Three types of network topology

Bus Topology – This topology consists of a Backbone cable connecting all nodes on a network without intervening connectivity devices. The single cable is called Bus and can support only one channel for communication; as a result, every node shares the bus’s total capacity. On a Bus Topology network, devices share the responsibility for getting data from one point to another. But only the intended recipient actually receives and processes the transmitted message while a device wants to send the broadcast message to all the devices that connected to the shared cable.

At the ends of each bus network are 50-ohm resistors known as terminators. The terminators stop signals after they have reached the end of the cable. Without these terminators, signals on a bus network would travel endlessly between the two ends of the network; this phenomenon is called signal bounce, and new signals could not get through.

Ethernet bus topology is actually easy to install and does not require much cabling and only a main shared cable is used for network communication. 10BASE2 and 10BASE-T are two popular types of the Ethernet cables used in the Bus topology. Also, Bus network works with very limited devices. Performance issues are likely to occur in the Bus topology if more than 12-15 computers are added in a Bus Network. In addition, if the Backbone cable fails then all network becomes useless and communication fails among all the devices.

Ring Topology – In ring network, each node is connected to the two nearest nodes so that the entire network forms a circle. In a ring network, all the communication messages are transmitted clockwise in one direction, around the ring. Each workstation accepts and responds to packets addressed to it, and then forwards the other packets to the next workstation in the network. Each workstation just acts as a repeater for the transmission, in other words, all workstations participate in delivery makes the ring topology an active topology. A ring topology has no “ends” and transmitted data stops at its destination. Twisted-pair or fiber-optic cabling is commonly used as the physical medium.

Any damage of the cable of any cable or device can result in the breakdown of the whole network. In addition, just as in a bus topology, the more workstations that must participate in data transmission, the slower the response time. Due to these shortcomings, ring topology now has become almost obsolete. FDDI, SONET or Token Ring Technology can be used to implement Ring Technology. Ring topologies can be found in office, school or small buildings.

Star Topology – In the computer networking world the most commonly used topology in local area networking is the star topology. Star topologies can be implemented in home, offices or even in a building. All the computers in the star topologies are connected to central devices like hub, switch or router. The functionality of all these devices is different. Computers in a network are usually connected with the hub, switch or router with the unshielded twisted-pair (UTP) or Shielded Twisted-pair cables. Star topology networks can support a maximum of only 1024 addressable nodes on a logical network.

Star topology requires more cabling than ring and bus. It also requires more configurations. However, because each node is separately connected to a central connectivity device, they are more fault-tolerant. A single malfunctioning workstation cannot disable an entire star network. A failure in the central connectivity device can take down a local area networking segment. Because they include a centralized connection point, star topology can easily be moved, isolated or interconnected with other networks through hubs, switches to form more complex topologies. Most Ethernet networks are based on the star topology.

Carrier Sense Multiple Access/Collision Detect (CSMA/CD)

Carrier Sense Multiple Access/Collision Detect (CSMA/CD) is a scheme where the hosts try to detect collisions, and if they occur, stop transmitting, and then try to re-transmit at a later time. Any host on the network segment that wants to send data must check the segment to see if another host is already sending data. This is called “listening to the wire”, and this is the Carrier Sense part of CSMA/CD. All the hosts on the network segment will do this – that is the Multiple Access part.

Collision Detection, which can be referred to as a host will be able to quickly detect any collision that its data is involved in. If two hosts listen to the wire at the exact same time, determine that no one else is sending data, and then send their data onto the segment at the same time, a collision will occur.

While collision happens, is actually a change in the voltage of the signals on the wire. The sending hosts detect this voltage changes and basically would know there is a collision, and both of the senders will send a jam signal over the network. This jam signal indicates to all other hosts on the network segment that there has been a collision, and they should not send data onto the wire.

After jam signals have been sent, each of the hosts will be waiting a random amount of time before listening to the wire again. The system invokes random time to help to ensure that the two hosts do not transmit data simultaneously again.

Difference between a hub and a switch

Hub repeats everything it receives and can be used to extend the network. However, this can result in a lot of unnecessary traffic being sent to all devices on the network. Hub passes on traffic to the network regardless of the intended destination; the workstations to which the packets are sent use the address information in each packet to work out which packets are meant for them. In a small network repeating is not a problem but for a larger, more heavily used network, another piece of networking equipment (such as a switch) may be required to help reduce the amount of unnecessary traffic being generated.

A hub transmits all data to all hosts

Switch is used to help reduce the amount of unnecessary traffic being generated. Switch controls the flow of network traffic based on the address information in each packet. A switch learns which devices are connected to its ports (by monitoring the packets it receives), and then forwards on packets to the appropriate port only. This allows simultaneous communication across the switch, improving bandwidth. This switching operation reduces the amount of unnecessary traffic that would have occurred if the same information had been sent from every port (as with a hub).

Switch only transmits data between communicating hosts

Switch and hub are often used in the same network; the hub extends the network by providing more ports, and the switch divides the network into smaller, less congested sections. Another advantage with using a switch is that it is possible to mix devices that communicate with different speeds, so one of the workstations connected to the switch could be running 10MB Half-duplex and another 100MB Full-duplex.