In this assignment I am going to list and describe the components that make up a LAN. I will also state how they can be interconnected and interact with each other. I will show some LAN topologies and illustrate them. Finally I will explain and describe how data travels around the various topologies and list the advantages and disadvantages of each topology.
I am going to look at the components that make the LAN and describe them. Ethernet, Hardware, NIC’s, Hubs/Switches/etc these are the list of components that make up the LAN and now I am going to explain each of these components.
NIC’s – Network Interface Card are used for connecting a RJ-45 cable. Nearly all the modem PC’s or PC’s which connect through dial-up connection need a NIC. On either ends is a RJ-45 connector which then allows you connect to another computer on the network or another device. The NIC is interconnected to other hardware such as the motherboard; this allows it to communicate with the computer. There are 3 ways the NIC can be connected to the computer. As mentioned before it can be integrated on the motherboard, connected through a PCI slot or connected using a ISA slot (Industry Standard Architecture).Order now
For the NIC’s to communicate with each other properly they need to have been connected with a RJ-45 connector on either end to one of the computers. The computer need to be switched on as well, this will allow the computer to send broadcast message to the computer.
Gateway – It is a network point which allows you to enter another network. Gateway also allows you to connect LAN or a WLAN to the internet or another WAN. The main role of a gateway is to convert protocols among communications networks. This means that if a computer receives a packet from Apple Talk the gateway will then convert to any format which it needs to send it to whether it is a TCP/IP, OSI or other.
Token Ring – Works at the data link layer protocol in a LAN. A token travels around a ring and when it gets the data to the specified computer it will go back to the original computer with a signature and then the data, with a signature and then it is destroyed, this means there is a free token which can be used. Data is transmitted via frames such as the addresses that are attached to the frame header and trailers. As soon as the frame has reached the specified computer the data is copied over.
Servers – These are high-performance computer which serve information to other computers. Also known as clients can connect to other servers on the internet via a LAN or a WAN network. There are lots of different servers which are hosted and can be connected to via the internet, normally they are used as web server, mail server or mail server using a STMP protocol. A web server is used to host a FTP/HTTP site which allows user to access information and download data if the server is not restricted to allow this to happen. On the other hand a mail server is used to store received and sent.
A file server is completely different to the previous one’s because this actually is made to only store files and let users download them at very high speeds. Any normal server can be transformed into a file server. Indeed a file server is mostly used in a private LAN network, which is probably password protected so user’s from outside the network cannot hack it. Ethernet cables are normally used to connect servers to a network. Servers work at application layer which is layer 7 in a OSI model and layer 4 in an TCP/IP model.
Workstations – A relatively high quality graphics quality and high power computer which can be used for modelling and simulation, engineering application (CAD/CAM), desktop publishing. A workstation can run multi-user operating systems. Workstations are connected to the internet via a router. As described in previous assignments a router is a device which lets you connect computer together. It routes where the data goes. It works at layer 3 of an OSI model.
Routers – It decides what path should be taken when the data is relayed on the network. This is done by using something called routing protocol, this helps it learn the network and it uses the algorithms, which helps the router understand the best route on its routing metrics. Router works at the Network layer of a OSI model which is the third layer. Router’s basic role in a network is to connect them together. The router does this by using the destination address which is integrated inside the network layer protocol. It will find it by stripping of the data link layer protocol from which it receives through one of its ports.
This will also help it know where to send the next packet. The data from the network layer protocol will be encapsulated into a new data link layer frame before it transmits or sends it. Packet can be directly transmitted by the router. Metric will be based either on its reliability or it be time based. The routes are regularly updated as a part of a packet switching solution two following packets may travel completely different routers. If you have a mobile device or a portable which has a NIC or a wireless connection maker than you connect your device directly rather than configuring it to work with the router.
Repeaters – A piece of device which has transmitter and receiver which allow the amplification of a signal and transmission of a signal. Repeater extends the length of the cable and also the physical size of a network. It extends the cable and network by having longer cables. Attenuation is a problem when the data travels longer distance because it reduces the strength of the cable.
Hubs – Hubs are used to connect twisted pair or fibre optic with Ethernet devices together which can be used as a single network segment. Hubs work at the physical layer which is layer 1 of an OSI model. Hubs strengthens signal if it detects them coming from one of it incoming ports, after it strengthens the signal it will send it back out from the forwarding ports. Hubs do not manage traffic on the network, but it can broadcast any packets that it receives from its incoming ports. Ethernet cable is connected to the hub and a cable from the hub is connected to a workstation or a server for the LAN to work.
Bridges – Similar to a switch a bridge allows two different network segments to be connected together. Mostly the same communication protocol is used. It also manages and decides which packet of data should be transported from one segment to another segment by analyzing the MAC address. A wireless bridge is also available which allows extension of wireless networks. Bridge also allows various networks to connect together on the data link layer.
Switch – Switches allow segments of network to link together. Switch filters and forward packets to various different segments of a LAN. Network performance is increased by reducing competition for available bandwidth by segmentations. Working at the network layer which is layer 3 in a OSI model, when the switch receives a signal it will define it as a data packet it will read the destined address from the data link layer protocol header and it will forward it to the computer. This is actualized by reducing the traffic and collisions on the network.
Token passing is the method used by nodes to send and receive data in a network. In token passing a node takes possession of a small frame called a token and attaches the data, control information and the specified target. Then the sending node will send it off, if a computer not targeted receives the token it will simply forward it off and when the target receives the data it opens the frame and reads the data and then send of the receipt of acknowledgement to the computer then transmitted the data and the token is destroyed and a new free token is created.
Network topologies are the way that nodes on a network are arranged. Topologies define how they are interconnected not only logically but physical as well, but mostly people concentrate on the logical part rather then physical part. An example of a network which normally concentrates on both is LAN the Local Area Network. In a topology a given topology except a Ring topology it will normally consist of two or more nodes connected to it. In a Ring topology each node is connected exactly twice.
Bus Topology – In a bus network the clients are connected via a shared communication line which is known as a bus. It is one of the easiest way of connect multiple clients. This topology consists of a main cable which either can be a twisted pair, coaxial or a fibre optic cable running through the network with a connector at each end of the cable. When two devices are transmitting at the same time results in collision on the network. When data is transmitted between two devices on the network it will also be broadcasted to other devices on the network virtually simultaneously.
* Easy to implement and extend, because it consist of a long cable running through the network.
* It is cheap to build because it requires a lot less cable to implement the topology. It is also good for temporary networks which need to be set up in a hurry.
* Transmission rate are relatively high compared to other network. The cable faults can also be easily identified.
* The data is transmitted to every computer when the data is just to be sent and received between two computers. Even though it is has a relatively high transmission rate it does not cope well with eavy traffic rates.
* There is a limited amount of cable and workstations that can be used and connected.
* If the network is cable is broken it can cripple the whole network, the computer will be functioning but cannot communicate with each other.
* Maintenance of the network in long period may be high, and also long term running more computers are added to the network the performance will drop.
Star Topology – Star topology is a central node with outlying nodes. In a star topology the messages are transmitted via a switch, hub or even a computer. The chance of being hit by a Network failure is quite low because all the computer systems are connected to the central node. Each of the devices connected have their own connection. The connections are made by network cable running from the central node directly to the device.
* Performance is a lot better because the data packets that are sent and goes through unnecessary node the star topology will block this from happening. Even if the central node has very high network utilization it will only affect the computer it is communicating with.
* Each device connected to the network is isolated; this prevents any non-centralized failure to affect other devices on the network.
* Expanding the network is very easy because of centralization. It can also analyze traffic and determine whether suspicious behaviour.
* This topology overall is very easy to understand, design and navigate. This topology discards the need of complex routing/message forwarding protocols
* There are setbacks of this type of topology because the network is dependent on the system functioning of the central hubs, If there is a problem with the central hub then the whole network will stop working.
* The whole network will also depend on the performance and capabilities of the central hub. The performance will depend on the number of connection that can be made simultaneously to the server.
* Centralization can be a drawback if it there is high utilizing of the network because it will result in performance drop in the traffic on the network.
Mesh Topology – In this type of topology each of the nodes are connected with each other. This means that the internet is the largest type of mesh topology. When the data is transmitted it will automatically know where the data is to be sent and it will take the shortest route.
* Even if there is a fault in the cable the traffic will still be sent because all the nodes on the network are joined together. The node can automatically reroute itself if the cable is broken because there is not gateway computer.
* Takes a long time to set up beach each node needs to be connected to each other, so data can flow around without problems.
* It requires a lot more cable than other types of topologies because all the node are connected, so the more nodes that are connected the more cable is needed.
* It can workout to be quite expensive to setup.
Ring Topology – The node are interconnected with each other like mesh but each node is exactly connected to two other node, which result in a pathway for signal that are transmitted. The data which is transmitted is handled by each node on the network and then sent back with an acknowledgement of receipt. When the node transmitting the data receives the receipt it destroys the token ring. Either a token ring or a small packet is continuously passed around the network so when a device needs to send something it will reserve the token ring.
* Network is not dependent on a centralized node because each of the nodes on the network can support the transmission rates around the network. Each of the nodes has the ability to take control of the transmission rate.
* Extremely high transmission rate can be achieved. Typically speed of transmission is around 10 Mbps which is quite quick. For e.g. it will take around 2 minutes to send a 100MB file which is pretty quick.
* Compared to a star topology it performs a lot better under heavy load.
* Network server is not required to manage the network.