Scribd
Upload
7 views

Unit 4 Chapter 5 Network Layer

The network layer, operating at Layer 3 of the OSI Model, is essential for routing data packets across networks, utilizing both connectionless (IP) and connection-oriented (virtual-circuit) services. Key concepts include routing algorithms like Dijkstra's for optimal pathfinding, addressing challenges such as the count-to-infinity problem in distance vector routing, and modern solutions like CIDR, NAT, and IPv6 to enhance address management and network efficiency. Internet Control Protocols like ICMP, ARP, and DHCP facilitate network operations and ensure reliable communication.

Uploaded by

Gunjan Arora
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
7 views

Unit 4 Chapter 5 Network Layer

The network layer, operating at Layer 3 of the OSI Model, is essential for routing data packets across networks, utilizing both connectionless (IP) and connection-oriented (virtual-circuit) services. Key concepts include routing algorithms like Dijkstra's for optimal pathfinding, addressing challenges such as the count-to-infinity problem in distance vector routing, and modern solutions like CIDR, NAT, and IPv6 to enhance address management and network efficiency. Internet Control Protocols like ICMP, ARP, and DHCP facilitate network operations and ensure reliable communication.

Uploaded by

Gunjan Arora
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Unit -4 Chapter-5: Network

Layer
The network layer is crucial for delivering data packets from a source to a
destination across multiple networks, operating at Layer 3 of the OSI Model. Key
aspects include store-and-forward packet switching, where routers store and
forward packets, and the debate between connection-oriented and
connectionless services. The Internet favors a connectionless model with IP,
while connection-oriented services like MPLS support Quality of Service (QoS).
Routing algorithms manage packet paths, balancing fairness and efficiency.

Created by : Devavrat Kumar Mahato

AE - 312 PMCS - 3rd Year

Submitted to : Ms. Nishu


Connectionless vs. Connection-Oriented Services
Virtual circuits require initial setup time but simplify routing. Datagram networks need
complex address lookups but no setup phase. Virtual circuits use short connection
numbers, while datagrams require longer addresses. Virtual circuits allow pre-allocation of
resources for QoS, while datagram networks struggle with congestion control. Datagram
networks are flexible and handle failures well, while virtual circuits are ideal for long-
running sessions.

The network layer offers connectionless (datagram) and connection-oriented (virtual-


circuit) services. In a datagram network, packets are routed independently using
destination addresses. The Internet Protocol (IP) is the dominant connectionless service.
Virtual-circuit networks establish a predefined path before sending data, with routers
assigning connection identifiers. Label switching, including Multi Protocol Label Switching
(MPLS), efficiently routes data based on labels.
Routing Algorithms: Optimality and Shortest Paths
Dijkstra's algorithm computes the shortest paths from a source node to all other nodes in a network.
It uses a graph representation with nodes as routers and edges as communication links. The
algorithm starts by marking the source node as permanent, updating labels of adjacent nodes with
the shortest known distances, and iteratively making the node with the smallest label permanent.
This process repeats until all nodes are permanent, ensuring the shortest paths are discovered.

The primary function of the network layer is routing packets from source to destination. Routing
involves selecting paths, while forwarding transmits packets based on routing tables. Desirable
routing algorithm properties include robustness, stability, fairness, and efficiency. Algorithms are
classified into nonadaptive (static) and adaptive (dynamic). The optimality principle states that if a
router is on the optimal path between two points, then the path from that router to the destination
is also optimal.
Addressing and Routing Challenges

The count-to-infinity problem occurs in distance vector routing when routers react quickly to good news (shorter paths) but slowly to
bad news (failed links or routers). It happens because routers rely on neighbor updates to estimate paths, which can lead to incorrect
routing loops.

IPv4 uses 32-bit addresses, assigning each network interface its own IP address. IPv4 addresses are hierarchical, divided into a network
portion (prefix) and a host portion. Prefixes reduce router table sizes, but the hierarchy can be wasteful of addresses. Subnetting divides a
larger block into smaller subnets for internal use, optimizing address usage. Classful addressing, with classes A, B, C, D, and E, has issues
such as address space wastage and routing table explosion.
Modern Solutions: CIDR, NAT, and IPv6
1. CIDR (Classless Inter-Domain Routing) solves routing table explosion by using route aggregation to combine smaller prefixes into larger
ones (supernets).
2. NAT (Network Address Translation) allows multiple devices to share a single public IP by mapping private IP addresses to a public one,
addressing IP address shortage.
3. IPv6 uses 128-bit addresses, ensuring an abundant supply of addresses, and includes features like simplified headers and enhanced
security.

Internet Control Protocols like ICMP, ARP, and DHCP assist in network operations. ICMP reports errors and tests network connectivity. ARP
maps IP addresses to MAC addresses. DHCP manages automatic IP address assignment to network devices. These protocols are essential
for efficient and reliable network communication.

You might also like