Assignment Help Hub | Assignment Help at 8.00 AUD/page ...



Table Of ContentsExecutive Summary Introduction Analysis Of User Requirements Description And Analysis Of Current Setup New Design For The SCU Campus Core Layer Distribution LayerAccess LayerConsiderations When Implementing The Hierarchical Network Model Detailed Network Design For SCU Campus Core Layer Configuration Distribution Layer Configuration Access Layer Configuration DNS Server For Blocking Restricted Websites Conclusions ReferencesExecutive Summary SCU now wants to upgrade their network infrastructure, after buying an old school that possesses an old-school type of network set-up. This network has to be replaced, owing to the fact that there is absolutely zero connectivity between the buildings, and the actual SCU buildings that are already in existence. This would mean that the students wouldn’t be able to share any information between the buildings via the network infrastructure. Therefore, a new network design is highly critical for the new building. Keeping in mind, the fact that these SCU buildings would meet the needs of a number of different kinds of users, it would be important to thoroughly realize the requirements of various communities of users. The new network design needs a great deal of performance, owing to the fact that different buildings have diverse requirements. In the first building, say, for the programmers, computers that run on a high-end operating system would be required. The present setup lacks any modern-day networking standards, as it does not offer any wireless connectivity and doesn’t interconnect each building. Each building comprises 30 computers. No buildings are connected to the main SCU buildings, which are situated 200 km away from the new campus. In this case, a hub/switch network is in existence, which is not efficient in nature. Hubs are often classified as Ethernet repeaters, have been in use for a great deal of time. They were one of the initial techniques utilized for connecting computers. A hub works at the physical layer. Hubs are often classified as dumb devices, owing to the fact that they do not have any forwarding or memory tables. However, utilizing Microsoft Server 2008 would cause a number of problems. The first thing to consider is that Microsoft’s official website states that mainstream support for windows 2008 server will finish in the year 2015. In physical servers, the physical hardware would be available for five years, before a hardware refresh. This would mean that the product support will end before upgrading the hardware. This demands a resilient network design, where every part of the network is connected i.e. every computing node is connected in a secure manner, and for improved performance, along with each building connected to provide communication. The new proposed design for the SCU campus demands a complete change in the manner the network is developed. The older design offered absolutely no performance and did not allow addition of nodes to the network. When it comes to developing a new design for the SCU campus network, the business goals of SCU are of vital importance and would form the basis for the selection of network model. The business goals of SCU could be explained in few simple words i.e. they want a network capable of supporting a greater number of users, and their diverse requirements. The business goals, along with the technical goals, would help to determine the suitable model for network to be designed. Keeping these factors in mind, the model chosen for the development of SCU’s new campus network would be the three-layer hierarchical model. The hierarchical network model was developed by Cisco, and has now become an industry standard, when it comes to designing large scale networks. The Cisco three-layer hierarchical network design contains three layers, and each level in the model deals with a set of problems. This makes it possible to increase the functionality of the hardware utilized in the network. The Cisco hierarchical model groups the network links and devices in three layers, which include the distribution layer, access layer, and the core layer. In the core layer, I would recommend that the SCU campus should use high-speed devices, such as Cisco 6500 or 6800. The Cisco Catalyst 6500 series is classified as a high-performance switch that might be used in application delivery and IP communications in an enterprise campus. The next thing to be taken into consideration, is that SCU wants to allow students to view their homework from their home. This means that they require the network to be developed in such a way that it facilitates remote access to resources in the classrooms present in different buildings. Additionally, a similar goal is that the university now desires for the new campus to be connected to the older campuses, which are situated around 200 km away from the new campus. Both goals are same in nature, and can be achieved by implementing a virtual private network (VPN). SCU would also be able to block certain websites from being accessed in their campus, by means of implementing OpenDNS on every router. Introduction SCU now desires to upgrade their network infrastructure, after purchasing an old school that possesses an old-school type of network set-up. This obviously has to be replaced, owing to the fact that there is absolutely no connectivity between the buildings in this old school, and the actual SCU network that is already in existence. Additionally, the buildings in this old school are unable to communicate with each other. This would prove extremely problematic, when there is a need to share information between each building. Keeping these factors in mind, I would be making certain suggestions about how a new network should be implemented. Through the course of this report, I would be first documenting the network design of the old school, which would make it possible to understand the flaws in the older design. This would play a critical role in proposing suggestions for the new network for the university. Analysis Of User Requirements The current network set-up at the school is fairly old-fashion, and one of the reasons behind this, is the fact that none of the buildings in the campus are linked. This would mean that users would not be able to share any kind of information between buildings via the network. Therefore, a new network design is extremely critical for the new building. Keeping in mind, the fact that these SCU buildings would serve the needs of several kinds of users, it would become critical to thoroughly understand the requirements of different communities of users. The new network design requires a great deal of performance, due to the fact that different buildings have varied requirements. In the first building, say, for the programmers, we would need the computers to run on a high-end operating system. In such cases, a Linux operating system would need to be installed on the respective PCs. In each building, there are activities being carried out, that would place a great deal of pressure on the network’s infrastructure. For example: Students in the multimedia building would need transmission of data in greater quantities, and those involved in programing would also require a considerable amount of speed. Therefore, switches would be required, that facilitate high speed transmission of data. The next thing to consider is the fact that since SCU is an educational institution, it would be essential to block certain websites, especially those that contain pornographic material or social networks. This would be highly detrimental to the overall functioning of the university, wasting the time of students. Therefore, an appropriate mechanism would need to be enforced that would restrict certain websites from being accessed by the students in the campus. One of the buildings in the campus deals with the university’s accounts, hence, it is critical that each building does not encounter failure in the network. Network failures would disrupt performance, and prevent time-critical operations from being executed in each of the buildings. Therefore, the network infrastructure should be designed in such a manner that it is resilient to failure. Connections between nodes ought to be redundant, so as to prevent network failure. Description And Analysis Of Current Setup The current setup lacks any modern networking standards, as it does not provide any wireless connectivity and does not interconnect each building. Each building contains 30 computers. None of these buildings are connected to the main SCU buildings, located 200 km away. Additionally, for each building, no special networking features have been added. By this, I mean, specific networking configuration does not exist to support the needs of different users. For example: Programmers might need to utilize a considerable amount of bandwidth, if they are testing any developed software. Students learning multimedia might need to download certain software from the internet, for which there is no explicit configuration included in the current setup for the building. Firstly, I shall begin by analysing the current equipment being utilized for the network design, which might include servers, computers, and networking devices. In this case, a hub/switch network exists, which is not the most efficient, to say the least. Hubs, often referred to as Ethernet repeaters, have been in use for a long time. They were one of the initial methods utilized for connecting computers together. A hub functions at the physical layer. Hubs are often referred to as dumb devices, due to the fact that they do not possess any forwarding or memory tables (Riley n.d.). In the present network setup, hubs are being utilized for connecting the computers or nodes in every building. This brings a number of performance issues in the network. The first thing to understand, is that hubs are classified as half-duplex devices, which means that they are only able to send or receive data at any given point of time. When a hub is used for connecting the computing nodes in the building, it would be required to constantly shift between receiving and sending data, which resulting in inefficiency in the data flow process. On the contrary, full-duplex devices offer a high level of performance. This is due to the fact that a network switch/hub, when it functions in full duplex mode, is able to send as well as receive data from every device, at once, without having to switch modes (Chron n.d.). Several routers are available in the market, which are able to function in full-duplex mode, if configured to do so (Ehow n.d.). Additionally, hubs share their bandwidth with all the devices they are connected to (Chron n.d.). In the context of the present network layout, we might assume that there are 5 hubs that connect to 6 computers each, and the hubs are connected to each other. If one hub possesses a bandwidth of 100 Mbps, it would be shared by the 6 connected devices. Therefore, if computer is transferring a very large file to another computer on the network, it would result in deprivation of performance, as the computers involved in the file transfer process would consume a lot of bandwidth, leaving none for the other computers. Additionally, hubs would prove highly detrimental for the network’s performance, due to the fact that when one computer tries to each another computer present on the network, the hub would transfer the message from the source, to all computers in the network, which would result in bandwidth consumption (Chron n.d.). For example: if there are 30 computers in a hub-based network, transmitting information from computer 1 to computer 14 would result in the packets being transferred to 29 computers in the network. Then, the computer for which the message is intended, will accept the packets. Therefore, relying on a hub-based network would result in a great deal of bandwidth being utilized for every transfer. For the current network, the usage of hubs might prove problems for the performance, which is why switches have been implemented.Switches, when implemented along with hubs, might worsen performance of the network as well. This is due to the fact that a switch should receive and parse the frame, and start searching, to find where the data packet needs to be forwarded. By storing the packet and then forwarding it to its destination, delays might occur during the transmission (Princeton 2009). The next thing to look into, is the fact that the present network currently utilizes a Windows 2008 server. Microsoft possesses nearly 75 percent of the software and OS market. This has led to Microsoft products being considered as the standard option for a majority of businesses. The present options for Microsoft servers are Microsoft Server 2008 and 2012. Utilizing Microsoft Server 2008, however, comes with its set of flaws. The first thing is that Microsoft’s official website reveals that mainstream support for windows 2008 server would end in the year 2015. In the case of physical servers, the physical hardware would be available for five years, prior to a hardware refresh. This would mean that the support for the product would end prior to upgrading the hardware. Utilizing an operating system after its lifecycle would increase the number of likely security issues, as it would no longer be secured by Windows updates. The next factor to consider with Windows 2008 implementations would be cost. This is due to the fact that the license fees needed are highly expensive. It would become more expensive for the school to expand with its current server, as the price increases when more users are added to the network (Techradar 2014). From the network diagram constructed above, it can be clearly seen that the four buildings are separated from each other, which means no form of communication exists between them. Each building has a hub and switch configuration. Not only would the hub and switch configuration hamper performance, in terms of bandwidth and speed. However, the main issue here still lies in the fact that no form of communication can be established between the buildings. This is a primary requirement for the university, as it is critical for each building to communicate. The usage of Windows 2008 servers would further bring about problems, due to the fact that support for the server OS would end in 2015. This calls for a resilient network design, where every component of the network is connected i.e. every computing node is connected securely, and for better performance, along with every building connected, so as to facilitate communication. New Design For The SCU Campus The new design for the SCU campus calls for a complete change in the way the network is laid out. The older design offered absolutely zero performance and did not permit the smooth addition of nodes to the network. When it comes to describing a new design for the SCU network, the university’s business goals are of vital importance and would form the basis for the choice of network model for the SCU campus. The business goals of SCU are as follows:Information should be protected across the network. Technology should be up-to-date. Information should be easily shared across the campus. Better communication in the campus. Should not prove highly expensive for the university. The next thing, we would need to consider, is the technical goals of the university. The technical goals, along with the business goals, would assist in determining the appropriate model for the network design. Some of the technical goals specified by SCU are listed below: ScalabilityReliabilityVideoconferencing facilitiesAvailabilityAccess to information from home. High Speed InterconnectivityKeeping these factors in mind, the model chosen for the development of SCU’s new campus network would be the three-layer hierarchical model. The hierarchical network model was developed by Cisco, and has now become an industry standard, when it comes to designing large scale networks. The Cisco hierarchical (three-layer) internetworking model is highly useful for designing networks that are scalable, reliable, and cost-efficient (Ciscopress 2014). A hierarchical network design model makes it possible to break down the complex aspects of network design into smaller, manageable portions. The Cisco hierarchical network design model comprises three layers, and every level in the hierarchy deals with a different range of problems. This makes it possible to optimize the hardware utilized in the network. The Cisco hierarchical model groups the network devices and links, according to the three layers it contains, which include the access layer, distribution layer, and the core layer. The three-layer model serves as a conceptual framework. It may be classified as an abstract picture of a network that is in line with the concept of Open System Interconnection (OSI) reference model. Layered models are highly useful, due to the fact that they provide modularity. In each layer, the devices present, have same, well-defined functions and attributes. Keeping in mind, the fact that SCU desires to double its network size, in the next five years, the hierarchical network model would prove highly useful. This is due to the fact that the hierarchical model makes it possible to add, remove, and replace individual parts of the network. This level of adaptability and flexibility assists in increasing the flexibility of the hierarchical network. However, the downside of utilizing a layered model for developing a network design lies in the fact that it is not easy to estimate the exact composition of every layer of the network. This is because every network might be different in its function, and hence, the design. Every layer in the hierarchical design model might include either a switch, a router, a link, or a combination of the three. The hierarchical model also permits grouping several layers into a single layer or omitting one layer fully (Cisco 2003). In the hierarchical network model, every layer can contain two or more devices or a single device could work across several layers. SCU would experience the following benefits from adopting the three-layered hierarchical model, includes:Higher performance: This network model would make it possible for SCU to have a high performance network, where only certain parts/layers of the network are prone to any form of congestion. Management and Troubleshooting: The hierarchical network model makes it possible to define network management and isolate reasons behind network issues. Creating policies: This network model also allows creation of policies and specifying filters as well as rules. Scalability: Scalability is by far the most important benefit for the new SCU campus. The hierarchical model assists the network designer in dividing the network into several functional areas. Predicting behaviour: The hierarchical model would help me gauge the effects of making certain adjustments to the network, which may be either adding or removing network components (Techtarget 2004). The next thing we would need to determine is the size of the network. It is critical to group a network, based on the number of nodes/ computers required. This is because network designs might fluctuate on the basis of requirements and size. The number of devices that would be a part of the network would have a direct influence on its complexity. For example: an organization with fewer devices would not be too complex to design a network for, unlike the network of a large organization. Small networks are those that offer services to around 200 devices. Medium-sized networks are those that provide services to around 200-1000 devices. Large networks support more than 1000 devices. Keeping in mind that SCU wishes to double the number of devices in the network, it would be reasonable to state that a large network would be required for supporting the functions of the university, and its desire to grow in the near future. Hierarchical network models would help in dividing the university’s network infrastructure into several buildings, where each building has its own services in operation. Before understanding how the network ought to be developed, it is necessary to have a fundamental understanding of the different layers of the hierarchical network model, along with its associated functions. Core LayerThe core layer is referred to as the core of the network, as it is present at the top of the network design (Tech-faq 2014). The core layer may also be classified as the backbone of the network. The core layer comprises high-speed network devices, which are made to switch packets as fast as possible. The core layer devices help to interconnect several campus components, which might include the data centre, WAN edge, service modules, and distribution modules. The core layer is absolutely important for interconnecting the distribution layer devices. The core layer should be made redundant and available. The core brings together the traffic from the devices present at the distribution layer (Ciscopress 2014). It is important to note that the sole objective of the core layer is to increase the speed of network traffic as much as possible. The traffic present at the core layer is common for a majority of users and user information is transferred to the distribution layers, which forwards the requests, if need be. The primary responsibility of the core layer is to handle a high amount of traffic. Fault tolerance is also important at this layer (Tech-faq 2014). Distribution Layer The distribution layer lies between the access layer and the core layer. This layer brings together the data it receives from the switches present at the access layer, prior to transmitting it to the core layer, after which it is routed to its end destination (Ciscopress 2014). The distribution layer is linked with filtering and routing, and serves as a communication point between the access and core layer. The distribution layer ought to be designed in a manner that it meets the requirements of the other layers. The distribution layer depicts a routing boundary between the core layer and access layer, serving as a connection point between the core layer and remote sites. The distribution layer is built with Layer 3 switching devices. The multilayer switches or routers present at the distribution layer offer several functions that are absolutely important for fulfilling the goals of the network design, must include: Filtering and handling flow of trafficEnforce access control policiesSummarize routes prior to advertising them to the core routersIsolate the core from the disruptions or failures at the access layerRouting between the VLANs present at the access layerIn the distribution layer, the devices are usually wired in a partial-mesh topology, which offers highly redundant paths that makes sure the network is able to function, even during link or device failure. In the distribution layer, multilayer switches, or routers are mostly installed as pairs, with access layer switches divided between them. This configuration is often called the departmental or building switch block. Every switch below functions independently. Therefore, when one device in the distribution layer fails, the entire network does not go down. Even if an entire switch block fails, it would not have a negative impact on a great number of users (Stewart et al. 2008). Access LayerThe access layer is also referred to as the desktop layer. This layer defines network and resources access to users and workgroups (Tech-faq 2014). The access layer provides end devices with access to the network, in a local area network environment. It may also offer remote sites and teleworkers with access to the corporate network across wide area network connections (Ciscopress 2014). The access layer makes it possible to control user access to the resources of the internetwork. The access layer might be influenced negatively by traffic, leading to poor performance, if not designed properly. The access layer depicts the network edge to which end devices are connected. The access layer devices and services are present in every building of a campus network (Stewart et al. 2008). Some of the functions provided by the access layer include: Handle access control and policy. Devise individual collision domains. Connection of the workgroup via the distribution layer (Tech-faq 2014). Considerations When Implementing The Hierarchical Network Model When it comes to implementing the hierarchical network model at SCU’s campus, it is important to consider few things. While the hierarchical model serves all requirements mentioned by SCU, there exist certain problems in this model. However, the good news is that these problems can be easily worked around, by applying certain networking tactics. Failures that occur at the core layer might negatively impact all users of the network. Certain features or additions need to be incorporated in the network design to reduce the impact of a core layer failure. This is due to the fact that the students should not have to be waiting to complete their daily tasks. It is critical to develop a network is failure-resistant, and recover from failure quickly as well. The routers and switches present at the core layer should comprise the following: Modular chassis-based designManagement modulesDual power supplies and fans Redundant parts cause costs to escalate, but they are worth investing in. The devices present at the core layer should possess hot-swappable components, as much as possible. Hot-swappable parts could be installed or removed sans needing to turn off the power supply to the device. Large scale networks could also benefit greatly from installing generators and large-scale uninterruptable power supply (UPS) devices. These devices are able to reduce the chances of minor power cuts, which could trigger large-scale damage (Stewart et al. 2008). Detailed Network Design For SCU Campus As described above, the hierarchical model would be utilized for designing the campus network for SCU. Since there are three layers in the hierarchical model, I would be dealing with each layer now, pointing to the way it should be designed. Core Layer Configuration In the core layer, I would suggest that the university campus should utilize high-speed devices, such as the Cisco 6500 or 6800. The Cisco Catalyst 6500 series is a high-performance switch that may be utilized in application delivery and IP communications in an enterprise campus. This switch has been designed to serve the needs of a large campus, and would help to reduce costs. This switch provides scalable performance and port density across a number of chassis configurations and WAN, MAN, and LAN interfaces. The Cisco Catalyst 6500 series provides a very high level of operational consistency, which enhances the usage of IT infrastructure, along with return on investment, thereby making it an ideal choice for the SCU campus. Additionally, this switch comes in 48-port to 576-port 10/100/1000 or 1152-port 10/100 Ethernet wiring closets, to hundreds-of-Mpps network cores supporting as much as 192 1-Gbps or 32 10-Gbps trunks (Andovercg 2004). The next thing to be taken into consideration, is the fact that SCU desires to make it possible for students to access their homework from their home. This means that they need the network to be designed in such a manner that it supports remote access to resources present in the classrooms in each building. Additionally, a similar goal is that the university now wishes for the new campus to be connected to the older campuses, which are located around 200 km away from the new campus. Both goals are highly similar in nature, and can be achieved by implementing a virtual private network (VPN). The reason why this information is being discussed in the ‘Core layer’ section lies in the fact that the core layer contains one or more links to the devices present at the enterprise edge that supports virtual private networks (Stewart et al. 2008). The reason behind choosing a VPN instead of other forms of remote communication lies in the fact that VPNs are highly secure. The students might utilize a private VPN service, commonly referred to as a VPN tunnel for protecting their identity and activity online. An anonymous VPN service would prove highly effective in encrypting the data and internet traffic of a student, thereby preventing eavesdroppers from sniffing internet activity of students. Additionally, Wi-Fi service would also be provided as a part of the network design, which might also benefit from implementing a VPN. A VPN service is highly useful, when trying to access Wi-Fi hotspots (Webopedia n.d.). The main goal of a VPN is to offer a secure as well as reliable connection between networks over an already present public network, which is mostly the internet. Additionally, a VPN can be configured in a manner that it handles the growth of the university, keeping in mind that the university desires to double its students in five years’ time. It can be stated that VPNs are highly scalable in nature. In the context of SCU, a remote-access VPN would be utilized, which permits individual users to connect securely with a remote computer network. The SCU students would be able to securely access resources on the network, as if they were connected directly to the servers of the SCU campus network. This type of VPN is referred to as a virtual private dial-up network (VPDN). There are two components needed for a remote VPN to function, and they include a network access server (NAS). NAS may also be referred to as a remote access server or media gateway. A network access server (NAS) could be a dedicated server or it could be running on a shared server (Howstuffworks n.d.). In the case of SCU, it is recommended that a shared server is utilized for the purpose of remote access, instead of a dedicated server, as it would reduce costs. It is important to understand that for a student to access a VPN, they would first need to access the NAS via the internet. Once they connect to the NAS, they would need to enter their credentials, which may be their student ID and password. The authentication of the student would take place via the authentication process of the NAS itself. The next thing that needs to be determined is the use of client software. The reason why client software is important lies in the fact it is required for the students to connect to the VPN at the SCU campus. Client software is needed to establish and maintain a connection to the virtual private network. The client software would assist in setting up tunnelled connection to the NAS, which is entered in the form of an internet address by the end-user (Howstuffworks n.d.). The type of VPN discussed previously is classified as a remote-access VPN. While the above mentioned remote NAS would prove beneficial for students, who desire to access their classroom resources at home, it would not be suitable for communication involving the university branches that are situated far away. To meet these needs, it is recommended that SCU should implement a site-to-site VPN. Site-to-site VPNs permit connectivity between the geographically dispersed sites of an organization. However, there are two types of site-to-site VPNs, which include intranet VPNs and extranet VPNs. For the case of SCU, it is recommended that an intranet VPN should be implemented. This is because an intranet VPN facilitates connectivity between different sites of one organization (Pearsoncmg n.d.). After determining the devices to be utilized at the core layer, it is necessary to understand the type of cabling that should be used. The type of cabling utilized, would play a critical role in determining the speed at the core layer. The core layer should incorporate fiber cables for interconnection (Mcmcse n.d.). A fibre optic cable is a kind of network cable that comprises strands of glass fibers covered by an insulated casing. These cables are made for long distance interconnection and facilitate very high bandwidth communications. These cables carry communication signals utilizing pulses of light. The reason why these cables would be suitable for the core layer interconnections would be the fact that they provide greater capacity and are less susceptible to electrical interference (About n.d.). Capacity would refer to the amount of data that can be carried by the cables. The attenuation of optical fiber is more than copper conductors, permitting longer cable runs as well as less repeaters. When compared with copper conductors that possess similar capacity to carry signals, fibre optical cables are much more easy to install, and weigh around 10-15 times less than copper equivalents (Amphenol-socapex 2004). However, the major disadvantage of utilizing fibre optical cables would lie in the fact that they are expensive to install. Additionally, they are fragile in nature, which would mean that can be damaged easily (Webopedia n.d.). Although fiber optical cables, due to their fragile nature and initial costs, might raise expenses, they are definitely ideal for the core layer, as they support the high performance demands present at this layer. Distribution Layer ConfigurationThe main function of the distribution layer is to make it possible for the access layer to establish a connection with the core layer. The distribution layer connects every device in the access layer to the core layer. This would enable the access layer devices to route data among themselves and to the core layer. The distribution layer deals with packet filtering, routing packets, and WAN connectivity. Some of the switches that may be implemented at the distribution layer include Cisco 5000 or 6500 switches. These two ranges of switches are ideal for the distribution layer, as they are able to route as well as switch (Techrepublic 2002). Keeping in mind the fact that SCU desires to keep its budget fairly low, the Cisco Catalyst 5000 switches would be implemented. This is because the distribution layer does not require a high amount of performance, compared to the core layer, where Cisco Catalyst 6500 switches would be implemented. The Cisco Catalyst 5000 series contains a Gigabit Ethernet and ATM-ready platform, which provides users with high-speed trunking technologies. The Cisco Catalyst 5000 Series also possesses a redundant architecture, and dynamic VLANs. The Cisco 5000 series makes it possible for network managers to provide high speed backbone access to host web browser-based traffic in the intranet (Cisco n.d. b). In the distribution layer, the cabling to be used would be coaxial cables. Coaxial cabling possesses a copper conductor in the centre, while a plastic layer offers insulation between a braided metal shield and a center conductor. While it is fairly difficult to install coaxial cables in the network, it is definitely worth it. This is because the metal shielding inside the cable helps to prevent interference from motors, computers, and fluorescent lights. These wires are highly resistant to signal interference. The next thing to consider would be the fact that the distance between the distribution layer switches and core layer switches might be fairly great. Coaxial cables support a greater cable length between devices (Usf n.d.). In total, the number of switches at the distribution layer would be four in number i.e. one switch for each building. Redundancy in network connections is obtained by connecting one distribution layer switch to every access level router, and one core layer switch to every distribution layer switch. Access Layer Configuration It is important to define the kind of networking devices that should be present at the access layer. This is because the access layer comprises devices that would allow the students and staff present in the different buildings to utilize the services offered by the core and distribution layers. The access layer presents network managers with options to utilize either hubs, repeaters, or a standard switch. It is necessary to note that if switches are utilized at the access layer, they are not the same ones, which are present at the core layer. Instead, the switches to be used at the access layer, are advanced versions of hubs (Mcmse n.d.). The switches might be deployed in the wiring closets of each building, which are then connected to the distribution layer switches. When it comes to selecting devices for the access layer of the network, certain things would be taken into consideration, and one of them would be the cost-per-port. In the case of SCU, it would be highly recommended to keep the cost-per-port low, owing to the fact that the university desires to double the number of students in five years. For example: If a switch costs around $480 and the number of ports is 24, the cost-per-port would be $40. In the context of SCU, we would have a choice between a hub and a switch for the access layer. The better choice here would be a switch, despite hubs being a cheaper alternative. The performance benefits of a switch, however, make it a better alternative, considering the high number of users. The switches that may be utilized at the access layer include the Cisco 1900, Cisco 2900, Cisco 4000, and Cisco 5000/5500. The Cisco 400 modular range of switches utilizes 10, 100 and 1000 MB connections. They might also be utilized for advanced telecommunications, which include unified messaging, and IP telephony (Techrepublic 2002). For the access layer of SCU, the device that would be utilized is the Cisco Catalyst 4006 switch. The Cisco 4006 switch facilitates redundant connections, which is a critical aspect of the proposed design. Additionally, the number of Ethernet ports available is 240, which is ideal for the each building. The price of the Cisco 4006 switch is $316, which is fairly reasonable, considering the performance it offers (Pcworld n.d.). In total, the number of Cisco 4006 switches to be installed for the four buildings of the campus would be 8, to facilitate redundancy in each building. DNS Server For Blocking Restricted Websites SCU’s campus also wants to keep track of the websites accessed by the students. Students should not have access to pornographic websites, along with social networking sites. Social networking sites would result in a great deal of time wastage, lifting the focus off their academic work. Additionally, certain websites could pose a threat to the internal security of the network, if accessed by the students. The next thing to understand is that an insider might intentionally access malware-infested websites, while certain users might access such websites unintentionally. Therefore, it is highly recommended that a suitable networking security mechanism be enforced, so as to prevent certain websites from being accessed by the students in the network. The ideal solution for blocking restricted websites in a network, would be a DNS. DNS stands for domain name system, and is a standard technology for handling public names of websites as well as other internet domains. DNS technology makes it possible for users to enter names in a web browser, and the computer finds the address on the internet automatically. A DNS server may be any computer that is registered to be a part of the Domain Name System. It is important to understand that a DNS server runs certain networking software, comprises a public IP address, along with a database of network addresses and names (About n.d.). DNS is defined as a protocol within the TCP/IP protocol suite that governs the exchange of information between private and public networks. The main function of the DNS is to convert an internet address in English into an internet protocol address that computers utilize to identify each other on the network. Computers as well as other networking devices utilize an IP address for routing the request to the website trying to be reached. A domain name server handles an extensive database that maps IP addresses to domain names (Howstuffworks n.d. b). A domain name system (DNS) is defined as a server application that guides end-users how to access local servers. DNS servers are hosted by internet service providers, but for the purpose of this network, third-party DNS servers should be utilized. This is due to the fact that they offer more features than ISP DNS servers. While there are many ways to block pornographic material or social websites in a network, but DNS servers are the most effective. It is recommended to configure web filtering by changing the DNS servers to Open DNS. The DNS server present on the routers of the network can be configured via the parental control settings present on the OpenDNS website. This would make it possible to block certain categories of websites and view the websites accessed in the campus network. OpenDNS may be configured on the routers present at the access layer. This would enforce the settings on each computer present in the respective buildings. By registering for an OpenDNS account, the campus would benefit from features such as access statistics and domain/content blocking. An account can be created using the ‘Create Account’ link present on the upper right hand corner of the website. After that, logging into the OpenDNS account would present ‘Settings’, where the users could click on ‘Add a Network’. This would make it possible for SCU to associate their network with the account. It is important to note that no IP address information needs to be provided, as it would be automatically detected by the software. To determine whether the network has been successfully added to the DNS software, the ‘Manage your Networks’ section would be checked. If it has been added, the network details would be present there. In the case of SCU, it is important to note that OpenDNS would not be as good as other paid software, but it would definitely help in cutting costs, which is a major concern in the university. To utilize the OpenDNS blocking facility, the ‘Adult Site Blocking’ feature should be clicked on. OpenDNS would then present a list of adult content categories, which can be blocked. After selecting the desired categories, the ‘Apply’ button can be clicked, so as to apply the filters for these categories. However, considering the fact that certain students in the campus would be rather inquisitive and try working around these controls by utilizing proxy or anonymous servers, OpenDNS would assist in blocking access to these sites as well. While the steps mentioned above are helpful for blocking general website categories, for example, porn, it may not block specific websites containing pornographic material. Therefore, in such cases, OpenDNS may be configured, so as to block access to specific domains. The ‘Domain Blocking’ option of OpenDNS makes it possible to block specific domains, by entering their respective names. However, it is important to note that the IT maintenance team should only make changes to the list of permitted websites at night. This is because OpenDNS might require around 10 minutes to update the blocking settings on its servers (Practicallynetworked n.d.). Conclusions SCU desires to upgrade its network infrastructure, since the older network does not connect any buildings together. Therefore, the network to be designed, would be such that it interconnects all buildings within the university campus. Through the course of this report, I proposed an effective solution to the problem faced by the university. The solution would make it possible to connect different buildings of the university, using high speed links. Additionally, it is important for the network to be resilient to failure. I have utilized the network hierarchical model by Cisco, to develop a network that is highly resistant to failure. This has been achieved by making redundant connections between devices present at each layer of the network. The hierarchical model was chosen for this network design, also due to its scalability. Hierarchical model’s scalability would make it possible for SCU to continue adding more computers to the network. References: Mcmcse, The Cisco Three-Layered Hierarchical Model, Viewed on 27/12/2014 Usf, Cabling, Viewed on 27/12/2014, Pcworld, Cisco Systems, Inc Catalyst 4006 Ethernet Switch, Viewed on 27/12/2014, Practicallynetworked, Block Web Sites and More With OpenDNS, Viewed on 26/12/2014, About b, What Is A DNS Server?, Viewed on 26/12/2014, Webopedia, Fiber Optics, Viewed on 26/12/2014, Howstuffworks b, How Domain Name Servers Work?, Viewed on 26/12/2014, Amphenol-socapex 2004, Understanding Fiber Optics, Accessed on 26/12/2014, About a, Fiber Optic Cable, Viewed on 26/12/2014, , What Is A Virtual Private Network?, Accessed on 26/12/2014, Ciscopress 2014, Cisco Networking Academy Connecting Networks Companion Guide: Hierarchical Network Design, Viewed on 25/12/2014, Tech-faq 2014, Understanding The Cisco Three-Layer Hierarchical Model, Viewed on 25/12/2014, Cisco 2008, Campus Network For High Availability Design Guide, Accessed on 25/12/2014, Andovercg 2004, Cisco Catalyst 6500 Series Switch, Accessed on 25/12/2014, Techtarget 2004, The Cisco Three-Layered Hierarchical Model, Viewed on 23/12/2014, Techradar 2014, Windows Vs Linux: Which OS is best for your business?, Viewed on 22/12/2014, Princeton 2009, Switches and Bridges, Accessed on 21/12/2014, K., IT Guru Lab 3: Hubs, Bridges and Switches, Accessed on 21/12/2014, Ehow, How To Configure Duplex on a Router, Viewed on 21/12/2014, , What are the Disadvantages of Using Hubs?, Viewed on 21/12/2014, Webopedia, VPN- Virtual Private Network, Viewed on 26/12/2014, Cisco, Cisco Catalyst 5000 Series Switches, Viewed on 26/12/2014, Techrepublic 2002, Build Your Skills: The Three-Layer Hierarchical Model, Viewed on 26/12/2014, ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download