System design for Global level application - For beginners

-

Basic System design for Global level application


1. Client and Server Roles in System Design

  • Client: The client is typically a user's device (mobile app, browser, etc.) that sends a request to get some data or perform some action.

  • Server: This is a machine (computer) that runs 24/7 and listens for client requests, processes them, and returns a response.

  • Think of it like ordering food online (client) and the restaurant kitchen preparing and delivering it (server).

2. Public IP Address

  • For a server to be accessed over the internet, it needs a public IP address.

  • This makes it globally accessible.

  • Example: Just like your phone has a unique number, servers have IPs.

3. Domain Names Instead of IP

  • IP addresses are hard to remember (like 142.250.78.206), so we use domain names like amazon.com.

  • These names are easier for humans to remember and use.

4. What is a DNS Server?

  • DNS (Domain Name System) is like a phonebook of the internet.

  • It converts domain names into IP addresses.

  • When you type google.com, DNS helps your computer find the right server to talk to.

5. DNS Resolution

  • This process of converting a domain name into an IP is called DNS resolution.

  • Your browser sends a query to a DNS server asking, "What is the IP of amazon.com?"

  • DNS replies with the correct IP.

6. Server Scaling (Why It's Needed)

  • If many users use your app at the same time, your server may slow down or crash.

  • Scaling means preparing your server to handle more users or traffic.

7. Vertical Scaling

  • Add more power to a single server: increase CPU, RAM, etc.

  • Called "scaling up".

  • Good for small to medium traffic.

  • But… it has limits.

8. Problems with Vertical Scaling

  • Server might need a restart after upgrades → downtime (bad for apps like Amazon).

  • Also, one big server = single point of failure.

9. Horizontal Scaling

  • Instead of one powerful server, you use multiple small servers.

  • Called "scaling out".

  • Spread traffic among them using a Load Balancer.

10. What is a Load Balancer?

  • Distributes traffic to multiple servers.

  • Uses strategies like:

    • Round Robin: turn by turn

    • Least Connections: least busy server

  • Ensures no single server is overwhelmed.

11. Health Checks by Load Balancer

  • Load balancers continuously check server health.

  • If a server crashes, it won’t send traffic there.

  • New servers must register with the load balancer.

12. Microservices Architecture

  • Break a large app into small services:

    • Auth Service, Order Service, Payment Service, etc.

  • Each service has its own server(s) and can be scaled independently.

13. API Gateway

  • A single entry point for all API requests.

  • Routes incoming requests to the right microservice.

  • Acts as a reverse proxy.

  • Helps with authentication, rate limiting, and logging.

14. Load Balancer Behind API Gateway

  • API Gateway forwards requests to the load balancer of the appropriate service.

  • Load balancer then sends the request to a healthy server (like an EC2 instance).

15. Authenticator with API Gateway

  • API Gateway can have an authentication layer.

  • Ensures only valid users can access the backend services.

16. Batch Processing

  • Used for background tasks like:

    • Sending emails

    • Generating reports

  • A worker (background process) reads from the database and performs these tasks.

17. Async Communication with Queues

  • Instead of calling services directly, one service can push a message to a queue (like SQS).

  • Worker pulls messages from queue and processes them.

  • Helps in decoupling services and retrying failed tasks.

18. Horizontal Scaling of Workers

  • You can run multiple workers in parallel.

  • Each can process a message simultaneously = faster processing.

19. Rate Limiting in Workers

  • You can set limits, like "send only 10 emails per second", to:

    • Avoid spam

    • Respect 3rd-party API limits

20. Async Communication Mechanisms

  • SQS (Simple Queue Service) can act as:

    • Push (messages sent to queue)

    • Pull (worker pulls when ready)

  • Good for decoupling microservices.

21. Pub/Sub Model

  • One event → multiple services react.

  • Example: On payment success:

    • Send email

    • Send SMS

    • Update dashboard

  • Each service subscribes to the same event.

22. Pub/Sub vs SQS

  • Pub/Sub → message goes to all subscribers.

  • SQS → message goes to one consumer.

23. Event-Driven Architecture

  • Microservices respond to events instead of direct calls.

  • Problem: Sometimes services may not acknowledge (ack) receipt of event.

24. Fan-Out Architecture

  • One event → message sent to multiple queues:

    • WhatsApp Queue

    • Email Queue

    • SMS Queue

  • Each queue has its own worker.

25. Fan-Out Use Case

  • Example: A video is uploaded.

    • Transcribe audio

    • Generate thumbnail

    • Process video

  • One event triggers multiple actions in parallel.

26. Rate Limiting

  • Helps prevent abuse or overload.

  • Example: Max 100 API calls per minute per user.

  • Protects backend from traffic spikes.

27. Database Scaling

  • One database can’t handle huge traffic alone.

  • Use read replicas to handle more traffic:

    • Primary handles writes

    • Replicas handle reads

28. Read Replicas

  • Help offload the main database.

  • But... have slight delay (eventual consistency).

29. Caching Layer

  • Frequently requested data is stored in cache (e.g., Redis).

  • Avoids repeated database hits.

  • Makes system faster and more efficient.

30. Content Delivery Network (CDN)

  • CDN stores static content (images, videos, CSS, JS) closer to users.

  • Reduces latency.

  • Example: CloudFront

    • If cache miss → fetches from server, stores in cache.

31. CloudFront and Anycast

  • Uses Anycast IP → same IP can reach nearest server.

  • User gets content from the closest location = faster delivery.

  • Saves bandwidth and improves user experience.

32. CDN Summary

  • Acts like a global cache layer.

  • Improves performance, reduces server load, and saves bandwidth.

Comments

Post a Comment

Popular posts from this blog

JavaScript Must-Read Topics for Senior Developer Interviews

-
JavaScript Must-Read Topics for Senior Developer Interviews Core JavaScript Variables and Scoping ( var , let , const ) ✅ Hoisting✅ Closures✅ The Event Loop✅ Promises✅ and Async/Await✅ Callbacks✅ Prototype and Prototypal Inheritance✅ this Keyword and Binding✅ Execution Context and Call Stack✅ Error Handling ( try-catch , finally , custom errors)✅ ES6+ Features (Destructuring✅, Spread/Rest✅, Arrow Functions✅, etc.) Type Coercion and Type Checking ✅ == vs ===  ✅ Object-Oriented Programming in JavaScript (Classes, Constructors)✅ Functional Programming Concepts (Higher-Order Functions✅, Pure Functions✅, Immutability✅) Module Systems ( require , import/export , CommonJS, ESModules)✅ JavaScript Design Patterns (Singleton, Factory, Observer, etc.) Garbage Collection✅ Advanced Topics Currying✅ Memoization✅ Throttling and Debouncing✅ Event Bubbling , Event Capturing and Event Delegation✅ Shadow DOM and Web Components Iterators, Iterables and Generator✅ WeakMap and WeakSet✅ Proxy and...
Read more

React.js Topics for a Mid-Senior Level Developer Interview Preparation

-
  React.js Topics for a Mid-Senior Level Developer Interview Preparation 1. Core Concepts React Component Types (Class vs Functional Components)✅ JSX and Rendering Elements✅ Props and Prop Validation✅ State Management (useState, class-based state)✅ Lifecycle Methods vs Hooks (useEffect)✅ 2. Advanced React Context API✅ Higher-Order Components (HOC)✅ Render Props✅ React Portals✅ Refs (createRef✅, useRef✅, forwardRef✅) Error Boundaries✅ Suspense and Lazy Loading✅ Strict Mode✅ Concurrent Rendering✅ 3. React Hooks Basic Hooks: useState, useEffect✅ Advanced Hooks: useContext✅, useReducer✅, useMemo✅, useCallback✅, useRef✅, useLayoutEffect✅ Custom Hooks (Creation and Use Cases)✅ 4. React Performance React.memo and useMemo✅ React Profiler Avoiding Re-renders (key, memoization, and immutability)✅ Code Splitting and Lazy Loading✅ Virtual DOM and Reconciliation✅ 5. Routing React Router (v6+): Route Configurations✅ Nested Routing✅ Dynamic Routing✅ Programmatic Navigation (useNavigate, useParams...
Read more

Recursive selection sort

-
Image
Recursive Selection sort : Program in C language #include<stdio.h> void selection_sort(int arr[],int start,int n) {     if(start>=n-1)     {         return;     }     int min_pos=start;     for(int i=start+1;i<n;i++)     {         if(arr[i]<arr[min_pos])         {             min_pos=i;         }     }     if(min_pos!=start)     {         int temp=arr[start];         arr[start]=arr[min_pos];         arr[min_pos]=temp;     }     selection_sort(arr,start+1,n); } void main() {     int n;     printf("Enter size of array:\n");     scanf("%d",&n);     int arr[n];     printf("Enter %d arra...
Read more