About Project

SPIDER WEB TOPOLOGY

Project Proposed Title:

Spider Web Topology

What is Topology in Network?

Network Topology is the layout pattern of interconnections of various computers or the nodes and links of the computer. Network Topology may be physical or logical.

Why – Spider Web Topology?

Since, there are many types of topologies used in interconnecting the computers, but there are many disadvantages. One of the problems is yet to solve is to design a network topology which is a fail-safe network.

So we tried to come up with a simple solution for designing a fail-safe network which resembles the design of a Spider’s web.

What is Spider Web Topology? 

The layout pattern of a network which resembles a same design of the natural spider’s web is called as Spider Web Topology.

The Layout of the network in which many Secondary Servers connected in wheels are also connected with adjacent servers is called as Spider Web Topology.

How it works?

The main components of this web topology are,

1. Back-Up and Communication Server

2. Secondary Servers

All the servers are connected to the Back Up and Communication server. The first wheel of the network is directly connected to the Back Up and Communication Server. The Other wheels are connected to it indirectly through routers. The circular path of the network connection is referred as Wheel. Each wheel is connected with other wheel. The connection line which connects the each wheel to the Back Up and Connection Server is called as Spokes.

The network which connected between the two secondary servers has series of Sub-Local Servers. Each Sub-Local Servers has its own local network.

The secondary servers are connected by two lines. They are

1. Main Link

2. Subordinate Link

Main Links are used to form Communication Wheel for the network and it also connects the Sub-Local Servers with the Secondary Servers of its adjacent sides.

Subordinate Links are used to connect one Secondary Servers with other four servers which are adjacent to it.

The network communication is simulated by Routers and Switches.

Main Links are used to make the communication between the Secondary Servers and the Sub-Local Servers. So, Secondary servers have the knowledge about the Sub-Local servers which are connected adjacent to them. If failure occurs in a Secondary server, the communication link will not get broken since all the Sub-Local Servers which are connected with the failure Secondary servers are also connected with other Secondary servers. So the failure of the network is avoided.

Subordinate Links are used to connect the Secondary Servers without the Sub-Local Servers. This link is a special link for the network, in order to overcome the fault in the network. The Secondary Servers communicate with each other, so it is easy to find when a server gets failure.
This is how the network works.

Keys of Spider Web Topology:

The keys of this web topology are,
1. One Secondary Server will have the knowledge about the neighbor Secondary servers only.

2. But, each Secondary Servers can communicate with any Secondary Servers in the network.

3. If failure occurs in one Secondary Server, other Secondary Server which is directly connected with it, take care of that particular network.

Routing of the network:

 Routing of Spider Web Topology should be based on the above keys of the topology. Basically, there are two types of routing algorithm. They are,

1. Distance Vector Algorithm and

                                                                  2. Link-State Routing Algorithm.

But, the properties of the Spider Web Topology are satisfied by the Link-State Routing Algorithm.

Conditions for Link-State Routing Algorithm:
1. Knowledge about the neighbor.
2.  Routing to whole network.

3. Information sharing when there is a change.

1. Knowledge about the neighbor:
Each router has the knowledge about their neighbor routers.

2. Routing to whole network:
Each router sends the information to every other router on the network. This can be done by flooding process. Flooding is the process in which a router sends its information to all of its neighbors.

3. Information sharing when there is a change:
Each router sends out the information about the neighbors when there is a change in the network.

The above three conditions of Link-State Routing Algorithm satisfies the properties of the Spider Web Topology.

Packet Cost:
Since Link-State Routing is a Lowest-Cost Algorithm, cost of the packets are weighted value based on a variety of factors such as security levels and state of the link.

Link State Packet:
When a router floods the network with information about its neighbor, it is called as Advertising. This is called as Link State Packet.

Link State Packet Structure:

Advertiser	Network	Cost	Neighbor
………………………………	………………………………	………………………………	………………………………
………………………………	………………………………	………………………………	………………………………
………………………………	………………………………	………………………………	………………………………

The above table is also called as Database Tables of Routers.

Link State Database:
Each router receives Link State Packet and they are stored as information into a Link-State Database. Every router in the network creates the same database and they are used to calculate their routing table. If a router gets added or deleted from the network, the whole database must be updated and shared for fast updating.

Dijkstra Algorithm:
Dijkstra Algorithm calculates the shortest path between two points on a network using a graph made up of nodes and arc.

Nodes are referred to as Network and Routers. Arc are referred to as the connection between the network and router (i.e.) router to network and network to router. The cost of the connection is applied only to the arc from router to network. The cost of the connection from arc to network to router is always zero.

Shortest path:
The Dijkstra Algorithm follows four steps to discover the routing table for each router. Each router must have their own routing table.

1. The root of the tree is identified and algorithm begins to build tree.

2. The algorithm starts to compare tree’s temporary arc and identifies the arc with lower cost. The lower cost of the arc will be the permanent part of the tree.

3. The algorithm reads the database table and each nodes are identified that can be reached from the root node.

4. The last two steps are repeated until the network gains the nodes as the permanent part of the tree. Thus the permanent arc is the shortest path of the root node to the ending node.