Unit 4 Suitable for applications like: 4.

ACK (from client)

🌐 Transport Layer Overview:-The Transport Layer is Layer 4 of the OSI model. It ensures Web (HTTP/HTTPS) UDP does not have connection management; it's stateless.
process-to-process communication — meaning, it allows applications on different hosts to Email (SMTP) 5. Flow Control and Retransmission
communicate. It provides services such as segmentation, flow control, error control, and
congestion control. File transfer (FTP) 🔹 Flow Control

1. Process-to-Process Delivery Key Features: Ensures that the sender does not overwhelm the receiver.

Unlike the network layer (which handles host-to-host delivery), the transport layer is Three-way handshake TCP uses a sliding window protocol:
responsible for delivering data to the correct process on the host. Sequencing and acknowledgment The sender can send multiple bytes before needing an acknowledgment.
Achieved using port numbers. Flow and congestion control Receiver advertises a window size (buffer size it can handle).
Example: HTTP uses port 80, HTTPS uses port 443. Retransmission of lost packets 🔹 Retransmission
Multiplexing and demultiplexing help handle multiple processes on a system. 3. Multiplexing and Demultiplexing If a packet is lost or corrupted, TCP retransmits it.
2. Transport Layer Protocols: UDP and TCP Multiplexing: Combining data from multiple applications at the sender and sending them Based on ACKs and timeouts.
over a single network connection.
🔹 UDP (User Datagram Protocol) Uses sequence numbers to keep track of segments.
Demultiplexing: Extracting data for specific applications at the receiver using port
Connectionless protocol. 6. Window Management
numbers.
Fast but unreliable – no error checking, flow control, or retransmission. TCP Window Size
Example:A user opens a web browser (port 80), email client (port 25), and a game (port
Suitable for time-sensitive applications like: 5000). All data share the same IP but use different port numbers to distinguish the Defines how many bytes can be sent without waiting for an acknowledgment.
DNS application.
Advertised window: Sent by receiver in ACKs.
Streaming (VoIP, video) 4. Connection Management
Helps in flow control and efficient transmission.
Online gaming In TCP, this includes:
Sliding Window Protocol
Key Features: 🔸Connection Establishment (Three-way Handshake):
Sender can send multiple frames before needing an ACK.
Simple header (8 bytes) 1. Client → Server: SYN
Receiver window slides forward as it receives data.
No guarantee of delivery 2. Server → Client: SYN-ACK
Benefits:
No order preservation 3. Client → Server: ACK
Better throughput
No congestion control 🔸 Connection Termination (Four steps):
Smooth data flow
🔹 TCP (Transmission Control Protocol) 1. FIN (from client)
7. TCP Congestion Control
Connection-oriented protocol. 2. ACK (from server)
TCP reduces transmission rate when the network is congested. Techniques include:
Reliable, ordered, and error-checked delivery of data. 3. FIN (from server)
🔹 Slow Start:

Starts with a small congestion window (cwnd), usually 1 MSS (Maximum Segment Size). Unit 5 HyperText Transfer Protocol (HTTP):

Doubles cwnd each RTT until threshold. 1. Domain Name System (DNS) HTTP is the application layer protocol used for communication between web browsers and
web servers.
🔹 Congestion Avoidance: The Domain Name System (DNS) is a distributed hierarchical system that translates
domain names (e.g., www.example.com) into IP addresses (e.g., 192.168.1.1), which are Key Features:-Stateless
After reaching threshold, increase cwnd linearly.
needed for locating and identifying services and devices on the Internet.
Text-based protocol
🔹 Fast Retransmit and Fast Recovery: DNS is essential because humans prefer to use readable domain names rather than
Uses TCP (port 80 for HTTP, 443 for HTTPS)
On receiving 3 duplicate ACKs, retransmit lost segment immediately (Fast Retransmit). numeric IP addresses. DNS operates at the application layer of the OSI model.
HTTP Methods:
Temporarily reduce cwnd but avoid going back to slow start (Fast Recovery). Key Components:-Domain Names: Organized hierarchically (e.g., root, top-level domains
like .com, .org). GET: Request a resource
TCP Congestion Phases:
DNS Server Types:- POST: Submit data to the server
1. Slow Start
Root DNS Servers PUT, DELETE, HEAD, etc.
2. Congestion Avoidance
Top-Level Domain (TLD) Servers HTTP Structure:
3. Fast Retransmit
Authoritative Name Servers Request: Method + URL + Headers + Optional body
4. Fast Recovery
Local DNS Resolver/Cache Response: Status code + Headers + Body (HTML, JSON, etc.)
8. Quality of Service (QoS)
Resolution Process: Versions:
Refers to the performance level of service provided to the applications or users.
1. User enters a URL. HTTP/1.1 (persistent connections)
Affects factors like latency, jitter, bandwidth, and packet loss.
2. Resolver checks local cache.
HTTP/2 (multiplexing)
Important for real-time applications (video/audio calls). 3. If not found, it queries root → TLD → authoritative server.
4. IP address is returned and used to contact the host. HTTP/3 (uses QUIC over UDP)
QoS Metrics:
Features: 3. Electronic Mail (Email)
Throughput – Amount of data transmitted per second
Distributed and fault-tolerant. Email is one of the oldest and most widely used applications on the Internet. It allows users
Latency – Delay between source and destination to send and receive messages over the network.
Uses UDP (port 53) for most queries, and TCP for large data.
Jitter – Variability in packet arrival time Architecture:
2. World Wide Web and HTTP (HyperText Transfer Protocol)
Packet Loss – Number of packets lost during transmission User Agent (UA): Interface for users (e.g., Outlook, Gmail)
World Wide Web (WWW):-The World Wide Web is a system of interlinked hypertext
QoS Techniques: Mail Transfer Agent (MTA): Transfers messages (e.g., Sendmail)
documents accessed via the Internet. These documents, or web pages, are identified using
Traffic Shaping Uniform Resource Locators (URLs) and viewed using web browsers.
Mail Server: Stores incoming/outgoing mail
Prioritization (e.g., VoIP over file transfer) Components:Web browsers (clients)
Email Protocols:
Resource Reservation Protocol (RSVP) Web servers SMTP (Simple Mail Transfer Protocol): Used for sending mail (port 25)
Differentiated Services (DiffServ) Protocols like HTTP/HTTPS
POP3 (Post Office Protocol v3): Download mail and delete from server (port 110)

IMAP (Internet Message Access Protocol): Access mail on the server (port 143) Uses TCP port 22 Lempel-Ziv-Welch (LZW)

Process: Encrypts all data, providing confidentiality and integrity Benefits:-Saves bandwidth

1. Sender composes and sends mail using SMTP. Supports file transfer (SCP, SFTP), port forwarding, and key-based authentication Reduces storage requirements
2. Mail is delivered to recipient’s server.
6. Network Management Speeds up data transfer
3. Recipient retrieves mail using POP3 or IMAP.
Network management includes the monitoring, configuring, and controlling of a computer Compression is especially important in multimedia transmission and cloud storage
network to ensure efficiency, security, and reliability. services.
4. File Transfer Protocol (FTP)
Key Functions: 8. Cryptography – Basic Concepts
FTP is used to transfer files between a client and a server on a network.
Fault Management: Detect and repair network issues Cryptography is the study and practice of techniques to secure communication in the
Features: presence of adversaries.
Configuration Management: Monitor network configurations
Uses TCP (port 21 for control, port 20 for data) Key Goals:
Performance Management: Monitor and improve network performance
Supports authentication (username/password) Confidentiality: Prevent unauthorized access
Security Management: Enforce security policies
Allows directory listing, upload, download, etc. Integrity: Ensure data is not altered
Accounting Management: Track resource usage
Modes: Authentication: Verify identity of users/devices
Protocol:
Active Mode: Server initiates data connection to client Non-repudiation: Prevent denial of actions
SNMP (Simple Network Management Protocol)
Passive Mode: Client initiates both control and data connections Types of Cryptography:
Uses UDP
Limitations: Symmetric Key Cryptography:
Components:
No encryption (in plain FTP) Same key is used for encryption and decryption.
Manager: Central controller
Secure alternatives include SFTP and FTPS Examples: AES, DES
Agent: Collects data
5. Remote Login (Telnet and SSH) Fast but key distribution is difficult.
MIB: Database of managed objects
Remote login allows a user to access and manage a remote computer as if logged in Asymmetric Key Cryptography:
7. Data Compression
locally.
Data compression reduces the size of data to save bandwidth and speed up transmission. Uses a public key for encryption and a private key for decryption.
Telnet:
Types: Examples: RSA, ECC
Uses TCP port 23
Lossless Compression: No data is lost (e.g., ZIP, PNG) More secure but slower.
Sends data in plain text
Lossy Compression: Some data is lost (e.g., JPEG, MP3) Common Applications:-SSL/TLS for secure web browsing
Allows command-line access to remote systems
Techniques:-Run-Length Encoding (RLE) Digital Signatures
SSH (Secure Shell):
Huffman Coding Encryption of Emails and Files
Secure alternative to Telnet
Unit 3 Subnetting:-Divides a network into smaller segments for better management and security. ICMP types include:-Type 0 (Echo Reply)
Subnet masks (e.g., /24) define how many bits are used for the network portion.
1. Point-to-Point Networks Type 3 (Destination Unreachable)
Importance:-Logical addresses allow communication across routers.
A point-to-point network is a direct communication link between two devices. It provides a Type 8 (Echo Request)
dedicated path for data transmission, with no intermediary devices between sender and Needed for routing decisions at the Network Layer.
Type 11 (Time Exceeded)
receiver.
Each device on a network must have a unique IP address.
4. Routing, Forwarding, and Delivery
Key Characteristics:
3. Basic Internetworking:-Internetworking involves connecting different networks into a) Routing-The process of finding the best path from source to destination.
Direct Connection: Only two endpoints (hosts or routers) are involved. a single large-scale network, such as the Internet. This requires multiple protocols
Uses routing tables built manually (static) or dynamically (via protocols).
and mechanisms:
Simplicity: Minimal configuration; the data travels without being routed through other
a) IP (Internet Protocol)-Primary protocol for the Network Layer. Happens at Layer 3 (Network Layer).
devices.
Responsibilities: Addressing, packet forwarding, and fragmentation. b) Forwarding-Action of sending a packet to the next hop based on the routing
Efficiency: Offers low latency and higher speed due to the absence of intermediate nodes.
table.
IPv4 vs. IPv6: IPv6 supports more devices and improved features.
Use Cases: Common in WAN links (e.g., leased lines between offices), serial links between
Performed by routers after examining destination IP.
routers, and in VPN tunnels. b) CIDR (Classless Inter-Domain Routing)-Allows flexible allocation of IP
addresses. c) Delivery-Ensuring the packet reaches its intended final host.
Advantages:-Easy to implement and manage.
CIDR notation: IP/prefix (e.g., 192.168.1.0/24). Two types:
Secure and private communication due to exclusivity.
Replaces class-based addressing to reduce IP wastage. Direct delivery: Both sender and receiver are on the same network.
Less susceptible to congestion.
c) ARP (Address Resolution Protocol)-Maps IP addresses to MAC addresses. Indirect delivery: Routers are used to reach remote networks.
Disadvantages:-Not scalable for larger networks.
Used in local area networks to find the MAC of the destination IP. 5. Static and Dynamic Routing
Wasted resources if the link is underused
a) Static Routing-Routes manually configured by a network administrator.
Stores results in the ARP cache for future use.
2. Logical Addressing:-Logical addressing is essential in identifying devices across
Suitable for small networks with simple topology.
different networks. Unlike physical (MAC) addresses that are hardcoded, logical d) RARP (Reverse ARP)-Resolves MAC addresses to IP addresses.
addresses can be assigned and changed. Advantages:-Simple and secure.
Used by diskless workstations to obtain an IP at boot.
IP Addressing: No extra CPU or bandwidth overhead.
Obsolete and replaced by DHCP.
IPv4: A 32-bit address divided into 4 octets (e.g., 192.168.1.1). Disadvantages:-Not adaptable to network changes.
e) DHCP (Dynamic Host Configuration Protocol)-Automatically assigns IP
Classes (A, B, C) were used traditionally. addresses to hosts. Maintenance is difficult in large networks.

Consists of network ID and host ID. Also provides subnet mask, default gateway, DNS. b) Dynamic Routing-Routers communicate using protocols to automatically
update routes.
IPv6: A 128-bit address represented in hexadecimal (e.g., Reduces administrative overhead in large networks.
2001:0db8:85a3:0000:0000:8a2e:0370:7334). Advantages:-Automatically adjusts to topology changes.
f) ICMP (Internet Control Message Protocol)-Used for diagnostics and error
Provides a vastly larger address space. reporting. Easier to manage in large networks.

Eliminates the need for NAT. Example: ping uses ICMP Echo Request/Reply. Disadvantages:-More complex.

May use bandwidth for protocol traffic. Allows bursts if enough tokens are available.

6. Routing Algorithms and Protocols More flexible than leaky bucket.

Routing Algorithms: c) RED (Random Early Detection)-Drops packets probabilistically before queues are
full.
a) Distance Vector Routing-Routers share entire routing tables with neighbors.
Helps prevent global synchronization and network collapse.
Metric: Number of hops.
d) TCP Congestion Control-Uses mechanisms like:
Example: RIP (Routing Information Protocol).
Slow Start
b) Link State Routing-Routers broadcast link status to all routers.
Congestion Avoidance
Builds a map of the network using Dijkstra’s algorithm.
Fast Retransmit
Example: OSPF (Open Shortest Path First).
Fast Recovery
Routing Protocols:
Adjusts window size dynamically.
a) RIP-Max 15 hops limit.
8. IPv6 (Internet Protocol version 6):-IPv6 addresses the limitations of IPv4,
Simple but slow to converge.
especially address exhaustion.
Uses UDP port 520.
Features:-128-bit address space (2^128 addresses).
b) OSPF-Hierarchical, fast-converging, uses areas.
Written in hexadecimal, separated by colons (e.g., fe80::1).
Suitable for large enterprise networks.
Header simplification: Fewer fields, making routers faster.
Uses Dijkstra’s algorithm for shortest path.
No need for NAT due to vast address space.
c) BGP (Border Gateway Protocol)-Used for routing between autonomous systems
Auto-configuration via SLAAC (Stateless Address Autoconfiguration).
(ASes) on the Internet.
IPSec is mandatory, offering built-in security.
Path vector protocol.
IPv6 Address Types:
Core protocol of the Internet.
Unicast: One-to-one.
7. Congestion Control Algorithms:-Congestion occurs when too many packets are
present in the network, causing packet loss, delay, or collapse. Multicast: One-to-many.

Key Algorithms: Anycast: One-to-nearest (in terms of routing).

a) Leaky Bucket-Enforces a fixed output rate. Transition Techniques:

Excess packets are discarded. Dual Stack: Devices run IPv4 and IPv6 simultaneously.

Smoothens bursty traffic. Tunneling: IPv6 packets are encapsulated in IPv4.

b) Token Bucket-Tokens are generated at a fixed rate. Translation: NAT64 translates IPv6 to IPv4 and vice versa.