ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780

Vol.9 (1), January (2019), pp. 532-538

Online available at zenithresearch.org.in
An Analytical Approach to the Classification and Optimization of Routing Algorithms in
Computer Networks.

Sudhir Agarmore1), Dr.Hemant Mahalle 2)

1
Assistant Professor
R.A.Arts, Shri M.K.Commerce & Shri S.R.Rathi Science College, Washim(MH), India.
2
Associate Professor
Phulsing Naik Mahavidyalaya, Pusad
Dist. Yavatmal (M.S) 445216.
Author Emails
1)
[email protected]
2)
[email protected]

Abstract.This paper offers a thorough overview of various routing algorithms working in different
network types, including Local Area Networks (LAN), Metropolitan Area Networks (MAN), and
Wide Area Networks (WAN). Algorithms play a crucial role in the effective and dependable
functioning of computer networks. These identify the best routes for data packets to move from
source to destination. Algorithms are classified based on their operational characteristics, such as
static and dynamic routing, distance vector and link state routing, and adaptive versus non-adaptive
routing, Centralized and Distributed Routing.
Keywords. Routing Algorithms, Static Routing, Dynamic Routing, Centralized Routing, Distributed
Routing, Scalability, Adaptability, Resource Usage, Complexity,Optimal Path.
Introduction. The main purpose of routing algorithms is to guarantee the efficient and precise
transmission of data packets throughout a network, from their starting point to their endpoint. Distinct
types of networks, such as LAN, MAN, and WAN, have specific needs and challenges that impact the
selection of routing algorithms. This paper study the different varieties of routing algorithms, their
operational characteristics, benefits, drawbacks, and their use in various network settings.

Classification of Routing Algorithms

1. Static vs. Dynamic Routing
Static Routing plays a vital role in networking, guiding data traffic between different networks. The
two primary forms of routing are static routing and dynamic routing, each with its own set of pros,
cons, and applications
Static routing entails the manual configuration of routes within a network. Here, a network
administrator creates the routing table, specifying distinct paths for data transfer. These routes remain
unchanged unless modified by the administrator.
Advantages
1. Static routing is straightforward to set up and comprehend, making it well-suited for smaller or less
complex networks.
2. As the routes are configured manually, the network paths are consistent and change only when an
administrator intervenes.
3. Static routing does not broadcast routes across the network, enhancing security and lowering the
risk of certain attacks.
4. It doesn’t demand the computing and storage resources that dynamic routing protocols require.
Disadvantages
1. Manually updating routes can be cumbersome, which can make static routing unsuitable for large
or frequently changing networks.
2. Static routes lack the ability to adapt to network changes or failures, potentially leading to outages
when static routes become inactive.
3. Any required network alterations necessitate manual updates, which can be time-consuming and
prone to errors.

532
ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780
Vol.9 (1), January (2019), pp. 532-538
Online available at zenithresearch.org.in
Applications
Best suited for small or uncomplicated networks with infrequent routing alterations, particularly in
instances where security and control are prioritized over flexibility.
Dynamic Routing
Dynamic routing employs algorithms and protocols to automatically refresh routing tables based on
modifications in the network structure. Routers collaborate by sharing information to determine the
optimal data routing paths. Dynamic routing predominantly serves as a backup route in expansive
networks.

Advantages
1. Dynamic routing is ideal for large, intricate networks since it automatically adjusts to alterations.
2. It swiftly responds to network failures and changes, ensuring uninterrupted data flow.
3. Once set up, dynamic routing protocols oversee route updates, minimizing the need for manual
oversight.
4. These protocols use algorithms to identify the most effective transmission routes for data.
Disadvantages
1. Configuring and comprehending dynamic routing protocols is generally more intricate than with
static routing.
2. It requires additional CPU and memory capacity to operate the routing algorithms and maintain up-
to-date routing tables.
3. Dynamic routing protocols can face vulnerabilities to specific attacks, like route poisoning,
particularly if not secured correctly.
Applications
1. Large, complex networks with constantly changing topologies.
2. Networks that demand high reliability and availability.
3. Situations where selecting the optimal routing path is crucial for maintaining performance.
Comparison between Static routing Vs Dynamic routing
Feature Static Routing Dynamic Routing
Configuration Manual Automated using routing protocols
Scalability Limited High
Adaptability Low High
Resource Usage Low Higher (requires CPU and memory)
Complexity Simple Complex
Maintenance Manual updates needed Automatic updates
Security High (not advertised) Can be lower if not secured properly
Dynamic path selection based on
Optimal Path Fixed paths
algorithms

2. Distance segmentation and distance vector link routing

Routing protocols are important in determining the best way for data to travel through a network. The
two main types of routing methods are distance vector routing and link-based routing. The two
methods are different and are used in different network situations.
Distance vector routing protocols determine the best route to a destination based on distance.
Each router has a routing table that lists all possible destinations on the network and the number of
hops (distance) to reach each destination. Routers share routing tables with their neighbors.
Key features
1. The metric used is the hop count, which indicates the number of routers a packet must traverse to
reach its destination.
2. A router broadcasts its entire routing table to directly connected neighbors.
3. Bellman-Ford algorithm is a commonly used algorithm in distance vector routing protocols.

533
ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780
Vol.9 (1), January (2019), pp. 532-538
Online available at zenithresearch.org.in
4. Convergence can be slow, especially on larger networks, and can easily cause routing loops and the
"count to infinity" problem.
Advantages
1. Distance vector routing techniques are relatively easy to implement and configure.
2. Less computational and memory overhead due to the number of hops considered.
Disadvantages.
1. Due to slow convergence and possible routing counters, it is not suitable for large networks.
2. No routing problems and endless computation problems may occur, eventually delaying network
convergence.
3. Frequent updates can result in unnecessary bandwidth usage, especially in larger networks.
Example. RIP (Routing Information Protocol). It is measured by the number of hops, the maximum
hops is 15, which reduces the network scale.
Link-state routing protocols use a complete view of the network topology to determine the
best route. Each router creates a network map by gathering information about all the routers and
connections in the network. This information is used to calculate the shortest path to each destination
using algorithms such as Dijkstra's algorithm.
Key features
1. Various metrics are available, including hop count, bandwidth, latency and load.
2. A router communicates about direct connections (connection status) with all other routers on the
network.
3. Dijkstra's algorithm is used to calculate the shortest path to each destination. 4. Inversion This
method is faster and more reliable than distance vector recognition.
Advantages
1. Due to the speed of convergence and the ability to process more signals, it is suitable for large and
complex networks.
2. It is less vulnerable to routing counters and provides a more accurate and reliable view of the
network topology.
3. Send updates whenever the network topology changes, thus reducing unnecessary bandwidth usage.
Disadvantages
1. Implementation and configuration are more complex than distance vector protocols.
2. Because of the need to maintain a complete network topology, more computing resources and
memory are required.
Example
1. OSPF (Open Shortest Path First)Uses connection information to build a network graph and
calculate the shortest path to each destination.
2. IS-IS (Intermediate System to Intermediate System)Similar to OSPF, used in large enterprises
and ISP networks.
Comparison between Distance Vector Routing Vs Linking Static Routing
Feature Distance Vector Routing Link State Routing
Metric Hop count (typically) Various (hop count, bandwidth, delay)
Routing Updates Periodic broadcast to neighbors Triggered updates, exchanged with all
Algorithm Bellman-Ford Dijkstra
Convergence Slower, subject to loops Faster, more reliable
Scalability Limited, not suitable for large nets High, suitable for large and complex nets
Complexity Simple to implement More complex
Low computational/memory Higher computational/memory
Resource Usage
requirements requirements
Stability Prone to loops and count-to-infinity Stable, accurate network view

3. Adaptive vs. Non-adaptive Routing

534
ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780
Vol.9 (1), January (2019), pp. 532-538
Online available at zenithresearch.org.in
Routing is an important part of network design and management because it determines the optimal
path for data to travel through the network. Routing algorithms can be broadly categorized as adaptive
(dynamic) routing and non-adaptive (static) routing. Each type has different characteristics and
advantages and disadvantages.
Adaptive Routing (Dynamic Routing)
Adaptive routing algorithms adapt the path of data packets to the current network conditions in real
time. These routes can change dynamically in response to changes in network topology, traffic load,
and other factors.
Features
Real-time adjustments. Routes are recalculated based on real-time network conditions such as
congestion, outages, and utilization.
Network awareness. Routers share information about the status of the network, allowing dynamic
route adjustments.
Algorithms. Common algorithms include link-state and distance-vector protocols.
Protocols Examples include OSPF (Open Shortest Path First), EIGRP (Enhanced Interior Gateway
Routing Protocol), and BGP (Border Gateway Protocol).
Advantages
Flexibility. Adapts to changes in the network such as B. Prevents connection failures and overloads
and ensures efficient data transmission.
Resilience. Provides better reliability and robustness by finding alternative routes in case of failures.
Efficiency. Optimizes network performance by balancing loads and avoiding congested paths.
Disadvantages
Complexity Its dynamic nature and continuous updates make it more complex to implement and
maintain. Resource intensive More processing power and storage are required to process routing
updates and maintain current network status.
Latency Route recalculation and frequent updates can cause latency, especially in large, complex
networks.
Applications
1. Large, dynamic networks with frequent topology changes.
2. Networks that require high reliability and performance optimization.
3. Real-time applications with changing network conditions.
Non-adaptive routing (static routing)
Non-adaptive routing algorithms use predefined, fixed paths for data packets. These routes are
manually configured by the network administrator and do not change unless they are manually
updated.
Characteristics
Fixed routes.Routes are static and do not adapt to network changes.
Manual configuration. Network administrators manually configure and update routes.
Simplicity. Easy to implement and understand due to robustness.
Advantages
Simplicity. Easy to implement and manage, especially in small or stable networks.
Predictability. Routes do not change automatically, so they are predictable and consistent.
Low resource consumption. Routes are fixed, so minimal processing power and memory are
required.
Disadvantages
Lack of flexibility. Inability to adapt to network changes such as outages and congestion can lead to
inefficient routing and outages.
Maintenance overhead. Any network changes must be updated manually, increasing management
overhead.
Scalability issues. Manual administration requirements make it impractical for large networks or
networks that change frequently.
Applications
1. Small, simple, and stable networks that change infrequently. Environments where security and
control of routing paths are most important.

535
ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780
Vol.9 (1), January (2019), pp. 532-538
Online available at zenithresearch.org.in
2. Backup routes for networks that primarily use dynamic routing.
Non-Adaptive Routing (Static
Feature Adaptive Routing (Dynamic Routing)
Routing)
Real-time, based on network
Route Adjustment Fixed, manual configuration
conditions
Flexibility High, adapts to changes Low, does not adapt
Network
Exchanges real-time information No real-time awareness
Awareness
Complexity High, due to dynamic nature Low, simple implementation
High, needs more computational Low, minimal resource
Resource Usage
power requirements
Lower, susceptible to network
Reliability High, can find alternative paths
changes
Maintenance Automated updates Manual updates required
Suitable for large and dynamic
Scalability Limited, not ideal for large networks
networks

4. Centralized vs. Distributed Routing

Routing is a fundamental aspect of network management that determines the best path for sending
data throughout the network. The methods for managing and determining routing information can be
categorized into two main typescentralized routing and distributed routing. Each approach has
different characteristics, advantages, and disadvantages.
Centralized Routing
In centralized routing, a single central node or controller makes all routing decisions for the entire
network. This node collects comprehensive information about the network topology and traffic
conditions and uses this information to calculate the optimal path for data packets.
Features
Single point of control A central node or controller manages routing decisions.
Global network viewsThe central node has a comprehensive view of the entire network, including all
nodes and links.
Algorithms Centralized algorithms such as Dijkstra's algorithm and Bellman-Ford can be used to
calculate routes.
Advantages
Optimal routingWith a global view of the network, the central node can calculate optimal paths,
which can lead to improved performance across the network.
Simplified network management Centralized management simplifies the management and
configuration of routing policies and updates.
Consistency Ensures consistent routing decisions across the network, reducing the chance of routing
loops and conflicts.
Disadvantages
Single point of failure A central node can affect the routing capabilities of the entire network if it
fails.
Scalability issues Due to the processing and communication overhead of a central node, central
routing may not scale well for very large networks.
Latency Routing decisions can cause delays because all traffic information must be sent to a central
node for processing.
Applications
1. Small to medium-sized networks where a central controller can effectively manage routing
decisions.
2. Networks where optimal routing and centralized management are important, such as Data centers
and enterprise networks.

536
ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780
Vol.9 (1), January (2019), pp. 532-538
Online available at zenithresearch.org.in
Distributed routing
In distributed routing, each router in the network makes its own routing decisions based on the
information available locally and the information exchanged with neighboring routers. There is no
central authority. Instead, routing information is decentralized and distributed throughout the network.
Characteristics
Distributed control Each router makes routing decisions independent of each other.
Local network view A router has a partial view of the network, usually limited to its neighbors.
Algorithms Distributed algorithms such as distance vector (e.g. RIP) or link state (e.g. OSPF) are
often used.
Advantages
Scalability Distributed routing is suitable for large networks because the processing load is
distributed across all routers.
Resilience There is no single point of failure. The network continues to function even if an individual
router fails.
Reduced latency Because routing decisions are made locally, the time required to calculate a route
may be reduced.
Disadvantages
Suboptimal routes Without a global view, routers may make suboptimal routing decisions, resulting in
less efficient paths.
Complexity Increased complexity as routers must exchange information and maintain up-to-date
routing tables.
Inconsistency Due to its distributed nature, temporary inconsistencies and routing loops may occur
during updates and convergence.
Applications
Scalability and resilience are important in large, dynamic networks like the Internet.
Networks that favor decentralized control and local decision-making.
Feature Centralized Routing Distributed Routing

Control Single central node or controller Decentralized, each router

independently

Network View Global, comprehensive view Local, limited to neighboring routers

Failure Impact Single point of failure No single point of failure

Scalability Limited scalability High scalability

Latency Higher due to central processing Lower due to local decision-making

Routing Potentially more optimal routes Potentially less optimal routes

Efficiency

Management Simplified centralized management Complex, requires distributed

coordination

Consistency High consistency Can have transient inconsistencies

Conclusion
Dynamic routing provides adaptability, scalability, and optimal path selection at the cost of increased
complexity and resource usage. The choice between static and dynamic routing depends on the
specific requirements and characteristics of the network. Distance vector protocols are simple and
have low overhead, making them suitable for small, simple networks. Link state protocols, on the
other hand, provide faster convergence, scalability, and a more accurate view of the network
topology.

537
ZENITH International Journal of Multidisciplinary Research ____________ISSN 2231-5780
Vol.9 (1), January (2019), pp. 532-538
Online available at zenithresearch.org.in

These protocols are suitable for large and complex networks but come with increased complexity and
resource requirements. The choice between these routing protocols depends on the specific needs and
characteristics of the network. For larger, more dynamic networks, link state protocols like OSPF and
IS-IS are more appropriate. Adaptive routing offers flexibility, reliability, and efficiency by adjusting
to real-time network conditions, making it ideal for large and dynamic networks.

Non-adaptive routing is simple and predictable, with minimal resource usage, making it suitable for
small, stable networks where changes are infrequent. The choice between adaptive and non-adaptive
routing depends on the specific needs of the network, including size, complexity, stability, and
performance requirements. Centralized and distributed routing each have their own benefits and
drawbacks, making them suitable for different network scenarios. Distributed routing offers
scalability and resilience, making it suitable for large, dynamic networks like the Internet.

The choice between centralized and distributed routing depends on the specific requirements and
characteristics of the network, including size, complexity, performance needs, and resilience
considerations.
References
1. Tanenbaum, A. S., & Wetherall, D. J. (2011). Computer Networks (5th ed.). Pearson.
2. Cisco Systems. (n.d.). Routing Protocols. Retrieved from Cisco Routing Protocols
3. Kurose, J. F., & Ross, K. W. (2017). Computer Networking: A Top-Down Approach (7th
ed.). Pearson.
4. Stallings, W. (2013). Data and Computer Communications (10th ed.). Pearson.
5. Static Routing is simple and predictable but lacks flexibility and adaptability, making it
suitable for small, stable networks.
6. Dynamic Routing offers adaptability and fault tolerance but comes with increased complexity
and overhead, ideal for large, dynamic networks.
7. Distance Vector Routing is easier to implement but suffers from slow convergence and
potential routing loops.
8. Link State Routing provides accurate routing information and fast convergence at the cost of
higher computational requirements.
9. Adaptive Routing dynamically adjusts to network conditions, optimizing performance and
resilience.
10. Non-Adaptive Routing is simple and predictable but inflexible and unable to respond to
network changes.
11. Centralized Routing can compute optimal routes but introduces a single point of failure and
scalability concerns.
12. Distributed Routing offers scalability and fault tolerance but may result in higher
communication overhead and potential suboptimal routes.
13. Understanding these characteristics helps network administrators choose the appropriate routing
algorithms based on the specific needs and conditions of their networks.

538