In the actual sense topology refers to both the physical and logical layout of a given network. The physical topology of a network refers to the actual layout of the computer cables and other network devices. On the other hand, the logical topology of a network refers to the way in which the network appears to the devices that use it (David & Skandier, 2005). The physical aspects of any network normally depend on the physical transport technology involved. The choice of the topology depends since there are number of LAN technologies to choose from. Legacy products are currently no longer used although technologies such as ARC net are still in use. ARCNET uses a token-bus scheme in managing the sharing among the nodes and other network devices connected to the LAN (Designing a Network Topology, 2004).
Empty messages frames are circulated continuously on a bus by a LAN server. When a device in the LAN wants to communicate it inserts a "token" in an empty frame, and at the same time it inserts the message in the frame. Once the targeted device reads the message, it does a reset to the token to 0 to make sure that the frame can be used by another device on the LAN. This scheme is very efficient especially when the traffic is high since all the devices are accorded same opportunity to use the shared network resources. ARC net is one of the major four LAN technologies inclusive of Ethernet, Token ring and FDDI. TCP/IP is the typical from which the worldwide internet down to the LAN. In addition, it is a communication protocol used to define how components in a network (computers) should be connected to the internet as well as how the data should be transmitted between these devices/nodes.
For a good network design the strategy and planning components are very vital. The overall project needs to be implemented in an orderly manner in order to achieve the goals set. Policies -statements about how something is going to be used- should be put in place when designing a network (Network Diagrams, 2010). Policies detail how the network resources are to be used by the users concerned. For instance, you might discover that the bandwidth of your network is underperforming because users are using the connection for no business needs. Another scenario that policies become a necessity is at a point is when you use a firewall as a security measure.
The security policy is well implemented by the use of the firewall technology. For instance if someone is not aware of the traffic that is supposed to go through the firewall then setting up one can be a difficult task in some instances because most of the secure websites prohibit the users in the internet to use for instance telnet- the standard application. All this have to be checked. Technical users responsible for managing the network require adequate training on procedures and the cost of training can be foregone if it is an in-house -conducted by the personnel who are familiar with the technology locally. It's important to keep in mind that technical staffs that will support the network are dependable and they should have adequate knowledge. When designing the network, the cost of the budget should also be a pre-requisite consideration and the overall benefits should always outweigh the costs. Besides, the cost should be within the company's financial obligations (Network management reference architecture, 2010).
To design a logical network, there are two options; one can either design or install a new network from the scratch or upgrade an existing an already existing logical network. However, a number of factors are supposed to be considered before commencing the design work. They include among others: What services will you be providing to the clients/users? What possible network traffic patterns that crop up as a result from your particular plan? What are the expected bottle necks and how to avoid? Another important consideration is to determine whether your internet connection is going to be for internal network users only or if outside users' domain will be given access to the network resources (Network Tutorials, 2010).
There exist a great number of network topologies that can be used which includes bus, hybrid, linear, mesh, ring, star and tree (Scots & Ogletree, 2003). Different topologies differ in terms of physical connections, transmission rates and the types of signals. Data transmission is one of the most distinguishable issues for any topology (Network Topologies, 2010). The following paragraphs explain in a nutshell the different kinds of topologies that exist and offers the advantage and disadvantage involved (Siamwall & Keshav, 2000). Bus network: - this uses a trunk or rather a backbone to which all the PCs are connected to. The PCs connects to this said backbone through T connector or taps. To avoid a reflection of signal the physical bus topology requires termination of each end of the physical bus. One of the most common implementation of this topology is the IEEE 802.3 -is a standard that establishes the physical and data links layers media access control of a wired Ethernet and it is usually a Local Area Network technology but borrows some application s from the Wide Area Network (Raza, 2002).
Advantages
It's cheap and easy to implement
Fewer cables are required as compared to other technologies
Disadvantages
Network disruption results if a computer is added or removed to/from the network
Since all the systems on the network are connected to a single backbone, a break in the cable normally results in preventing all the systems from accessing the network.
The scalability and extensibility of are important factors to consider and thus a number of requirements have to be looked upon in any network. Network design -whether logical and physical- involves meeting certain standards while at the same time working within the reach of certain determinants (Raza, 2002). These may include: Economic constraints -in most cases one has to compromise the capability of Wide Area Network links, the switching routers' capabilities, the kind of interfaces used, and at some instances the of redundancy achieved; Labor -in either of the network design, the degree of effort and the level of knowledge required to admit a new client to the new network and/or expand the capacity of the network infrastructure are of great concern (Craft, 2006).
Johnson (2002) in his research work states that the labor costs may be reduced by investing in network management tools; The processor speed increases on regular basis and therefore makes the internet traffic levels to rapidly increase. Computational limitations that most often engulf any particular network design are correlated to processing of route updates, security filtering, encryption, translation of address and packet forwarding; Space consists of the value of air-conditioned points of references (POPs) an/or the collocation facilities (Scots & Ogletree, 2003). Besides, it also entails the subtler for instance the Wide Area Network link band width; Time-to-market is another issue to be considered as while designing a network, clients might be have looked for alternatives by the time it becomes operational (Mueller & Ogletree, 2003).
As earlier mentioned, scalability and extensibility of the network design are important factors to consider (Thomas, 2008). They are the major determinants of a good network design and that is why the network routing is crucial in the design process. Other issues involved includes, the engineering cycles, an overly zealous requirement definition phase among others.
Since planning and strategy are important elements in network design, this paper then forms an important basis for the development of a logical or physical network that apart from being effective, it's reliable. This paper has successfully analyzed the different components of logical network design representing IP structures of the network such as Class A, B, or C address scheme. It has also addressed the components of physical network design including topology design such as Fiber, ISDN and Ethernet. The paper has successfully identified various factors revolving around network design so as to develop a network that is compliant to IT standards and that it remains critical all year round as it provides network redundancy from both a logical and physical perspective.
It is also evident from the research work that the process of designing a network requires a handful of necessities such as: internet connectivity, TCP/IP addressing through the use of IP subnets, the perimeter of the network and finally the remote access infrastructure to be put in place for easy accessibility. This would a long way in offering a network developer with the best working environment. Scalability and extensibility during the process of network design are key elements. The effective network design process would however be limited by factors such as social-economic concerns, labor, time and space. The impact that network design process can have on the daily operations of a reliable network is thus undeniable (Programming Languages: Logical and Physical Network Design, 2004).
