Fragmentation is the process of dividing a large IP datagram into smaller fragments so that it can be transmitted through a network whose Maximum Transmission Unit (MTU) is smaller than the size of the original packet.
IPv4 can have a maximum packet size of 2¹⁶ − 1 = 65,535 bytes, because the Total Length field is 16 bits. However, the Data Link Layer may support a smaller MTU. For example, Ethernet MTU = 1500 bytes. Therefore, a large IP packet may need to be fragmented before forwarding.
Why is Fragmentation Required?
Fragmentation is required when the size of an IPv4 packet is greater than the MTU of the outgoing network. The router divides the original datagram into smaller fragments that can fit within the available MTU.
Important IPv4 Fragmentation Fields:
Identification (ID): A unique value used to identify fragments belonging to the same original IP datagram. All fragments of one original packet have the same Identification value.
DF (Don't Fragment) Flag: If DF = 1, fragmentation is not allowed. If the packet is larger than the outgoing MTU, the router cannot forward it and must drop it.
MF (More Fragments) Flag: If MF = 1, more fragments are still to follow. The last fragment has MF = 0.
Fragment Offset: It indicates the position of a fragment's data within the original datagram. The Fragment Offset field is 13 bits. The offset is represented in units of 8 bytes, so fragment payload sizes (except the last fragment) must be multiples of 8 bytes.
Key Formula for Fragment Offset:
Fragment Offset = Starting Byte of Fragment ÷ 8
Example: If the first fragment contains 3976 bytes, then the next fragment starts at byte 3976.
Therefore:
3976 ÷ 8 = 497
So, the Fragment Offset of the second fragment is 497.
Important Numerical Example:
Suppose the original IP packet contains 10,000 bytes of payload. It can be fragmented into:
3976 + 3976 + 2048 = 10,000 bytes
The three fragments will have:
Fragment 1: Payload = 3976 bytes, Offset = 0, MF = 1
Fragment 2: Payload = 3976 bytes, Offset = 497, MF = 1
Fragment 3: Payload = 2048 bytes, Offset = 994, MF = 0
All three fragments have the same Identification (ID).
Why 3976 Bytes?
The minimum IPv4 header size is 20 bytes. If the MTU is 4000 bytes, the maximum payload would initially appear to be:
4000 − 20 = 3980 bytes
But 3980 is not divisible by 8. Since the Fragment Offset uses 8-byte units, the payload must be reduced to 3976 bytes.
3976 ÷ 8 = 497
Therefore, 3976 bytes is used instead of 3980 bytes for the fragment payload.

Given:
Total IP Packet Length = 4500 bytes
IPv4 Header = 20 bytes
TCP Header = 40 bytes
MTU = 600 bytes
First Fragment Offset = 0
Step 1: Calculate IP Payload
IP Payload = Total IP Packet Length − IP Header
IP Payload = 4500 − 20
IP Payload = 4480 bytes
Note: The 40-byte TCP header is already included inside the IP payload, so it is not subtracted separately.
Step 2: Calculate Maximum Data per Fragment
Maximum Data = MTU − IP Header
Maximum Data = 600 − 20
= 580 bytes
Since the Fragment Offset is measured in units of 8 bytes, the data size of each fragment (except the last) must be divisible by 8.
580 is not divisible by 8, so:
Maximum Fragment Data = 576 bytes
Step 3: Calculate Fragment Offsets
1st Fragment:
Starting Byte = 0
Fragment Offset = 0 ÷ 8 = 0
2nd Fragment:
Starting Byte = 0 + 576 = 576
Fragment Offset = 576 ÷ 8 = 72
3rd Fragment:
Starting Byte = 576 + 576 = 1152
Fragment Offset = 1152 ÷ 8 = 144
Therefore:
Fragmentation Offset Value in the Third Fragment = 144
Final Answer: 144

The IPv4 header option field can have a maximum length of 40 bytes. In the Record Route option, 3 bytes are used for the option type, length, and pointer fields. Each IPv4 address requires 4 bytes.
Therefore:
(40 − 3) / 4 = 37 / 4 = 9
Hence, the maximum number of IPv4 router addresses that can be recorded is 9.

Given:

• IP Datagram Size = 1000 bytes
• MTU = 100 bytes
• IP Header Size = 20 bytes

Step 1: Calculate the data size
Data Size = Total Datagram Size − Header Size
Data Size = 1000 − 20 = 980 bytes
Step 2: Calculate the maximum data size per fragment
Each fragment must fit within the MTU of 100 bytes.
Maximum data per fragment = MTU − Header Size
= 100 − 20 = 80 bytes
Step 3: Calculate the number of fragments
Number of fragments = 980 / 80
= 12.25
Since the last fragment can contain the remaining data, we need 13 fragments.

Distance Vector Routing is a routing algorithm in which a router determines the best path based mainly on information received from its neighboring routers.
The term Distance Vector has two parts:
Distance means the cost or distance required to reach a destination. The distance may be measured using metrics such as hop count.
Vector means the direction or next-hop router through which the destination can be reached.
In Distance Vector Routing, each router maintains a routing table containing information about the destination, distance/cost, and next hop.
Example: If Router R1 sends a packet through Router R4 to reach Router R5, then R4 can be considered the next-hop router and the associated cost represents the distance.
Algorithm Used: Bellman-Ford Algorithm
Common Protocols: RIP (Routing Information Protocol) and IGRP (Interior Gateway Routing Protocol).
Working of Distance Vector Routing:

• 1. Each router creates and maintains its routing table.
• 2. The router identifies its directly connected or neighboring routers.
• 3. The router determines the distance or cost to its neighboring routers.
• 4. Routers exchange routing information with their neighboring routers.
• 5. Each router updates its routing table based on the information received from its neighbors.
• 6. The process is repeated until the routers find the best available paths.
• 7. The shortest or lowest-cost path is selected for packet forwarding.

Characteristics of Distance Vector Routing:

• • Uses information mainly about neighboring routers.
• • Uses the Bellman-Ford algorithm.
• • Routing information is exchanged periodically with neighbors.
• • Convergence is relatively slow.
• • It may suffer from the Count-to-Infinity problem.
• • It generally requires less network information than Link State Routing.

Step 1: Initial Routing Table of Router A
Router A has a direct connection only with Router E, and the cost of A → E = 5. Therefore, the other routers are initially considered unreachable from A.

Destination	Distance	Next Hop
A	0	—
B	∞	—
C	∞	—
D	∞	—
E	5	E

Explanation:
A → A = 0 because the source and destination are the same router.
A → E = 5 because A has a direct link to E.
A → B, A → C, and A → D are initially ∞ because there are no direct links from A to these routers.

---

Router E's Routing Table
Router E is directly connected to A, B, and D. Therefore:

• E → A = 5
• E → B = 4
• E → D = 7
• E → C = ∞ initially because there is no direct E-C link.

Destination	Distance	Next Hop
A	5	A
B	4	B
C	∞	—
D	7	D
E	0	—

Step 2: Router A Receives Router E's Distance Vector
The cost from A to E is:

c(A,E) = 5
Router A now uses the Bellman-Ford equation:

DA(Y) = min { DA(Y), c(A,E) + DE(Y) }
We calculate the distance to each destination using Router E as the intermediate/next-hop router.
1. Destination A
DA(A) = min { 0, 5 + 5 } = 0
2. Destination B
DA(B) = min { ∞, 5 + 4 }
DA(B) = min { ∞, 9 } = 9
Therefore, the path is: A → E → B
3. Destination C
DA(C) = min { ∞, 5 + ∞ } = ∞
Router E currently has no direct known route to C in the given distance vector.
4. Destination D
DA(D) = min { ∞, 5 + 7 }
DA(D) = min { ∞, 12 } = 12
Therefore, the path is: A → E → D
5. Destination E
DA(E) = min { 5, 5 + 0 } = 5
Therefore, A reaches E directly with a cost of 5.

---

Updated Routing Table of Router A

Destination	Old Distance	Via E	New Distance	Next Hop
A	0	5 + 5 = 10	0	—
B	∞	5 + 4 = 9	9	E
C	∞	5 + ∞ = ∞	∞	—
D	∞	5 + 7 = 12	12	E
E	5	5 + 0 = 5	5	E

Final Result:
After receiving Router E's distance vector, Router A learns routes to B and D through E.
A → E → B = 5 + 4 = 9
A → E → D = 5 + 7 = 12
Router A still has no route to C at this stage because Router E's current distance vector shows C = ∞.
Therefore, the updated routing table of A is:

Destination	Distance	Next Hop
A	0	—
B	9	E
C	∞	—
D	12	E
E	5	E

Link State Routing is a routing algorithm in which each router builds a complete view or topology map of the network and uses that information to calculate the shortest path.
Unlike Distance Vector Routing, a router does not depend only on information about its immediate neighbors. Each router eventually obtains information about the entire network topology.
Algorithm Used: Dijkstra's Shortest Path First (SPF) Algorithm
Common Protocols: OSPF (Open Shortest Path First) and IS-IS (Intermediate System to Intermediate System).
Working of Link State Routing:

• 1. Each router discovers its neighboring routers.
• 2. The router measures the cost or distance of its links to neighboring routers.
• 3. The router creates a Link State Packet (LSP) containing information about its links and neighbors.
• 4. The LSP is distributed to other routers through a process called flooding.
• 5. Every router receives sufficient link-state information to build a complete network topology map.
• 6. Each router runs the Dijkstra algorithm to calculate the shortest path to each destination.
• 7. Each router creates its routing table based on the calculated shortest paths.

Characteristics of Link State Routing:

• • Provides a view of the entire network topology.
• • Uses the Dijkstra (Shortest Path First) algorithm.
• • Uses Link State Packets (LSPs).
• • Uses flooding to distribute link-state information.
• • Converges faster than Distance Vector Routing.
• • Does not have the traditional Count-to-Infinity problem of Distance Vector Routing.
• • Requires more processing power and memory because each router maintains topology information.

Step	N'	D(B), P(B)	D(C), P(C)	D(D), P(D)	D(E), P(E)	D(F), P(F)
0	{A}	(2, A)	(5, A)	(1, A)	(∞, —)	(∞, —)
1	{A,D}	(2, A)	(4, D)	(1, A)	(2, D)	(∞, —)
2	{A,D,B}	(2, A)	(4, D)	(1, A)	(2, D)	(∞, —)
3	{A,D,B,E}	(2, A)	(3, E)	(1, A)	(2, D)	(4, E)
4	{A,D,B,E,C}	(2, A)	(3, E)	(1, A)	(2, D)	(4, E)
5	{A,D,B,E,C,F}	(2, A)	(3, E)	(1, A)	(2, D)	(4, E)

• A → B: Distance = 2, Predecessor = A
• A → D: Distance = 1, Predecessor = A
• A → E: Distance = 2, Predecessor = D
• A → C: Distance = 3, Predecessor = E
• A → F: Distance = 4, Predecessor = E

Therefore, the shortest path from A to F is:

A → D → E → F

Total Path Cost = 1 + 1 + 2 = 4
Final Answer: The shortest path from A to F is A → D → E → F, and the minimum total cost is 4.

1. Knowledge:
Distance Vector Routing uses mainly local/neighbor information, whereas Link State Routing maintains knowledge of the entire network topology.
2. Algorithm:
Distance Vector Routing uses the Bellman-Ford algorithm, whereas Link State Routing uses the Dijkstra algorithm.
3. Information Exchange:
Distance Vector routers exchange routing information mainly with their neighbors, whereas Link State routers distribute Link State Packets throughout the network using flooding.
4. Convergence:
Distance Vector Routing generally has slower convergence, whereas Link State Routing generally has faster convergence.
5. Count-to-Infinity Problem:
Distance Vector Routing can suffer from the Count-to-Infinity problem, whereas Link State Routing does not have this traditional problem.
6. Network View:
Distance Vector Routing has a limited view of the network, whereas Link State Routing has a complete topology view.
7. Resource Requirement:
Distance Vector Routing generally requires less memory and processing, whereas Link State Routing requires more memory and processing to maintain the topology database and calculate shortest paths.
8. Examples:
Distance Vector: RIP, IGRP
Link State: OSPF, IS-IS

Comparison Table:

Feature	Distance Vector Routing	Link State Routing
Network Knowledge	Neighbor/local information	Complete network topology
Algorithm	Bellman-Ford	Dijkstra
Information Exchange	With neighboring routers	Link State Packets are flooded
Convergence	Slower	Faster
Count-to-Infinity	Possible	Not a traditional issue
Complexity	Relatively simple	More complex
Resource Requirement	Lower	Higher
Examples	RIP, IGRP	OSPF, IS-IS

Routing Protocol is a set of rules used by routers to exchange routing information and determine the best path for forwarding packets from a source network to a destination network.
Routing protocols are mainly used in dynamic routing, where routers automatically learn, update, and maintain routing information.
Three important routing protocols are:
1. RIP — Routing Information Protocol
2. OSPF — Open Shortest Path First
3. BGP — Border Gateway Protocol

RIP is a Distance Vector routing protocol. It determines the best route mainly using hop count as its routing metric.
In RIP, a router exchanges routing information with its neighboring routers and learns the distance to different destination networks.
Important Features of RIP:

• • Type: Distance Vector
• • Metric: Hop Count
• • Algorithm: Bellman-Ford based
• • Maximum usable hop count: 15
• • 16 hops: Considered unreachable
• • Transport: UDP
• • Port: UDP 520
• • Mainly suitable for small networks.

Example:
If one route requires 3 hops and another route requires 5 hops, RIP normally selects the 3-hop route, regardless of whether the 5-hop route has higher bandwidth.
Advantages of RIP:

• • Simple to configure.
• • Easy to understand.
• • Requires relatively low resources.
• • Suitable for small and simple networks.

Disadvantages of RIP:

• • Limited to 15 usable hops.
• • Slow convergence compared with OSPF.
• • Does not consider bandwidth when selecting the route.
• • Not suitable for large and complex networks.

OSPF is a Link State routing protocol. It builds a topology database of the network and uses Dijkstra's Shortest Path First (SPF) algorithm to calculate the shortest path.
Unlike RIP, OSPF does not simply count the number of routers. It uses a cost metric, which is commonly related to link bandwidth.
Important Features of OSPF:

• • Type: Link State
• • Algorithm: Dijkstra / SPF
• • Metric: Cost
• • Protocol Number: IP Protocol 89
• • Used mainly as an Interior Gateway Protocol (IGP).
• • Suitable for medium and large networks.
• • Supports hierarchical network design using Areas.
• • Area 0 is known as the backbone area.

How OSPF Works:

• 1. Routers discover their neighboring routers.
• 2. Routers exchange Link-State Advertisements (LSAs).
• 3. Each router builds a Link-State Database (LSDB).
• 4. The router creates a complete view of the network topology.
• 5. Dijkstra's SPF algorithm is used to calculate the shortest paths.
• 6. The router creates its routing table.

Advantages of OSPF:

• • Fast convergence.
• • Highly scalable.
• • Suitable for large enterprise networks.
• • Supports hierarchical routing using areas.
• • Supports VLSM and CIDR.
• • Provides efficient path selection using cost.

Disadvantages of OSPF:

• • More complex than RIP.
• • Requires more CPU and memory resources.
• • Configuration and network design can be more complicated.

BGP is a Path Vector routing protocol. It is mainly used to exchange routing information between different Autonomous Systems (AS).
BGP is the primary routing protocol used for inter-domain routing on the Internet.
Important Features of BGP:

• • Type: Path Vector
• • Routing Scope: Inter-domain / Between Autonomous Systems
• • Transport Protocol: TCP
• • Port: TCP 179
• • Uses multiple path attributes and routing policies.
• • Highly scalable and suitable for Internet-scale routing.

Important BGP Path Attributes:

• • AS-PATH
• • NEXT-HOP
• • LOCAL-PREFERENCE
• • MED (Multi-Exit Discriminator)

Types of BGP:
eBGP (External BGP): Used for exchanging routing information between different Autonomous Systems.
iBGP (Internal BGP): Used for exchanging BGP routing information within the same Autonomous System.
Advantages of BGP:

• • Extremely scalable.
• • Suitable for Internet-scale routing.
• • Provides powerful routing policy control.
• • Supports route filtering and route aggregation.
• • Can connect different Autonomous Systems.

Disadvantages of BGP:

• • Complex to configure and manage.
• • Requires significant routing knowledge.
• • Convergence is generally slower than OSPF.
• • Requires more powerful network equipment in large deployments.

Feature	RIP	OSPF	BGP
Full Form	Routing Information Protocol	Open Shortest Path First	Border Gateway Protocol
Type	Distance Vector	Link State	Path Vector
Algorithm	Bellman-Ford based	Dijkstra / SPF	Path Vector / Policy-based
Metric	Hop Count	Cost	Path Attributes / Policies
Routing Scope	Intra-domain	Intra-domain	Inter-domain
Suitable For	Small Networks	Medium/Large Networks	Internet / ISP Networks
Scalability	Low	High	Very High
Convergence	Slow	Fast	Generally slower, stability/policy focused
Transport	UDP 520	IP Protocol 89	TCP 179