Network topology can be defined as the way of how to arrange and map the elements of a network, especially the physical and logical interconnections between nodes. A LAN, is an example of a network that has both a physical topology but also a logical one. Each node from the local network has one or more links to one or more nodes in the network. In order to determine the physical topology of the network, all nodes and links must be represented in graph form. Also, through the representation of the data flow between nodes in the form of a graph we determine the logical topology of the network.
For any given network, the logical and physical topology may be identical, but they can also be different. In general, local network technology is based on graph theory.
Types of topologies:
– physical topologies – it deals with the spatial aspect and the physical organization of the network stations and the cables
– signal topologies
– logical topologies – refers to the way in which the communication in the network is realized, to how the data flows between stations.
The topology of a network directly affects its performance, so choosing a topology over others influences:
– the type of equipment needed
– the fundamental equipment characteristics
– the circumstances about expanding the network
– the process of managing the network
1. Point-to-Point Topology
The simplest topology in this category is a permanent link (or link) between any two endings. Switched point-to-point topologies are the basic models of ordinary telephony. The definitive value of point-to-point networks is a guaranteed value between the two endings. The value of on-demand point-to-point connections is proportional to the number of potential subscriber pairs and has been expressed in Metcalfe’s Law.
2. The Bus topology
The type of network topology in which all the nodes of the network are connected to a common transmission medium that has exactly two terminations. All the data that is transmitted between the nodes in the network is transmitted during this common part of the transport and the environment so as to be received by all nodes in the network, almost simultaneously, regardless of widespread delays.
The two terminations that are part of the communal transport bus are normally stopped, with a device called terminal. The respective device absorbs the energy that remains in the signal, thus preventing the reflection or propagation of the signal in the opposite direction, which may cause interference or even lead to signal degradation.
Bus topologies are the easiest way to connect multiple clients, but they often have problems when two clients simultaneously want to transmit data on the same bus. Thus, systems that use bus network architecture have designed some schemes to avoid data collisions on the common bus, such as the Carrier Sense Multiple Access method that controls the shared resources of the common bus is most often used.
Carrier Sense Multiple Access (CSMA) is a Media Access Control (MAC) protocol in which each node, before transmitting the information on the common bus, tests the presence of other traffic from the common transmission environment.
Advantages of the bus topology:
– Easy to implement and extend
– Requires less cable length than star networks
– They are well adapted for temporary and small networks that do not require high speeds, in addition, they are easy to configure.
– They are less expensive because only one cable is used.
Disadvantages of the Bus Topology:
– The length of the cable is limited and so are the number of stations
– If there are problems with the cable, the whole network “crashes”
– Maintenance costs can be high over a long period of time
– Performance degrades if too many computers are connected
– The correct termination of the signal (and the cable) is required
– Significant loading capacity (each transaction must reach its destination)
– Works better with a limited number of nodes
– It is slower than other topologies
– If a computer fails then the whole network will “crash”
3. The Star Topology
It is the type of network topology in which each of the network nodes is connected to a central node, called a hub or switch. All data that is transmitted between the nodes in the network is first transmitted to this central node and is only then relayed to some or all other nodes in the network. This centralized connection allows a permanent connection even if a device is disconnected from the network. The only danger is the removal of the central node, which would lead to the loss of the connection with the entire network.
– Increased performance: Passing data packets through unnecessary nodes is prevented by this topology. This topology in itself induces a large load on the central node, however, if this node has the respective capacity, then intensive use by one device in the network will not affect the other devices in the network.
– Device isolation: Each device is inherently isolated by the link that connects to the central node. This makes isolating individual devices quite easily, and allows them to be disconnected at any time from the central node. This isolation procedure prevents any non-centralized failure that will affect the entire network.
The Extended Star topology shows the multiplication of the central nodes, allowing the work of the network even if one of the central nodes fails. It has a higher performance than the Star topology.