Advanced Computer Networks

Unit – I
Network Architecture
Network architecture refers to the design of a computer network,
encompassing the physical components, communication protocols, and
configuration that facilitate data transfer and communication between devices.
 Components: Includes devices like routers, switches, servers, and clients,
as well as the transmission media (e.g., fiber-optic cables, wireless).
 Layers: Networks are typically organized into layers (e.g., OSI model,
TCP/IP model), with each layer having specific functions such as data
link control, routing, and application communication.
Key Aspects of Network Architecture:
1. Topology: The physical or logical layout of devices in a network (e.g.,
star, bus, mesh, ring).
2. Protocols: Standardized rules governing data exchange (e.g., TCP/IP,
HTTP, Ethernet).
3. Switching and Routing: Mechanisms for determining the path that data
takes through a network.

Performance: Bandwidth and Latency

 Bandwidth: The maximum rate at which data can be transmitted over a
network. It is usually measured in bits per second (bps).
o High bandwidth allows for faster data transfer, reducing
bottlenecks in network traffic.
o Examples: Gigabit Ethernet supports up to 1 Gbps, while fiber-
optic networks can support multiple terabits per second.
 Latency: The time it takes for a data packet to travel from the sender to
the receiver. It is measured in milliseconds (ms).
o Low latency is essential for real-time communication (e.g., VoIP,
online gaming).
 o Latency is influenced by factors like the distance between devices,
network congestion, and the number of hops (i.e., routers/switches)
the data must pass through.
Key Performance Metrics:
1. Throughput: Actual data transfer rate, which can be lower than the
theoretical bandwidth due to congestion or inefficiencies.
2. Packet Loss: Occurs when packets are dropped due to network
congestion or errors.
3. Jitter: Variation in packet arrival times, which is critical for time-
sensitive applications like video conferencing.

High-Speed Networks
High-speed networks are designed to handle large volumes of data at faster
speeds, typically employing advanced technologies to minimize delays and
improve performance.
 Fiber-Optic Networks: Use light signals to transmit data, enabling high-
speed, long-distance communication. Fiber-optic cables offer much
higher bandwidth and lower latency compared to copper cables.
 5G Networks: Offer ultra-low latency and high bandwidth, suitable for
applications such as autonomous vehicles, augmented reality (AR), and
the Internet of Things (IoT).
 Data Centers: High-speed networks within data centers use high-
bandwidth connections (e.g., InfiniBand) to ensure efficient
communication between servers.

Network-Centric View
A network-centric view focuses on the role of the network as the backbone of
modern communication. It considers how devices interact over the network,
how data flows, and the essential role of protocols and infrastructure in ensuring
seamless communication across systems.
 Cloud Computing: Relies on high-performance networks to connect
users to data and services hosted remotely in data centers.
  Internet of Things (IoT): Network-centric thinking is critical in enabling
the massive number of connected devices to communicate and share data
efficiently.

Error Detection and Reliable Transmission

Error Detection ensures that data transmitted over a network is accurate and
not corrupted. Common techniques include:
1. Checksums: Simple error-checking method that adds a value to the data,
allowing the receiver to verify its integrity.
2. Cyclic Redundancy Check (CRC): A more robust error-detection
method that uses polynomial division for error-checking.
3. Parity Bits: A bit added to a data word to make the number of ones either
even or odd, used for simple error detection.
Reliable Transmission is a mechanism that ensures data is delivered without
loss, duplication, or errors. It involves:
1. Acknowledgments: The receiver confirms receipt of data.
2. Retransmission: If the acknowledgment is not received within a
specified time, the data is resent.
3. Flow Control: Manages the rate of data transmission to prevent
congestion and packet loss.
Protocols like TCP (Transmission Control Protocol) implement reliable
transmission by using sequence numbers, acknowledgments, and
retransmissions.

Ethernet and Multiple Access Networks

Ethernet is the most common local area network (LAN) technology, used for
connecting devices within a building or campus. It operates using packet
switching and is typically based on IEEE 802.3 standards.
 Access Method: Ethernet traditionally uses Carrier Sense Multiple
Access with Collision Detection (CSMA/CD), where devices listen for
traffic before transmitting to avoid collisions.
o Modern Ethernet networks, especially over switches, do not rely
on CSMA/CD as much, since collisions are rare.
Multiple Access Networks are networks where multiple devices share the same
communication medium (e.g., wireless channels, bus topology). They employ
mechanisms to prevent collisions or ensure fair access to the medium, such as:
1. CSMA/CD (Ethernet).
2. TDMA (Time Division Multiple Access) for cellular networks.
3. FDMA (Frequency Division Multiple Access) for radio communications.
4. CDMA (Code Division Multiple Access) for mobile communications.

Overlay Networks
An overlay network is a virtual network built on top of another network
(usually the Internet). It is designed to provide additional services such as
routing, content distribution, or security.
Routing Overlays:
 Overlay networks can be used to implement custom routing schemes that
are optimized for specific tasks, like content delivery, security, or network
resilience. Examples include VPNs (Virtual Private Networks), Tor, or
content distribution networks (CDNs).
Peer-to-Peer Networks:
 In a P2P network, each device (peer) can act as both a client and a server,
sharing resources (e.g., files, processing power) directly with other
devices, without relying on a centralized server.
o Examples: BitTorrent (file sharing), Skype (communications),
and Blockchain networks (cryptocurrency).
o Advantages: Decentralization, fault tolerance, scalability.
Content Distribution Networks (CDNs):
 CDNs are specialized overlay networks that distribute content (e.g.,
videos, images, websites) to users from locations geographically closer to
them, reducing latency and improving access speeds.
o Example: Akamai, Cloudflare.

Client-Server Networks
In a client-server network, clients (e.g., computers, mobile devices) request
services from centralized servers (e.g., web servers, databases, application
servers).
1. Characteristics:
o Centralized Control: Servers manage and control data, user
access, and resources.
o Scalable: More clients can be added easily without changing the
server architecture.
o Reliability: Servers can be configured for high availability and
fault tolerance (e.g., load balancing, redundancy).
2. Common Protocols:
o HTTP (Hypertext Transfer Protocol) for web-based services.
o FTP (File Transfer Protocol) for file sharing.
o SQL for database access.

Delay-Tolerant Networks (DTNs)

Delay-Tolerant Networks are designed to operate in environments where
traditional networking protocols (like TCP/IP) may not be practical due to
intermittent connectivity or high latency. These networks store data and forward
it when a connection becomes available.
1. Use Cases:
o Space Networks: Communication between Earth and spacecraft,
where communication is intermittent and has high delay.
o Disaster Recovery: In areas with poor infrastructure, where
connections are unreliable.
o Rural Connectivity: For low-cost, low-reliability connectivity in
remote areas.
2. Key Features:
o Store-and-Forward: Nodes temporarily store data and forward it
when possible.
o Bundle Protocol: A protocol used in DTNs to handle data transfer
across networks with intermittent connectivity.
 o Buffering: Nodes buffer data until a viable communication path is
available.

Summary of Key Concepts

 Network Architecture: Involves the design, components, and protocols
that enable data communication.
 Bandwidth and Latency: Key performance metrics for data
transmission.
 High-Speed Networks: Networks designed to support high-volume, low-
latency data transmission.
 Network-Centric View: Focuses on the network's role in modern
systems, such as cloud computing and IoT.
 Error Detection and Reliable Transmission: Ensures the accuracy and
integrity of data in transit.
 Ethernet and Multiple Access Networks: Protocols and techniques for
sharing network resources among multiple devices.
 Overlay Networks: Virtual networks designed to add services like
routing and content distribution.
 Client-Server Networks: A model where clients request services from
centralized servers.
 Delay-Tolerant Networks: Networks designed for environments with
intermittent connectivity or high latency.

Unit – II
Switching: Types of Networks
Switching refers to the process by which data is routed through a
network. The method used for switching determines how data is
transmitted between devices and how the network handles traffic. The
four main types of switching technologies are:

1. Circuit-Switched Networks
 A Circuit-Switched Network establishes a dedicated communication
path between two endpoints for the duration of the communication
session. Once the circuit is established, it remains open until the session
ends.
 Example: Traditional telephone networks (PSTN).
 Characteristics:
o Dedicated Path: A fixed communication path is established
between sender and receiver.
o Fixed Bandwidth: The bandwidth is reserved for the entire
duration of the call, whether or not data is being transmitted.
o Low Latency: Once established, communication happens with
minimal delay.
o Inefficiency: Resources are wasted when no data is being
transmitted.
Advantages:
 Predictable quality of service (QoS).
 Suitable for voice calls, where continuous, reliable communication is
needed.
Disadvantages:
 Not efficient for bursty data traffic.
 Requires more infrastructure and setup time for each call.

2. Datagram Networks
In a Datagram Network, each packet (datagram) is treated
independently, and there is no need to establish a dedicated path. Each
datagram can take a different route to the destination.
 Example: The Internet (uses IP - Internet Protocol).
 Characteristics:
o Connectionless: No prior setup or dedicated path is needed for
communication.
 o Packet Routing: Each packet can be routed independently, and
may take different routes through the network.
o Best-Effort Delivery: No guarantees for delivery or ordering of
packets.
Advantages:
 Efficient use of network resources.
 Scalability: Can handle a large number of devices with no need for
dedicated paths.
 Flexible routing: Data can find alternate routes if one path fails.
Disadvantages:
 Unreliable delivery: Packets may be lost, duplicated, or arrive out of
order.
 No guaranteed QoS, making it unsuitable for real-time services (like
voice or video).

3. Virtual-Circuit Networks
A Virtual-Circuit Network combines aspects of both circuit-switching
and datagram switching. It establishes a logical path (virtual circuit)
between sender and receiver before data transfer begins. However, data is
transmitted in packets like in a datagram network.
 Example: Frame Relay, ATM (Asynchronous Transfer Mode), and
MPLS (Multiprotocol Label Switching).
 Characteristics:
o Connection-Oriented: A virtual circuit is established between
devices before communication begins, and packets follow this
predefined path.
o Routing Information: Each packet contains routing information
that directs it along the established path.
o Stateful: Each intermediate router or switch maintains state
information about the virtual circuit.
Advantages:
 More reliable than datagram networks due to established routes.
 Allows for QoS guarantees and better resource management.
 Flexible: The virtual circuit can handle errors and congestion more
effectively than circuit-switching.
Disadvantages:
 Requires setup time to establish the virtual circuit.
 Less flexible than datagram networks because of the fixed path.

4. Message-Switched Networks
In a Message-Switched Network, the entire message is transmitted as a
single unit, with no segmentation into smaller packets. The network
routes the message through various nodes, and the message is stored at
each node until it can be forwarded to the next node.
 Example: Telegraph systems and some store-and-forward email
systems.
 Characteristics:
o Store-and-Forward: Messages are stored at intermediate nodes
until they can be forwarded to the next hop.
o No Fixed Path: Like datagram networks, there’s no fixed path
between sender and receiver.
o Message Delivery: Messages may experience delays, as they are
temporarily stored before being forwarded.
Advantages:
 Ideal for environments where real-time transmission is not critical (e.g.,
email).
 Allows for intermittent connectivity.
Disadvantages:
 Slow delivery due to store-and-forward operation.
 Potential delays in message delivery, which makes it unsuitable for real-
time applications.

5. Asynchronous Transfer Mode (ATM)

 ATM is a high-speed, connection-oriented switching technology that
operates at the data link layer and supports both voice and data services.
It transmits data in fixed-size cells (53 bytes), providing uniform size for
all types of traffic.
 Evolution: ATM evolved from the need for an integrated service network
that could handle voice, video, and data traffic over a single
infrastructure. It was designed to provide high-performance, low-latency
communication across wide-area networks (WANs).
 Benefits:
o Cell-based: ATM uses small, fixed-size cells, which simplifies
switching and ensures fast, deterministic data transmission.
o Quality of Service (QoS): Supports guaranteed QoS, making it
suitable for real-time applications like voice and video.
o Scalability: Can scale efficiently to handle large amounts of data
and diverse applications (e.g., broadband multimedia).
o Multimedia Support: Can simultaneously support voice, video,
and data, offering a versatile network solution.
 Concepts:
o Cell Structure: ATM uses 53-byte cells, with 5 bytes for header
and 48 bytes for payload.
o Virtual Circuits: Like virtual-circuit networks, ATM establishes
logical connections before data transmission.
o Traffic Classes: ATM supports different types of traffic (e.g.,
constant bit rate for voice, variable bit rate for data).

Broadband Integrated Services Digital Network (B-ISDN)

B-ISDN is a broadband network service designed to provide high-speed
data, voice, and video communication over a single network. It is part of
the Integrated Services Digital Network (ISDN) concept but supports
broadband, as opposed to the narrower-band services in traditional ISDN.
 Evolution: B-ISDN evolved from the need for more efficient, higher-
capacity systems capable of handling modern multimedia traffic.
 Benefits:
 o High-Speed Data Transmission: Supports high-bandwidth
applications such as video conferencing, high-definition TV, and
large data transfers.
o End-to-End Digital Transmission: Offers fully digital
transmission from the source to the destination, minimizing
distortion and signal degradation.
o Flexibility: Supports a variety of services (voice, video, data) using
ATM as the core transport technology.
 Concepts:
o Broadband: Refers to high-speed communication that supports a
wide range of services.
o ATM for Transport: ATM is used as the underlying technology
for B-ISDN to ensure fast, efficient data transfer.

Layer and Adaptation Layer

The Adaptation Layer is a crucial component of B-ISDN and ATM
networks. It sits between the higher layers (such as the network layer) and
the ATM layer, providing the necessary functionality to adapt the
communication to the needs of different applications.
 Function:
o Multiplexing: The adaptation layer handles the conversion and
multiplexing of different types of data (e.g., voice, video) into ATM
cells.
o Traffic Management: Manages the flow of data to ensure the
appropriate quality of service.
o Segmentation and Reassembly: Breaks larger messages into
smaller ATM cells and reassembles them at the destination.
 ATM Layers:
o ATM Layer: Deals with the transmission of fixed-size cells.
o Adaptation Layer: Transforms data into a format that can be
handled by ATM, ensuring that different types of traffic (e.g.,
video, voice) can be transmitted over the same network.
 Summary of Key Switching Concepts
 Circuit-Switched Networks: Dedicated paths for communication,
commonly used in traditional telephone networks.
 Datagram Networks: Connectionless, with each packet routed
independently (e.g., Internet).
 Virtual-Circuit Networks: Logical path setup between sender and
receiver, suitable for applications requiring reliability and QoS (e.g.,
Frame Relay).
 Message-Switched Networks: Store-and-forward systems, often used in
messaging systems.
 ATM: High-speed, connection-oriented technology designed for
broadband multimedia services, using small fixed-size cells for efficient
data transmission.
 B-ISDN: Broadband version of ISDN that supports integrated services
(voice, video, data) over ATM-based networks.
 Adaptation Layer: Provides functionality for multiplexing, traffic
management, and converting different data types to be transmitted over
ATM.
These switching methods and technologies are designed to handle
different types of traffic (real-time voice, video, and data) and cater to
different application needs in modern communication networks.

Unit – III

Switching: Types of Networks

Switching refers to the process by which data is routed through a
network. The method used for switching determines how data is
transmitted between devices and how the network handles traffic. The
four main types of switching technologies are:

1. Circuit-Switched Networks
 A Circuit-Switched Network establishes a dedicated communication
path between two endpoints for the duration of the communication
session. Once the circuit is established, it remains open until the session
ends.
 Example: Traditional telephone networks (PSTN).
 Characteristics:
o Dedicated Path: A fixed communication path is established
between sender and receiver.
o Fixed Bandwidth: The bandwidth is reserved for the entire
duration of the call, whether or not data is being transmitted.
o Low Latency: Once established, communication happens with
minimal delay.
o Inefficiency: Resources are wasted when no data is being
transmitted.
Advantages:
 Predictable quality of service (QoS).
 Suitable for voice calls, where continuous, reliable communication is
needed.
Disadvantages:
 Not efficient for bursty data traffic.
 Requires more infrastructure and setup time for each call.

2. Datagram Networks
In a Datagram Network, each packet (datagram) is treated
independently, and there is no need to establish a dedicated path. Each
datagram can take a different route to the destination.
 Example: The Internet (uses IP - Internet Protocol).
 Characteristics:
o Connectionless: No prior setup or dedicated path is needed for
communication.
 o Packet Routing: Each packet can be routed independently, and
may take different routes through the network.
o Best-Effort Delivery: No guarantees for delivery or ordering of
packets.
Advantages:
 Efficient use of network resources.
 Scalability: Can handle a large number of devices with no need for
dedicated paths.
 Flexible routing: Data can find alternate routes if one path fails.
Disadvantages:
 Unreliable delivery: Packets may be lost, duplicated, or arrive out of
order.
 No guaranteed QoS, making it unsuitable for real-time services (like
voice or video).

3. Virtual-Circuit Networks
A Virtual-Circuit Network combines aspects of both circuit-switching
and datagram switching. It establishes a logical path (virtual circuit)
between sender and receiver before data transfer begins. However, data is
transmitted in packets like in a datagram network.
 Example: Frame Relay, ATM (Asynchronous Transfer Mode), and
MPLS (Multiprotocol Label Switching).
 Characteristics:
o Connection-Oriented: A virtual circuit is established between
devices before communication begins, and packets follow this
predefined path.
o Routing Information: Each packet contains routing information
that directs it along the established path.
o Stateful: Each intermediate router or switch maintains state
information about the virtual circuit.
Advantages:
 More reliable than datagram networks due to established routes.
 Allows for QoS guarantees and better resource management.
 Flexible: The virtual circuit can handle errors and congestion more
effectively than circuit-switching.
Disadvantages:
 Requires setup time to establish the virtual circuit.
 Less flexible than datagram networks because of the fixed path.

4. Message-Switched Networks
In a Message-Switched Network, the entire message is transmitted as a
single unit, with no segmentation into smaller packets. The network
routes the message through various nodes, and the message is stored at
each node until it can be forwarded to the next node.
 Example: Telegraph systems and some store-and-forward email
systems.
 Characteristics:
o Store-and-Forward: Messages are stored at intermediate nodes
until they can be forwarded to the next hop.
o No Fixed Path: Like datagram networks, there’s no fixed path
between sender and receiver.
o Message Delivery: Messages may experience delays, as they are
temporarily stored before being forwarded.
Advantages:
 Ideal for environments where real-time transmission is not critical (e.g.,
email).
 Allows for intermittent connectivity.
Disadvantages:
 Slow delivery due to store-and-forward operation.
 Potential delays in message delivery, which makes it unsuitable for real-
time applications.

5. Asynchronous Transfer Mode (ATM)

 ATM is a high-speed, connection-oriented switching technology that
operates at the data link layer and supports both voice and data services.
It transmits data in fixed-size cells (53 bytes), providing uniform size for
all types of traffic.
 Evolution: ATM evolved from the need for an integrated service network
that could handle voice, video, and data traffic over a single
infrastructure. It was designed to provide high-performance, low-latency
communication across wide-area networks (WANs).
 Benefits:
o Cell-based: ATM uses small, fixed-size cells, which simplifies
switching and ensures fast, deterministic data transmission.
o Quality of Service (QoS): Supports guaranteed QoS, making it
suitable for real-time applications like voice and video.
o Scalability: Can scale efficiently to handle large amounts of data
and diverse applications (e.g., broadband multimedia).
o Multimedia Support: Can simultaneously support voice, video,
and data, offering a versatile network solution.
 Concepts:
o Cell Structure: ATM uses 53-byte cells, with 5 bytes for header
and 48 bytes for payload.
o Virtual Circuits: Like virtual-circuit networks, ATM establishes
logical connections before data transmission.
o Traffic Classes: ATM supports different types of traffic (e.g.,
constant bit rate for voice, variable bit rate for data).

Broadband Integrated Services Digital Network (B-ISDN)

B-ISDN is a broadband network service designed to provide high-speed
data, voice, and video communication over a single network. It is part of
the Integrated Services Digital Network (ISDN) concept but supports
broadband, as opposed to the narrower-band services in traditional ISDN.
 Evolution: B-ISDN evolved from the need for more efficient, higher-
capacity systems capable of handling modern multimedia traffic.
 Benefits:
 o High-Speed Data Transmission: Supports high-bandwidth
applications such as video conferencing, high-definition TV, and
large data transfers.
o End-to-End Digital Transmission: Offers fully digital
transmission from the source to the destination, minimizing
distortion and signal degradation.
o Flexibility: Supports a variety of services (voice, video, data) using
ATM as the core transport technology.
 Concepts:
o Broadband: Refers to high-speed communication that supports a
wide range of services.
o ATM for Transport: ATM is used as the underlying technology
for B-ISDN to ensure fast, efficient data transfer.

Layer and Adaptation Layer

The Adaptation Layer is a crucial component of B-ISDN and ATM
networks. It sits between the higher layers (such as the network layer) and
the ATM layer, providing the necessary functionality to adapt the
communication to the needs of different applications.
 Function:
o Multiplexing: The adaptation layer handles the conversion and
multiplexing of different types of data (e.g., voice, video) into ATM
cells.
o Traffic Management: Manages the flow of data to ensure the
appropriate quality of service.
o Segmentation and Reassembly: Breaks larger messages into
smaller ATM cells and reassembles them at the destination.
 ATM Layers:
o ATM Layer: Deals with the transmission of fixed-size cells.
o Adaptation Layer: Transforms data into a format that can be
handled by ATM, ensuring that different types of traffic (e.g.,
video, voice) can be transmitted over the same network.
 Summary of Key Switching Concepts
 Circuit-Switched Networks: Dedicated paths for communication,
commonly used in traditional telephone networks.
 Datagram Networks: Connectionless, with each packet routed
independently (e.g., Internet).
 Virtual-Circuit Networks: Logical path setup between sender and
receiver, suitable for applications requiring reliability and QoS (e.g.,
Frame Relay).
 Message-Switched Networks: Store-and-forward systems, often used in
messaging systems.
 ATM: High-speed, connection-oriented technology designed for
broadband multimedia services, using small fixed-size cells for efficient
data transmission.
 B-ISDN: Broadband version of ISDN that supports integrated services
(voice, video, data) over ATM-based networks.
 Adaptation Layer: Provides functionality for multiplexing, traffic
management, and converting different data types to be transmitted over
ATM.
These switching methods and technologies are designed to handle
different types of traffic (real-time voice, video, and data) and cater to
different application needs in modern communication networks.
Would you like to dive deeper into any specific switching technology, or
perhaps see how ATM works in practice?
4o mini
You said:
IPv4: Address Space, Notations, Classful, Classless, Network Address
Translation, Datagram, Fragmentation and Checksum IPv6 Addresses:
Structure, Address Space, Packet Format and Extension Headers, ICMP,
IGMP, ARP, RARP, Congestion Control and Resource Allocation:
Problem, Issues, Queuing, TCP?
ChatGPT said:
ChatGPT
IPv4: Address Space, Notations, Classful, Classless, and More
 IPv4 Address Space
IPv4 (Internet Protocol version 4) uses a 32-bit address space, which
allows for a total of 2^32 (4,294,967,296) unique addresses. These
addresses are typically written in dotted decimal notation, which breaks
the 32-bit address into four 8-bit segments (called octets), each
represented by a decimal number between 0 and 255.
 Example: 192.168.1.1
Each IPv4 address consists of two parts:
1. Network portion: Identifies the specific network.
2. Host portion: Identifies the device within that network.

IPv4 Notations
 Dotted Decimal Notation: The most common representation of IPv4
addresses. For example, 192.168.0.1.
 CIDR Notation (Classless Inter-Domain Routing): A more efficient
way of representing network addresses by appending a slash (/) followed
by the number of bits used for the network portion. For example,
192.168.0.0/24.

Classful Addressing
IPv4 addresses were historically divided into classes to simplify routing
and network organization:
1. Class A: Addresses from 0.0.0.0 to 127.255.255.255. Supports 16 million
hosts per network. The first bit of Class A addresses is always 0.
o Default Subnet Mask: 255.0.0.0
2. Class B: Addresses from 128.0.0.0 to 191.255.255.255. Supports 65,000
hosts per network. The first two bits are 10.
o Default Subnet Mask: 255.255.0.0
3. Class C: Addresses from 192.0.0.0 to 223.255.255.255. Supports 254
hosts per network. The first three bits are 110.
o Default Subnet Mask: 255.255.255.0
4. Class D: Addresses from 224.0.0.0 to 239.255.255.255. Reserved for
multicast communication.
5. Class E: Addresses from 240.0.0.0 to 255.255.255.255. Reserved for
future use.

Classless Addressing (CIDR)

Classless Inter-Domain Routing (CIDR) replaced the rigid classful
system and allows for more flexible address allocation, helping to reduce
the depletion of IPv4 addresses. CIDR notation specifies an IP address
followed by a slash (/) and the number of bits in the network portion
(subnet mask).
 Example: 192.168.1.0/24, where 24 indicates the number of bits in the
network portion.
CIDR also helps to subnet networks into smaller parts (subnets) without
worrying about the traditional class boundaries.

Network Address Translation (NAT)

Network Address Translation (NAT) is used to map private IP
addresses to a public IP address. This is common in home and corporate
networks, where devices use private IP addresses (e.g., 192.168.0.0/16)
internally but share a single public IP address for external
communication.
 Types of NAT:
o Static NAT: A one-to-one mapping between a private IP address
and a public IP address.
o Dynamic NAT: A pool of public IP addresses is used and mapped
dynamically to private IP addresses.
o PAT (Port Address Translation): Multiple devices can share a
single public IP address, differentiated by port numbers (often
referred to as "overloading").
Benefits of NAT:
 Provides security by hiding internal IP addresses.
 Conserves public IP addresses.
 IPv4 Datagram, Fragmentation, and Checksum
IPv4 Datagram
An IPv4 Datagram consists of:
1. Header: Contains routing information, such as the source and destination
addresses, protocol, and checksum.
2. Payload: The actual data being transmitted.
Header Structure (fixed fields):
 Version: Indicates the IP version (IPv4 or IPv6).
 Header Length: The length of the header.
 Type of Service (ToS): Quality of service information (used for
prioritization).
 Total Length: Total length of the datagram, including the header and
payload.
 Identification: Identifies fragmented packets.
 Flags and Fragment Offset: Used for packet fragmentation and
reassembly.
 Time-to-Live (TTL): Limits the packet's lifetime to avoid infinite loops.
 Protocol: Specifies the higher-layer protocol (e.g., TCP, UDP).
 Header Checksum: A simple error-checking mechanism to ensure data
integrity.
Fragmentation and Reassembly
 Fragmentation occurs when a datagram is too large to be transmitted
over a network segment with a smaller Maximum Transmission Unit
(MTU).
 The IPv4 header includes fragmentation fields (e.g., Flags, Fragment
Offset) to allow the original datagram to be reassembled at the
destination.
 Reassembly occurs at the destination by using the fragment identification
and offset to reconstruct the original packet.
Checksum
 The IPv4 checksum is used for error detection in the header. It checks
whether the data in the IPv4 header is intact during transmission. The
checksum is recalculated at each hop.

IPv6: Addresses and Packet Format

IPv6 Address Space
IPv6 uses a 128-bit address space, providing a massive address pool of
2^128 addresses, far more than IPv4's 32-bit space. This helps address
the issue of address depletion in IPv4.
 IPv6 Address Representation: Written as eight groups of four
hexadecimal digits, separated by colons. Leading zeros in each group can
be omitted.
Example: 2001:0db8:85a3:0000:0000:8a2e:0370:7334 can be shortened
to 2001:db8:85a3::8a2e:370:7334.
IPv6 Packet Format
An IPv6 packet consists of:
 Version: Always 6 for IPv6.
 Traffic Class: Similar to Type of Service (ToS) in IPv4, for quality of
service.
 Flow Label: Used for special handling of packets belonging to the same
flow.
 Payload Length: Length of the data (excluding the header).
 Next Header: Identifies the type of the next header (e.g., TCP, UDP).
 Hop Limit: Similar to TTL in IPv4, specifies the maximum number of
hops.
 Source Address: The IPv6 address of the sender.
 Destination Address: The IPv6 address of the receiver.

IPv6 Extension Headers

 IPv6 allows extension headers, which provide additional functionality
not required for basic packet routing. These headers are placed after the
main IPv6 header.
 Types of Extension Headers:
o Hop-by-Hop Options Header: Used to pass information to all
routers along the path.
o Routing Header: Specifies intermediate nodes the packet must
pass through.
o Fragment Header: Used for fragmentation and reassembly,
similar to IPv4.
o Destination Options Header: Allows optional data for the
destination node.
o Authentication and Encapsulation Headers: Used for IPsec
security.

ICMP (Internet Control Message Protocol)

ICMP is used for diagnostic and error reporting purposes in IPv4
networks.
 Common ICMP Messages:
o Echo Request and Echo Reply: Used for ping operations to test
connectivity.
o Destination Unreachable: Reports that a destination cannot be
reached.
o Time Exceeded: Indicates that the TTL value has expired.

IGMP (Internet Group Management Protocol)

IGMP is used by IPv4 hosts to manage membership in multicast groups.
It enables devices to join or leave multicast groups for efficient data
distribution (e.g., video streaming).

ARP (Address Resolution Protocol)

 ARP is used to map IPv4 addresses to MAC addresses within a local
network. When a device wants to communicate with another device on
the same network, ARP resolves the destination device's MAC address.

RARP (Reverse Address Resolution Protocol)

RARP is used to map a MAC address to an IPv4 address. It is used by
devices that don't have a fixed IP address (e.g., diskless workstations) to
obtain an IP address from a RARP server.

Congestion Control and Resource Allocation

Congestion Control: The mechanism to prevent network congestion by
managing the rate of data flow to avoid overloading the network.
 Issues: Congestion occurs when too much data is injected into the
network, causing delays and packet loss.
 Queuing: Data is queued at routers or switches when there is congestion,
which leads to increased delays.
 TCP (Transmission Control Protocol):
o Flow Control: Ensures that the sender does not overwhelm the
receiver.
o Congestion Window: Controls the amount of data that can be sent
based on network conditions.
o TCP Congestion Control Algorithms: Include Slow Start,
Congestion Avoidance, Fast Retransmit, and Fast Recovery.

Unit – IV
Congestion Control, Congestion-Avoidance Mechanisms, and Quality of
Service (QoS)

Congestion Control
Congestion Control refers to mechanisms that prevent network congestion or
mitigate its effects once it occurs. Network congestion happens when there is
too much data traffic in the network, which can lead to delays, packet loss, and
inefficient communication.
Key Concepts in Congestion Control:
1. Feedback Mechanism: Routers or switches provide feedback to the
sender about congestion levels in the network, usually by dropping
packets or sending congestion notifications.
2. Traffic Shaping: Traffic shaping involves controlling the amount and the
rate of data entering the network to avoid congestion.
3. Window-Based Flow Control: Mechanisms like TCP use a window of
available bandwidth to adjust the rate of data transmission based on
network congestion levels.

Congestion-Avoidance Mechanisms
Congestion Avoidance aims to prevent congestion before it happens, rather
than reacting to it. These mechanisms are crucial for maintaining a stable
network performance, especially in scenarios with fluctuating traffic volumes.
Key Congestion-Avoidance Mechanisms:
1. Slow Start (TCP): TCP begins transmission with a small congestion
window, gradually increasing it as it detects that the network is not
congested. If congestion is detected (packet loss), it reduces the window
size.
2. Congestion Window (TCP): The sender adjusts the amount of data sent
based on feedback from the network. The congestion window increases
or decreases depending on packet loss and acknowledgment.
3. Random Early Detection (RED): RED is used in network routers to
detect congestion early and drop packets randomly before the network
reaches full congestion, thus signaling senders to slow down.
4. Explicit Congestion Notification (ECN): ECN allows routers to mark
packets (instead of dropping them) to notify the sender about congestion.

Quality of Service (QoS)

Quality of Service (QoS) refers to the mechanisms that provide guaranteed
service levels, such as bandwidth, delay, and packet loss. It is used to ensure
that important traffic (like voice and video) gets the necessary resources to meet
its requirements, even in congested network conditions.
Key QoS Techniques:
1. Traffic Classification and Marking: Packets are classified based on
their importance (e.g., voice, video, or data) and are marked for special
handling.
2. Traffic Policing and Shaping: These mechanisms regulate the flow of
traffic entering the network, ensuring that data conforms to the specified
traffic profile.
3. Queue Management: Routers and switches manage multiple queues for
different traffic types. Higher-priority queues are given preferential
treatment in terms of delay and packet loss.
4. Differentiated Services (DiffServ): DiffServ marks packets with
different priority levels, ensuring high-priority traffic (e.g., VoIP) gets
preferential treatment.

Internetworking: Intra-Domain and Inter-Domain Routing

Internetworking involves connecting different networks (domains) to enable
communication between them. Routing within and between these domains is
essential to direct data from the source to the destination across multiple
networks.
Intra-Domain Routing:
Intra-domain routing refers to routing within a single autonomous system
(AS), typically controlled by one organization or entity.
1. Interior Gateway Protocols (IGP): Used for routing within an AS.
Examples include:
o RIP (Routing Information Protocol)
o OSPF (Open Shortest Path First)
o EIGRP (Enhanced Interior Gateway Routing Protocol)
Inter-Domain Routing:
Inter-domain routing refers to routing between multiple ASes. This type of
routing helps in determining the best path for data to travel between different
networks, typically using:
 1. Exterior Gateway Protocols (EGP): Used for routing between different
ASes. The primary protocol used for inter-domain routing is:
o BGP (Border Gateway Protocol), specifically BGP-4 for IPv4.

Unicast Routing Protocols

Unicast Routing protocols are used for routing data packets from a single
sender to a single destination.
1. RIP (Routing Information Protocol):
o Type: Distance-vector protocol.
o Operation: RIP uses hop count as the metric for determining the
best path to a destination. It limits the number of hops to 15, which
can make it unsuitable for larger networks.
o Versions: RIP v1 (classful) and RIP v2 (classless, supports
VLSM).
o Limitations: Slow convergence and scalability issues.
2. OSPF (Open Shortest Path First):
o Type: Link-state protocol.
o Operation: OSPF uses a more sophisticated algorithm (Dijkstra’s
algorithm) to compute the shortest path based on cost, which is
typically the bandwidth of links.
o Features: Scalable, faster convergence, supports VLSM and
CIDR.
o Used in: Large enterprise networks.
3. BGP (Border Gateway Protocol):
o Type: Path-vector protocol, used for inter-domain (inter-AS)
routing.
o Operation: BGP uses AS-path as a metric, choosing the best path
based on policy and rules set by network administrators.
o Types: IBGP (Internal BGP, within an AS) and EBGP (External
BGP, between ASes).
o Used in: Internet backbone, large-scale ISPs.
Multicast Routing Protocols
Multicast routing protocols are designed for routing data from a single sender to
multiple receivers (group communication). These protocols are used to
efficiently deliver data, such as video streaming, to multiple users.
1. DVMRP (Distance Vector Multicast Routing Protocol):
o Type: Distance-vector multicast routing protocol.
o Operation: DVMRP uses a distance-vector algorithm to construct
multicast trees, with each router maintaining its own routing table.
o Limitations: Scalability issues in large networks.
2. PIM-DM (Protocol Independent Multicast - Dense Mode):
o Type: Protocol-independent multicast routing.
o Operation: PIM-DM uses a flooding and pruning mechanism to
create a multicast distribution tree. It assumes that multicast traffic
is needed everywhere, so it initially floods the network and then
prunes unused branches.
o Use Case: Suitable for networks with dense multicast receivers.
3. PIM-SM (Protocol Independent Multicast - Sparse Mode):
o Type: Protocol-independent multicast routing.
o Operation: PIM-SM assumes that multicast receivers are sparsely
distributed across the network. It uses a shared tree for routing and
then switches to a source-specific tree for optimized delivery.
o Use Case: Ideal for networks with sparse multicast receivers.
4. CBT (Core-Based Trees):
o Type: Multicast routing protocol.
o Operation: CBT builds a shared distribution tree with a central
"core" router, and all multicast traffic flows through this core.
o Advantages: Reduces routing overhead by using a single tree for
multicast traffic.
5. MSDP (Multicast Source Discovery Protocol):
o Type: Protocol used in conjunction with PIM-SM.
 o Operation: MSDP allows sources to be discovered across different
multicast domains, allowing for inter-domain multicast.
6. MOSPF (Multicast Open Shortest Path First):
o Type: An extension of OSPF to support multicast.
o Operation: MOSPF allows OSPF routers to propagate multicast
routing information.

Spanning Tree Algorithm

The Spanning Tree Algorithm (STA) is used in Ethernet networks to prevent
looping in the network topology. The algorithm ensures that there is only one
active path between any two switches in a network by creating a spanning tree.
 Operation: STA selects a root bridge (switch) and computes the shortest
path to all other switches, blocking any redundant paths that could form
loops.
 Protocol: The Spanning Tree Protocol (STP) is the standard used to
implement the spanning tree algorithm in Ethernet networks.
 Key Concepts:
o Root Bridge: The central switch from which all paths are
computed.
o Port Roles: Ports can be designated as root port, designated port,
or blocked port.
o Bridge Protocol Data Units (BPDU): Messages exchanged
between switches to maintain the spanning tree.

Summary
1. Congestion Control: Prevents or mitigates congestion using feedback
and flow control techniques, such as TCP congestion window, RED, and
ECN.
2. QoS: Ensures that critical traffic (e.g., VoIP) gets necessary bandwidth
and low latency.
3. Routing:
 o Intra-Domain: Uses protocols like RIP, OSPF, and BGP for
routing within and between ASes.
o Multicast Routing: Protocols like DVMRP, PIM, and CBT
optimize data delivery to multiple receivers.
4. Spanning Tree: Ensures loop-free network topology in Ethernet-based
networks.
Unit – V
Optical Networking: SONET/SDH Standards
Optical networking uses fiber optic cables to carry high-speed data signals
over long distances with minimal signal degradation. SONET (Synchronous
Optical Network) and SDH (Synchronous Digital Hierarchy) are the two
primary standards for optical networking, used to transport digital signals over
optical fiber.
SONET/SDH Standards:
 SONET: Primarily used in North America, SONET provides
standardized data rates for transmission in optical networks.
 SDH: A similar standard used worldwide, except in North America. SDH
and SONET are compatible, and SDH was developed to standardize
international networks.
Key Features of SONET/SDH:
1. Hierarchical Structure: Both SONET and SDH organize signals in a
hierarchy, with each level representing a multiple of base transmission
rates (e.g., OC-1 or STM-1).
2. Synchronous Transmission: The use of a synchronous clock allows
multiple signals to be multiplexed over the same optical fiber, making it
efficient and less prone to signal interference.
3. Scalability: SONET/SDH can handle data rates from low-speed links
(like 1.5 Mbps) up to ultra-high-speed ones (up to 40 Gbps or higher).
4. Protection: Features such as automatic protection switching ensure that
the network can recover from failures quickly by rerouting data.

Traffic Engineering
Traffic Engineering refers to the process of optimizing the flow of data across
a network to ensure efficient usage of resources and maintain performance
requirements.
Traffic Engineering Requirements:
1. Capacity Planning: Ensuring that the network has sufficient bandwidth
to handle peak traffic loads.
2. Traffic Sizing: Determining the amount of traffic a network segment can
carry without degradation.
3. Characterization: Analyzing traffic types (e.g., voice, video, data) to
optimize network design and performance.
Key Factors in Traffic Engineering:
 Protocols: Protocols like MPLS (Multiprotocol Label Switching) and
RSVP (Resource Reservation Protocol) are used for path selection, load
balancing, and resource reservation.
 Time and Delay Considerations: Networks should minimize delays,
particularly for latency-sensitive applications like VoIP and real-time
streaming.
 Throughput: The network's ability to carry data efficiently over time,
which needs to be optimized for all traffic types.
 Availability and Reliability: Traffic engineering also ensures high
availability (uptime) and reliability through redundancy and fault-tolerant
paths.

Multimedia Over Internet

Multimedia over the Internet involves transmitting multiple forms of media
(audio, video, images) over the internet. Key technologies include IP
Multicasting and VoIP.
IP Multicasting:
 IP Multicast is a communication method used to send data to multiple
destinations efficiently. Instead of sending multiple copies of the same
data to each individual receiver (as with unicast), multicast sends a single
copy to all recipients who subscribe to a multicast group.
  Protocols like PIM (Protocol Independent Multicast) and IGMP
(Internet Group Management Protocol) are used to manage multicast
group memberships and the forwarding of multicast traffic.
VoIP (Voice over IP):
 VoIP allows voice communication to be transmitted over IP networks.
Key protocols include:
o SIP (Session Initiation Protocol): A signaling protocol used to set
up and manage communication sessions.
o RTP (Real-time Transport Protocol): Used to carry the voice
data packets.
o QoS: Ensures that voice traffic receives the necessary priority to
maintain voice quality.

Domain Name System (DNS)

The Domain Name System (DNS) is a hierarchical system that translates
human-readable domain names (e.g., www.example.com) into IP addresses,
which computers use to communicate with each other.
DNS Structure:
 Name Space: The DNS name space is a tree structure, with the root at the
top and subdomains branching down (e.g., .com, .org).
 Domains: Each level of the tree represents a domain. Domains are used
to organize the DNS name space into manageable units.
 Resolution: When you query a domain name, DNS resolves it by
contacting the appropriate servers to return the corresponding IP address.
 Dynamic DNS: Unlike static DNS, Dynamic DNS (DDNS) allows DNS
records to be updated in real time, enabling applications where IP
addresses may change frequently (e.g., for remote workers or dynamic
IPs).

SNMP (Simple Network Management Protocol)

SNMP is a protocol used for network management, allowing for the monitoring
and control of devices such as routers, switches, and servers.
  Operations: It uses a manager-agent model. The manager queries or
configures agents on network devices.
 MIB (Management Information Base): A database of objects that
represent the devices’ data (e.g., CPU usage, bandwidth utilization).
 Versions: SNMPv1, SNMPv2, and SNMPv3, with SNMPv3 providing
enhanced security.

Security Protocols
1. IPSec (Internet Protocol Security):
o IPSec provides secure communication over IP networks through
encryption and authentication mechanisms.
o Modes:
 Transport mode: Encrypts only the payload of the packet.
 Tunnel mode: Encrypts the entire packet, providing secure
communication for VPNs.
2. SSL/TLS (Secure Sockets Layer/Transport Layer Security):
o SSL/TLS is a cryptographic protocol that ensures privacy and
integrity of data during transmission over networks.
o Commonly used for securing web communications (HTTPS).
3. PGP (Pretty Good Privacy):
o PGP is used for email encryption and file encryption. It ensures
data privacy through public-key cryptography.
4. Firewalls:
o Firewalls are security systems that control the incoming and
outgoing network traffic based on predetermined security rules.
o Types:
 Packet filtering firewalls: Inspect packets and determine if
they should be allowed or blocked.
 Stateful inspection: Keeps track of the state of active
connections and allows packets to flow based on the state.
Datacenter Design and Interconnection Networks
Datacenter Design focuses on the physical and logical infrastructure of
datacenters, including servers, storage, networking, and power systems.
Key Aspects:
1. Scalability: The datacenter should be able to scale with growing business
needs.
2. Redundancy: Critical components should be redundant to ensure high
availability.
3. Efficiency: Power and cooling systems should be optimized for energy
efficiency.
Interconnection Networks:
 Datacenter Networks (DCNs) connect servers and storage systems
within a datacenter. These networks typically use technologies such as:
o Ethernet: A common technology for local interconnections in a
datacenter.
o InfiniBand: A high-speed interconnection technology used for
low-latency, high-throughput networking.
o Leaf-Spine Architecture: A two-tier architecture where leaf
switches connect to spine switches, allowing for efficient data flow.

Summary
 SONET/SDH standards enable high-speed optical networking with
synchronous transmission.
 Traffic Engineering ensures efficient data transmission by considering
traffic size, protocols, and delay.
 Multimedia over the Internet includes technologies such as IP
Multicasting and VoIP for transmitting voice and video data efficiently.
 DNS translates domain names to IP addresses and supports dynamic
updates.
 SNMP facilitates network management through monitoring and
configuration.
 Security protocols like IPSec, SSL/TLS, and PGP ensure secure
communication over networks.
 Datacenter Design and Interconnection Networks focus on building
efficient, scalable, and redundant infrastructures for data storage and
processing.