Computer Networks

[Link] Interview Guide

Based on InterviewBit syllabus + real [Link] campus & off-campus experiences

OSI & TCP/IP · DNS & HTTP · TCP vs UDP · What happens when you type [Link]

Subnetting · Routing · Sockets · Security · CDN & Load Balancing

Computer Networks — [Link] Interview Guide InterviewBit Topics

Contents
1. OSI Model — All 7 Layers
Layer roles, devices, protocols per layer

2. TCP/IP Model
4-layer model, comparison with OSI

3. IP Addressing & Subnetting

IPv4/IPv6, classes, CIDR, NAT

4. DNS — How domain names resolve

■ Most asked at [Link]

5. HTTP & HTTPS

Methods, status codes, TLS handshake

6. TCP vs UDP
■ Standard [Link] question

7. What happens when you type [Link]

■ The #1 [Link] question — full flow

8. Routing & Switching

Routers, switches, ARP, routing tables

9. Socket Programming Concepts

Ports, sockets, 3-way handshake

10. Network Security Essentials

Firewalls, SSL/TLS, attacks

11. CDN, Load Balancing & Proxies

Relevant to [Link]'s ad-tech domain

12. Quick-Fire Q&A; Cheat Sheet

30 crisp answers for rapid revision

Prepared from InterviewBit + [Link] Campus Experiences Page 2

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 1 — OSI Model (7 Layers)

OSI (Open Systems Interconnection) is a conceptual framework that standardises how different network systems
communicate. Every [Link] interviewer expects you to know all 7 layers, their responsibilities, and the
associated protocols/devices.

Layer # Name Key Responsibility Protocols / Devices

7 Application User-facing services, data representation HTTP, FTP, SMTP, DNS, DHCP

6 Presentation Encryption, compression, format translation SSL/TLS, JPEG, ASCII, MPEG

5 Session Manages sessions (open/close/maintain) NetBIOS, RPC, PPTP

4 Transport End-to-end delivery, segmentation, error control

TCP, UDP — ports live here

3 Network Logical addressing, routing between networks

IP, ICMP, ARP, Routers

2 Data Link Node-to-node transfer, MAC addressing, framing

Ethernet, Switches, MAC

1 Physical Bits over physical medium Cables, Hubs, Wi-Fi signals

Mnemonics to remember layer order

Top→Bottom (7→1): All People Seem To Need Data Processing
Bottom→Top (1→7): Please Do Not Throw Sausage Pizza Away

Important Interview Q&A;

Q. At which OSI layer does a router operate?
Layer 3 (Network). Routers read IP addresses to forward packets between networks.

Switches operate at Layer 2 (Data Link) using MAC addresses.

Hubs are Layer 1 — they just broadcast bits to all ports.

Q. What is the difference between a hub, switch, and router?

Hub (L1): Broadcasts every packet to all ports. Dumb device, causes collisions.

Switch (L2): Learns MAC addresses and forwards frames only to the correct port.

Router (L3): Routes packets between different IP networks using a routing table.

Q. What is encapsulation in networking?

Each OSI layer adds its own header (and sometimes trailer) as data travels down the stack.

Application data → Segment (Transport adds port) → Packet (Network adds IP) → Frame (Data Link
adds MAC) → Bits (Physical).

De-encapsulation happens at the receiver in reverse.

Prepared from InterviewBit + [Link] Campus Experiences Page 3

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 2 — TCP/IP Model

The TCP/IP model is the practical model that the actual Internet uses. It has 4 layers and maps roughly onto the 7
OSI layers.

TCP/IP Layer Maps to OSI Layers Common Protocols

Application 5, 6, 7 HTTP, HTTPS, FTP, DNS, SMTP, SSH

Transport 4 TCP, UDP

Internet 3 IP, ICMP, ARP, RARP

Network Access 1, 2 Ethernet, Wi-Fi, MAC

OSI vs TCP/IP — Key Differences

Aspect OSI Model TCP/IP Model

Layers 7 4

Type Conceptual/reference Practical/implemented

Transport layer Guarantees delivery Does not guarantee

Session/Presentation Separate layers Merged into Application

Protocol Generic Specific (IP, TCP, UDP)

Prepared from InterviewBit + [Link] Campus Experiences Page 4

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 3 — IP Addressing & Subnetting

IPv4 Address Classes

Class First Octet Range Default Subnet Mask # of Networks # Hosts/Net Use

A 1 – 126 [Link] 126 16,777,214 Large enterprises

B 128 – 191 [Link] 16,384 65,534 Medium orgs

C 192 – 223 [Link] 2,097,152 254 Small networks

D 224 – 239 N/A — — Multicast

E 240 – 255 N/A — — Reserved/Research

Private IP Ranges (Non-routable on Internet)

Class Private Range CIDR

A [Link] – [Link] /8

B [Link] – [Link] /12

C [Link] – [Link] /16

Subnetting — CIDR Quick Formula

Given a /n prefix: Number of hosts = 2^(32-n) − 2 (subtract network & broadcast).

Example: [Link] /26

Hosts available = 2^(32-26) - 2 = 2^6 - 2 = 64 - 2 = 62
Network address : [Link]
Broadcast : [Link]
Usable range : [Link] → [Link]

NAT — Network Address Translation

Q. What is NAT and why is it used?

NAT translates private IP addresses to a public IP before packets leave a local network.

It allows many devices to share a single public IP address — crucial since IPv4 addresses are limited.

Types: Static NAT (1-to-1), Dynamic NAT (pool), PAT/Overloaded NAT (many-to-one using ports).

IPv6 — Key Points

• 128-bit address (vs 32-bit IPv4) — written as 8 groups of 4 hex digits.
• No broadcasts — replaced by multicast and anycast.
• Built-in IPSec support for security.
• No NAT needed — every device gets a globally unique address.
• Auto-configuration via SLAAC (Stateless Address Autoconfiguration).

Prepared from InterviewBit + [Link] Campus Experiences Page 5

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 4 — DNS (Domain Name System)

■ [Link] favourite: DNS resolution is asked in almost every [Link] interview. Know every step.

DNS translates human-readable domain names (e.g. [Link]) into IP addresses. It is a distributed,
hierarchical database.

DNS Resolution — Step by Step

Browser cache check
1 Browser checks if it already knows the IP. If yes, done.

OS/hosts file
2 Checks /etc/hosts (local override file) and OS DNS cache.

Recursive Resolver
3 Your ISP's DNS resolver (or [Link]) takes over. It does the heavy lifting.

Root Name Server

4 Resolver asks a Root Server — 13 sets globally. Returns address of TLD server.

TLD Name Server

5 Handles .com, .in, .org etc. Returns the authoritative name server for the domain.

Authoritative Name Server

6 Holds actual DNS records. Returns the final IP address.

Response cached
7 Resolver caches the result per TTL. Browser connects to the IP.

DNS Record Types

Record Purpose Example

A Maps hostname → IPv4 [Link] → [Link]

AAAA Maps hostname → IPv6 [Link] → 2607:f8b0:...

CNAME Alias for another domain www → [Link]

MX Mail server for domain [Link] → [Link]

NS Authoritative name server [Link] NS [Link]

TXT Arbitrary text (SPF, DKIM) v=spf1 include:...

PTR Reverse DNS (IP → name) 142.250.x.x → [Link]

■ TTL (Time To Live) controls how long a DNS record is cached. Low TTL = frequent updates, high TTL = faster
resolution.

Prepared from InterviewBit + [Link] Campus Experiences Page 6

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 5 — HTTP & HTTPS

HTTP (HyperText Transfer Protocol) is the foundation of data communication on the Web. HTTPS = HTTP + TLS
encryption.

HTTP Methods
Method Purpose Idempotent? Safe?

GET Retrieve a resource Yes Yes

POST Create a resource / submit data No No

PUT Replace entire resource Yes No

PATCH Partially update resource No No

DELETE Remove a resource Yes No

HEAD GET without body (check headers) Yes Yes

OPTIONS List allowed methods for a URL Yes Yes

HTTP Status Codes

Range Meaning Common Codes

1xx Informational 100 Continue, 101 Switching Protocols

2xx Success 200 OK, 201 Created, 204 No Content

3xx Redirection 301 Moved Permanently, 302 Found, 304 Not Modified

4xx Client Error 400 Bad Request, 401 Unauthorised, 403 Forbidden, 404 Not Found

5xx Server Error 500 Internal Server Error, 502 Bad Gateway, 503 Service Unavailable

TLS Handshake (HTTPS)

• 1. Client Hello — Browser sends supported TLS versions + cipher suites.
• 2. Server Hello — Server picks cipher, sends its SSL certificate (with public key).
• 3. Certificate Verification — Browser verifies cert against trusted CAs.
• 4. Key Exchange — Browser generates a pre-master secret, encrypts with server's public key, sends it.
• 5. Session Keys — Both sides derive the same symmetric session key independently.
• 6. Encrypted Communication — All further data is encrypted with the session key.

HTTP/1.1 vs HTTP/2 vs HTTP/3

Feature HTTP/1.1 HTTP/2 HTTP/3

Multiplexing No (one req/conn) Yes (streams) Yes (QUIC)

Header compression No HPACK QPACK

Server Push No Yes Yes

Transport TCP TCP UDP (QUIC)

Head-of-line blocking Yes Partially solved Solved

Prepared from InterviewBit + [Link] Campus Experiences Page 7

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 6 — TCP vs UDP

■ [Link] favourite: TCP vs UDP is asked in nearly every [Link] CN round. Know the 3-way
handshake cold.

Feature TCP UDP

Full form Transmission Control Protocol User Datagram Protocol

Connection Connection-oriented (3-way handshake) Connectionless

Reliability Guaranteed delivery + ACKs No guarantee

Order Ordered delivery No ordering

Speed Slower (overhead) Faster

Error checking Yes (retransmission) Yes (checksum only, no retry)

Flow control Yes (sliding window) No

Congestion control Yes No

Header size 20–60 bytes 8 bytes

Use cases HTTP, FTP, SSH, Email DNS, Video streaming, Gaming, VoIP

TCP 3-Way Handshake

Establishes a reliable connection before data transfer.

Client Server
|--- SYN (seq=x) -----------> | Step 1: Client requests connection
|<-- SYN-ACK (seq=y,ack=x+1) --| Step 2: Server acknowledges + sends its seq
|--- ACK (ack=y+1) ----------> | Step 3: Client acknowledges server
|======= DATA TRANSFER ======== | Connection established!

TCP 4-Way Termination

Client Server
|--- FIN -------------------> | I want to close
|<-- ACK -------------------- | Acknowledged
|<-- FIN -------------------- | Server also wants to close
|--- ACK -------------------> | Acknowledged → Connection closed

TCP Flow Control vs Congestion Control

Q. What is the difference between flow control and congestion control in TCP?
Flow Control prevents the sender from overwhelming the receiver. TCP uses a sliding window — the
receiver advertises how much buffer it has.

Congestion Control prevents the sender from overwhelming the network. Algorithms: Slow Start,
Congestion Avoidance, Fast Retransmit, Fast Recovery (AIMD principle).

Prepared from InterviewBit + [Link] Campus Experiences Page 8

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 7 — "What happens when you type

[Link]?"
■ [Link] favourite: This is THE most asked question at [Link]. Give a complete, layered answer.

URL Parsing
1 Browser parses the URL: scheme=https, host=[Link], path=/

HSTS Check
2 Browser checks its HSTS preload list — if [Link] is listed, it forces HTTPS.

DNS Resolution
Browser checks cache → OS cache → /etc/hosts → Recursive DNS resolver → Root NS → TLD
3
NS (.com) → Authoritative NS → IP returned (e.g. [Link]). Result cached per TTL.

ARP (if needed)

OS checks if the destination is on the same subnet. If not (it isn't), sends ARP to find the default
4
gateway's MAC address.

TCP Connection
5 OS initiates a TCP 3-way handshake with [Link]:443 (HTTPS port).

TLS Handshake
Client Hello → Server Hello + Certificate → Key Exchange → Session key derived → Encrypted
6
tunnel established.

HTTP Request
7 Browser sends: GET / HTTP/2 Host: [Link] (plus cookies, user-agent, etc.)

Server Processing
8 Google's load balancer receives request, routes to an available server. Server generates response.

HTTP Response
9 Server responds: 200 OK with HTML content, headers (Content-Type, Cache-Control, etc.).

Rendering
Browser parses HTML, discovers CSS/JS/image resources, fires additional requests. DOM +
10
CSSOM → Render Tree → Layout → Paint → Composite.

Connection Reuse
11 TCP connection stays open (Keep-Alive) for subsequent requests.

■ In the interview, walk through OSI layers as you describe each step — it shows depth. Mention: DNS (App
layer), TCP (Transport), IP routing (Network), Ethernet/ARP (Data Link).

Prepared from InterviewBit + [Link] Campus Experiences Page 9

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 8 — Routing & Switching

ARP — Address Resolution Protocol

Q. How does ARP work?

ARP resolves an IP address to a MAC address on a local network.

1. Host broadcasts: 'Who has IP [Link]? Tell [Link]'

2. The target host replies: 'I have that IP. My MAC is AA:BB:CC:DD:EE:FF'

3. Requester caches IP→MAC in its ARP table.

RARP does the reverse (MAC → IP). DHCP has now replaced RARP for address assignment.

Routing Protocols
Protocol Type Algorithm Metric Notes

RIP Distance Vector Bellman-Ford Hop count Max 15 hops; slow convergence

OSPF Link State Dijkstra (SPF) Cost (bandwidth) Fast convergence; scalable

BGP Path Vector Best path selection AS path, policies Internet backbone protocol

EIGRP Hybrid DUAL Bandwidth + delay Cisco proprietary

Static vs Dynamic Routing

• Static Routing: Routes manually configured by admin. No overhead, but doesn't adapt to failures.
• Dynamic Routing: Routes learned automatically via protocols (RIP, OSPF, BGP). Adapts to topology
changes.
• Default Route: [Link]/0 — used when no specific route matches (the 'catch-all' gateway).

DHCP — Dynamic Host Configuration Protocol

DHCP automatically assigns IP addresses. Process: DORA

D — Discover : Client broadcasts 'I need an IP!'

O — Offer : DHCP server offers an available IP
R — Request : Client requests the offered IP
A — Acknowledge: Server confirms; client gets IP, subnet, gateway, DNS

Prepared from InterviewBit + [Link] Campus Experiences Page 10

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 9 — Socket Programming Concepts

Q. What is a socket?
A socket is a combination of IP address + port number — it uniquely identifies a communication endpoint.

Every network connection has two sockets: one at each end (client socket and server socket).

Formally: socket = (IP address, port, protocol). Example: ([Link], 80, TCP)

Well-Known Ports
Port Protocol Service

20, 21 TCP FTP (data, control)

22 TCP SSH

23 TCP Telnet

25 TCP SMTP (email sending)

53 TCP/UDP DNS

67, 68 UDP DHCP (server, client)

80 TCP HTTP

110 TCP POP3

143 TCP IMAP

443 TCP HTTPS

3306 TCP MySQL

5432 TCP PostgreSQL

6379 TCP Redis

27017 TCP MongoDB

Ephemeral Ports
When a client initiates a connection, it is assigned a temporary high port (1024–65535) called an ephemeral port.
The server listens on its well-known port (e.g. 80), while the client uses its ephemeral port.

Prepared from InterviewBit + [Link] Campus Experiences Page 11

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 10 — Network Security Essentials

Firewall

Q. What is a firewall and how does it work?

A firewall monitors and filters incoming/outgoing network traffic based on predefined security rules.

Packet filtering: Checks IP/port headers. Stateless.

Stateful inspection: Tracks active connections — knows if a packet belongs to an established session.

Application-layer (WAF): Inspects HTTP content, blocks XSS, SQL injection etc.

Common Network Attacks

Attack How it works Defence

DDoS Flood server with traffic from many sources Rate limiting, CDN, scrubbing centres

Man-in-the-Middle Attacker intercepts communication between two parties

TLS/HTTPS, certificate pinning

DNS Spoofing / PoisoningAttacker injects fake DNS records DNSSEC, use trusted resolvers

ARP Spoofing Attacker sends fake ARP replies to link their MAC toDynamic
another IP
ARP Inspection (DAI)

SYN Flood Send many SYN packets without completing 3-way handshake
SYN cookies,
(half-open
rate limits
conns)

IP Spoofing Fake source IP in packet headers Ingress filtering (BCP38)

Replay Attack Re-send captured valid packets Timestamps, nonces, HTTPS

VPN — Virtual Private Network

• Encrypts traffic and tunnels it through a public network as if on a private network.
• Site-to-site VPN: Connects two office networks securely over the internet.
• Remote access VPN: Employee connects to company network from home.
• Protocols: IPSec, OpenVPN, WireGuard, L2TP.

Prepared from InterviewBit + [Link] Campus Experiences Page 12

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 11 — CDN, Load Balancing & Proxies

This chapter is especially relevant for [Link], which operates a global ad delivery network.

CDN — Content Delivery Network

Q. What is a CDN and how does it reduce latency?

A CDN is a geographically distributed network of servers (PoPs — Points of Presence) that cache
content close to users.

When a user requests content, DNS routes them to the nearest CDN edge server instead of the origin.

Cache hit: Content served from edge — very fast. Cache miss: Edge fetches from origin, caches it.

Benefits: Lower latency, reduced origin load, DDoS protection, better availability.

Examples: Cloudflare, Akamai, AWS CloudFront, Fastly.

Load Balancing Algorithms

Algorithm How it works Best for

Round Robin Requests distributed in rotation Equal-capacity servers

Weighted Round Robin Heavier servers get more requests Different-capacity servers

Least Connections Route to server with fewest active connections Variable request duration

IP Hash Same client IP always hits same server Session persistence

Random Random selection Simple stateless services

Forward Proxy vs Reverse Proxy

Q. What is the difference between a forward proxy and a reverse proxy?
Forward Proxy: Sits between client and the internet. Client sends all requests through it. Used for
anonymity, content filtering, bypassing geo-blocks. Client knows about the proxy.

Reverse Proxy: Sits in front of servers. Clients don't know they're hitting a proxy. Used for load
balancing, SSL termination, caching, security. Examples: Nginx, HAProxy.

Prepared from InterviewBit + [Link] Campus Experiences Page 13

Computer Networks — [Link] Interview Guide InterviewBit Topics

Chapter 12 — Quick-Fire Q&A; Cheat Sheet

30 crisp questions and answers for last-minute revision before your [Link] interview.

Q1. What is bandwidth vs throughput?

Bandwidth = max theoretical data rate (capacity of the pipe). Throughput = actual data transferred per
second (real-world). Throughput ≤ Bandwidth.

Q2. What is latency?

Time taken for a packet to travel from source to destination. Components: propagation + transmission +
processing + queuing delays.

Q3. What is a MAC address?

A 48-bit hardware address burned into a NIC. Written as 6 hex pairs: AA:BB:CC:DD:EE:FF. Unique per
device on a LAN segment.

Q4. Difference between TCP and IP?

IP handles logical addressing and routing (getting packets to the right machine). TCP handles reliable,
ordered delivery between processes (on top of IP).

Q5. What is a subnet mask?

Determines which part of an IP is the network portion vs host portion. [Link] means first 3 octets =
network, last = host.

Q6. What is ICMP?

Internet Control Message Protocol — used for diagnostics. ping uses ICMP Echo Request/Reply.
traceroute uses ICMP TTL Exceeded messages.

Q7. What is TTL in IP packets?

Time To Live — decremented by 1 at each router. When it hits 0, the packet is dropped and ICMP Time
Exceeded is sent back. Prevents infinite loops.

Q8. What is a broadcast address?

An address that targets all hosts on a subnet. For [Link]/24, broadcast = [Link]. Routers do
not forward broadcasts.

Q9. What is the difference between connection-oriented and connectionless?

Connection-oriented (TCP): establish → transfer → teardown. Reliable. Connectionless (UDP): just send.
No setup. Faster but unreliable.

Q10. What is congestion in a network?

When routers/links receive more traffic than they can handle — packets queue up, then get dropped. TCP
detects this via packet loss and slows down.

Prepared from InterviewBit + [Link] Campus Experiences Page 14

Computer Networks — [Link] Interview Guide InterviewBit Topics

Q11. What is HTTPS and how does it differ from HTTP?

HTTPS = HTTP over TLS. Data is encrypted in transit. TLS provides: confidentiality, integrity, and server
authentication via certificates.

Q12. What is a session in computer networks?

A semi-permanent interactive information interchange between two devices. OSI Layer 5 manages
session creation, maintenance, and termination.

Q13. What is SSL vs TLS?

SSL (Secure Sockets Layer) is the older, deprecated predecessor to TLS (Transport Layer Security). TLS
1.2 and 1.3 are current standards. People still say 'SSL' informally to mean TLS.

Q14. What is a ping?

A tool that sends ICMP Echo Requests to a host. If the host replies, network is reachable. Measures
round-trip time (RTT).

Q15. What is traceroute?

Traces the path packets take to a destination, listing each hop (router). Sends packets with incrementing
TTL values to reveal each router along the path.

Q16. What is the difference between IPv4 and IPv6?

IPv4: 32-bit, ~4.3B addresses. IPv6: 128-bit, virtually unlimited. IPv6 has no NAT, built-in IPSec, simpler
header, auto-configuration.

Q17. What are unicast, multicast, and broadcast?

Unicast = one-to-one. Multicast = one-to-many (group). Broadcast = one-to-all (subnet). IPv6 has no
broadcast — replaced by multicast.

Q18. What is network topology?

Physical/logical arrangement of nodes. Types: Bus (single cable), Star (central switch), Ring (loop), Mesh
(every node connected), Hybrid.

Q19. What is a VLAN?

Virtual LAN — logically segments a physical network into separate broadcast domains without needing
separate physical infrastructure.

Q20. What is a CDN?

Content Delivery Network — distributes cached content geographically so users are served from the
nearest edge server, reducing latency.

Q21. What is socket?

An endpoint for communication — identified by (IP address, port number, protocol). A connection has two
sockets, one at each end.

Prepared from InterviewBit + [Link] Campus Experiences Page 15

Computer Networks — [Link] Interview Guide InterviewBit Topics

Q22. What happens if two devices on a network have the same IP?

IP conflict! Both devices may stop working correctly. ARP will have conflicting entries, packets will go to
unpredictable destinations.

Q23. What is QoS?

Quality of Service — mechanisms to prioritise certain traffic types (e.g. voice over bulk downloads) to
reduce latency/jitter for time-sensitive apps.

Q24. What is a half-duplex vs full-duplex connection?

Half-duplex: can send OR receive at a time (walkie-talkie). Full-duplex: can send AND receive
simultaneously (phone call).

Q25. What is the difference between HTTP GET and POST?

GET retrieves data — params in URL, cached, bookmarkable, idempotent. POST sends data in request
body — not cached, for mutations, not idempotent.

Q26. What is SMTP?

Simple Mail Transfer Protocol — used to send emails (port 25/587). For receiving, IMAP (port 143) or
POP3 (port 110) is used.

Q27. What is the difference between a hub and a switch?

Hub (Layer 1) broadcasts packets to all ports. Switch (Layer 2) learns MAC addresses and sends frames
only to the correct port — more efficient.

Q28. What is a default gateway?

The router that a host sends traffic to when the destination is outside its local subnet. Typically the first
address in a subnet: [Link].

Q29. What is packet switching vs circuit switching?

Circuit switching: dedicated path reserved (old telephone). Packet switching: data split into packets that
independently route through the network (internet). More efficient.

Q30. What is the difference between a Layer 2 and Layer 3 switch?

L2 switch: forwards frames based on MAC addresses. L3 switch: can route packets based on IP
addresses — acts as both switch and basic router.

Prepared from InterviewBit + [Link] Campus Experiences Page 16