI-generated.

━━━━━━━━━━━━━━━━━━━━
🔥 OPENING HOOK
━━━━━━━━━━━━━━━━━━━━

Start with a strong hook that makes developers want to read further.

Example:

“🚦 What happens when 100,000 users suddenly hit your application at the same time?

One server can’t handle everything.

This is where a Load Balancer becomes one of the most important components in System Design. ⚖️”

Then briefly introduce the concept.

━━━━━━━━━━━━━━━━━━━━
⚖️ WHAT IS A LOAD BALANCER?
━━━━━━━━━━━━━━━━━━━━

Explain in simple language:

A Load Balancer sits between clients and backend servers and intelligently distributes incoming requests across multiple servers.

Architecture:

👥 Clients
↓
⚖️ Load Balancer
↓
🖥️ Server 1
🖥️ Server 2
🖥️ Server 3

Explain that the goal is not simply to “divide traffic equally,” but to keep the system reliable, responsive, scalable, and available.

━━━━━━━━━━━━━━━━━━━━
📸 EXPLAIN IMAGE 1
━━━━━━━━━━━━━━━━━━━━

Use the first image as the architecture explanation.

Explain:

Client → Load Balancer → Multiple Backend Servers

Requests can come from:
🌐 Web browsers
📱 Mobile applications
💻 Desktop/Laptop applications
🔗 Other services/APIs

The Load Balancer receives these requests and routes them to appropriate backend servers.

Mention that real-world load balancers can use server health, connection count, routing rules, weights, and other factors when deciding where to send traffic.

━━━━━━━━━━━━━━━━━━━━
👥 EXPLAIN IMAGE 2 — 100 USERS EXAMPLE
━━━━━━━━━━━━━━━━━━━━

Use the second image to explain the concept with a simple example.

Imagine:

👥 100 users are accessing an application simultaneously.

❌ WITHOUT LOAD BALANCER

100 users → 🖥️ One Server

Possible problems:
• Server becomes overloaded
• Response time increases
• CPU/RAM resources are exhausted
• Requests may fail
• Server may crash
• Application availability is affected

Then explain:

✅ WITH LOAD BALANCER

100 users
↓
⚖️ Load Balancer
↙ ↓ ↘
🖥️ S1 🖥️ S2 🖥️ S3

The requests can be distributed across multiple healthy servers.

For example:

Server 1 → ~33 users
Server 2 → ~34 users
Server 3 → ~33 users

Clarify that this is a simplified example. Real systems do NOT necessarily distribute traffic perfectly equally.

━━━━━━━━━━━━━━━━━━━━
🛡️ WHY DO WE NEED A LOAD BALANCER?
━━━━━━━━━━━━━━━━━━━━

Create a visually attractive section:

⚡ Better Performance
🛡️ High Availability
📈 Horizontal Scalability
🔄 Traffic Distribution
❤️ Health Checks
🚨 Fault Tolerance
🔧 Easier Infrastructure Scaling

Briefly explain each benefit in one sentence.

━━━━━━━━━━━━━━━━━━━━
⚙️ COMMON LOAD BALANCING ALGORITHMS
━━━━━━━━━━━━━━━━━━━━

Explain these in simple developer-friendly language:

🔹 Round Robin
Requests are distributed sequentially across servers.

🔹 Weighted Round Robin
More powerful servers receive a larger share of traffic.

🔹 Least Connections
Traffic goes toward the server handling fewer active connections.

🔹 IP Hash
A client's IP is used to determine the destination server.

🔹 Least Response Time
Requests can be directed toward servers responding faster.

Mention that the best strategy depends on the application's architecture and traffic pattern.

━━━━━━━━━━━━━━━━━━━━
❤️ HEALTH CHECKS
━━━━━━━━━━━━━━━━━━━━

Explain why health checks matter.

For example:

If Server 2 becomes unhealthy:

Client → Load Balancer
↓
❌ Server 2
↙ ↘
Server 1 Server 3

The Load Balancer can detect unhealthy instances and stop sending new traffic to them, depending on the configuration.

━━━━━━━━━━━━━━━━━━━━
🏗️ REAL-WORLD USE CASES
━━━━━━━━━━━━━━━━━━━━

Mention:

🛒 E-commerce
🏦 Banking
📱 Mobile applications
🌐 SaaS platforms
☁️ Cloud applications
🎥 Streaming platforms
🤖 AI/ML applications
🏢 Enterprise systems
🚀 High-traffic web applications

━━━━━━━━━━━━━━━━━━━━
🧠 SYSTEM DESIGN TAKEAWAY
━━━━━━━━━━━━━━━━━━━━

Create a strong takeaway:

“Load Balancing is not just about distributing requests.

It is about designing systems that can continue serving users reliably as traffic grows and individual servers fail.”

Then explain that Load Balancers are commonly combined with:

⚡ Auto Scaling
🗄️ Databases
⚡ Caching
📨 Message Queues
🌐 CDN
🔐 API Gateway
☁️ Cloud Infrastructure
📊 Monitoring & Observability

This shows how Load Balancing fits into a larger System Design architecture.

━━━━━━━━━━━━━━━━━━━━
📚 MY LEARNING JOURNEY
━━━━━━━━━━━━━━━━━━━━

Add a personal section:

“📚 What I’m Learning”

“I’m currently strengthening my understanding of System Design, scalable backend architecture, cloud infrastructure, DevOps, distributed systems, and high-availability applications.

Every concept — from Load Balancers and Caching to Database Scaling and Message Queues — helps me understand how real-world applications handle millions of requests.”

Keep this authentic and concise.

━━━━━━━━━━━━━━━━━━━━
👨‍💻 ABOUT ME
━━━━━━━━━━━━━━━━━━━━

Add:

👨‍💻 Subham Khandual

Full Stack Developer | MERN Stack Developer | AI/ML Engineer | DevOps Engineer

💡 Interested in:
• Full Stack Development
• MERN Stack
• AI/ML
• DevOps
• Cloud Computing
• System Design
• Distributed Systems
• Scalable Applications

🔗 LinkedIn:
https://www.linkedin.com/in/subham-khandual/

💻 GitHub:
https://github.com/subham-khandual

━━━━━━━━━━━━━━━━━━━━
💬 ENGAGEMENT
━━━━━━━━━━━━━━━━━━━━

End with an engaging question:

“Which System Design concept should I break down next? 👇

1️⃣ Caching
2️⃣ Database Scaling
3️⃣ API Gateway
4️⃣ Message Queues
5️⃣ Microservices

Let me know in the comments! 🚀.

Suraksha Setu: Project overview

Subham Khandual | MERN Stack Developer — Fri, 15 May 2026 13:50:01 +0000

🚨 Suraksha Setu: Building a Smart Tourist Safety System with the MERN Stack

👋 Introduction

Hi, I’m Subham Khandual — a MERN Stack Developer and Final Year B.Tech CSE student from Bhubaneswar, Odisha, India.

Recently, I built one of the most meaningful projects of my development journey: Suraksha Setu — a Smart Tourist Safety System designed to help travelers stay safe through emergency assistance, live safety alerts, and real-time travel monitoring.

🔴 Live Demo: https://suraksha-setu-oxkw.vercel.app

🐙 GitHub: https://github.com/subham-khandual
🌐 Portfolio: https://portfolio-nine-alpha-8nkzp7nnk6.vercel.app

🌍 The Idea Behind Suraksha Setu

India attracts millions of tourists every year, but traveler safety remains a major concern — especially in unfamiliar cities, remote destinations, and emergency situations.

As someone from Odisha, I wanted to build something that could genuinely help people during travel emergencies.

“What if tourists had a digital safety companion that could assist them instantly during emergencies?”

That idea became Suraksha Setu.

🛡️ Features of Suraksha Setu

🚨 Emergency Assistance

One-tap SOS functionality
Emergency contact access
Quick response support system

📡 Real-Time Safety Alerts

Area-based safety notifications
Weather and risk-level alerts
Local travel advisories

📍 Smart Travel Monitoring

Live travel dashboard
Monitoring support for solo travelers
Mobile-friendly responsive interface

🛠️ Tech Stack Used

Layer	Technology
Frontend	React.js + Tailwind CSS
Backend	Node.js + Express.js
Database	MongoDB
Authentication	JWT / Clerk
Deployment	Vercel

🚧 Challenges I Faced

1️⃣ Designing for Emergency Situations

In safety applications, users may be stressed or panicked. I focused on reducing clicks and making critical actions visible immediately.

2️⃣ Real-Time Alert Handling

Implementing live safety updates taught me a lot about state management and real-time data flow in React.

3️⃣ Mobile Responsiveness

Since tourists mostly use smartphones, building a responsive UI with Tailwind CSS became a top priority.

4️⃣ Privacy & Location Data

Handling user location responsibly was important. I worked on creating a transparent and trustworthy experience around location permissions.

📚 What I Learned

Building Suraksha Setu improved my skills in:

MERN Stack architecture
REST API development
Responsive UI/UX design
Real-time data handling
Deployment using Vercel
Problem-solving through real-world applications

🔮 Future Improvements

I plan to add:

📲 Progressive Web App (PWA) support
🗺️ Google Maps integration
🔔 Push notifications
🌐 Multi-language support
🤝 Integration with emergency services

🌟 Other Projects

🏥 Swasthya Setu

AI-powered healthcare monitoring platform using MERN + Machine Learning.

🔴 Live Demo: https://swasthyasetu-tsbt.vercel.app

🩸 Blood Cancer Prediction

Machine learning model for cancer risk detection.

🌾 Crop Recommendation System

AI-powered crop suggestion platform for farmers.

🤖 AI Assistant

Python-based conversational AI assistant.

🤝 Let’s Connect

🌐 Portfolio: https://portfolio-nine-alpha-8nkzp7nnk6.vercel.app

🐙 GitHub: https://github.com/subham-khandual
💼 LinkedIn: https://www.linkedin.com/in/subham-khandual

If you enjoyed this project, feel free to ⭐ the repository and connect with me!

Project overview

Subham Khandual | MERN Stack Developer — Thu, 14 May 2026 03:33:00 +0000

How I Built Swasthya Setu — An AI-Powered Health App Using the MERN Stack

By Subham Khandual | MERN Stack Developer | Final Year CSE Student, Bhubaneswar, India

👋 Introduction

Hi, I'm Subham Khandual, a Final Year B.Tech Computer Science student from Bhubaneswar, Odisha, India. I'm a MERN Stack Developer passionate about using technology to solve real-world problems.

In this post, I want to share the story behind Swasthya Setu — an AI-powered health monitoring web application I built as my final year project. I'll walk you through the problem I was trying to solve, the tech stack I chose, the challenges I faced, and what I learned along the way.

👉 Live Demo: https://swasthyasetu-tsbt.vercel.app
👉 GitHub: https://github.com/subham-khandual

🏥 What is Swasthya Setu?

Swasthya Setu (meaning "Health Companion" in Hindi/Odia) is an AI-powered health app designed to help patients and healthcare providers monitor health data intelligently.

The core idea was simple: millions of people in India, especially in Tier 2 and Tier 3 cities, don't have easy access to regular health check-ins. What if a web app could act as a smart health companion — tracking vitals, flagging risks, and giving users actionable insights?

Key Features:

🤖 AI-based health risk analysis
📊 Patient dashboard with real-time health metrics
🔒 Secure authentication and user data privacy
📱 Fully responsive design for mobile users
🏥 Healthcare provider interface

🛠️ Tech Stack

Here's everything I used to build Swasthya Setu:

Layer	Technology
Frontend	React.js, Tailwind CSS, Vite
Backend	Node.js, Express.js
Database	MongoDB (Atlas)
Authentication	JWT, Clerk
AI/ML	Python-based prediction model
Deployment	Vercel (Frontend), Render (Backend)

I chose the MERN stack because it allows full JavaScript development across the entire application — from the React frontend to the Node.js backend to MongoDB. This consistency made it much easier to develop and debug as a solo developer.

💡 Why I Built This

During my third year, I noticed how underserved rural healthcare really is in India. A close family experience made me think: what if patients had a smart app that could monitor their health data and alert them before things got serious?

That thought became Swasthya Setu.

I also wanted to challenge myself technically — integrating AI/ML into a MERN app was something I hadn't done before, and I knew it would push my skills significantly.

🚧 Challenges I Faced

1. Integrating Python ML Model with Node.js Backend

The biggest challenge was connecting a Python-based machine learning prediction model with my Node.js backend. I solved this by running the ML model as a microservice and communicating via REST API calls.

2. Handling Real-Time Data

Displaying health metrics in real time on the patient dashboard required careful state management in React. I used useEffect hooks and polling to keep the dashboard updated without overloading the server.

3. Designing for Non-Tech Users

Healthcare users — especially older patients — aren't always tech-savvy. I had to keep the UI extremely simple and intuitive, which meant several redesigns of the dashboard.

4. Deployment Configuration

Getting the frontend (Vercel) and backend (Render) to communicate correctly in production, with proper CORS settings and environment variables, took longer than expected. But it was a valuable DevOps lesson.

📚 What I Learned

Building Swasthya Sathi as a solo MERN Stack Developer taught me:

Full Stack Architecture — designing a system end-to-end, from database schema to UI
API Design — creating clean, RESTful APIs with proper error handling
AI Integration — connecting ML models to web applications
React State Management — handling complex data flows in a real app
Deployment & DevOps — managing production environments across multiple platforms
User-Centric Design — building for real users, not just for developers

🔗 Other Projects I've Built

Alongside Swasthya Sathi, I've been working on several other projects:

🩸 Blood Cancer Prediction — ML model using Jupyter Notebook to predict blood cancer risk
🌾 Crop Recommendation System — ML-based crop suggestion app for farmers
🛡️ Suraksha Setu — Another web app focused on safety (Live Demo)
🤖 AI Assistant — A conversational AI assistant built with Python

All projects are open source on my GitHub: github.com/SubhamKhandual007

🚀 What's Next for Swasthya Setu

I'm planning to add:

Real-time notifications via WebSockets
Wearable device data integration
Multilingual support (Odia, Hindi, English)
Doctor appointment booking module

🤝 Let's Connect

I'm Subham Khandual, a MERN Stack Developer actively looking for opportunities — internships, collaborations, or full-time roles after graduation.

If you're working on something interesting in web development, AI, or healthtech, I'd love to connect!

🌐 Portfolio: subhamportfoliodemo.vercel.app
💼 LinkedIn: linkedin.com/in/subham-khandual
🐙 GitHub: github.com/subham-khandual
📧 Email: subhamkhandual215@gmail.com

If you found this post helpful or interesting, drop a ❤️ and follow me for more content on MERN Stack development, AI integration, and real-world project builds!

Tags: #mernstack #react #nodejs #mongodb #webdev #healthtech #ai #javascript #beginners #showdev