Networking - Statistics Project Example

? Full Paper Network LAN Topologies Network topologies are essential before implementing a network as per requirements of an organization. Topology is a framework defining the arrangements of every object on the network. This includes workstations, network components, servers, WAN devices and many more. There are total five topologies to design a computer network. However, all of these five topologies share certain factors. Each topology demonstrates disadvantages as well as advantages that will be discussed further. The five different topologies are illustrated below: 1.1 Star Topology Star topology is recommended for the wired local area network. It is the most widely adopted topology. The star topology supports the centralized provision of network resources and services. The support staff can manage the network administrative and troubleshooting tasks centrally. Star topology helps to implement centralized security architecture for improved and enhanced security of the network. The network implementation cost can be saved by provisioning the core systems located centrally. The security controls and backup systems are also located centrally for better troubleshooting and management. 1.2 Ring Topology The ring topology shares the same network cable to connect workstations, servers, printers and peripheral devices on the network. The gateway for data exchange is the hub that is based on multi station unit or multi station access unit. The functionality of exchanging data on the network includes a token ring that passes via each node on the network. The required data is encapsulated in the token ring, which delivers the data in an orderly manner to the recipient. 1.3 Bus Topology The bus topology shares a single communication channel via all the nodes available on the network. However, before transmission, collision detection and multiple accesses (CSMA/ CD) sense the current activity on the bus, in order to prevent collision between the two communication channels. If CSMA /CD sense that the data is already available on the bus, then it waits and counters checks in order to transmit data on the network, when the bus is free. 1.4 Tree Topology The tree topology is based on two topologies i.e. star topology and bus topology. Moreover, tree topology facilitates the network administrator to expand the network easily. However, if the network expands, it becomes more difficult for the network engineers to manage the network. 2 Issues of Tree Topology As there is no centralized provision of services, it is difficult for the network engineers to tackle issues related hardware, software and network services. Physical intervention is required as the hub backs up each network segment, if the hub become nonoperational, the workstations of the entire department becomes unavailable on the network. Moreover, it is not necessary that each network segment is using the same cabling scheme (Chapter 5: Topology). Due to compatibility issues, cross cabling or straight cabling is required, consequently making difficult for network engineers to troubleshoot network issues. Moreover, due to separate network segments on the network, some network segments may become vulnerable to threats, as network administrators do not have a centralized access to the network devices. These considerations must be analyzed during the network design stage of an organization. 3 Network Infrastructure Recommended LAN topology is a star topology. As star topology shares many advantages, the most prominent one is the centralized administration and monitoring of the network. Moreover, star topology provides high data speeds along with easy installation of cables. Any component on the network can be removed or plugged in without any interruption. Likewise, centralized administration facilitates network administrators and engineering to detect issues and faulty components. Hubs may deteriorate network performance. As they can only operate on the maximum data transmission speed of 10 Mbps and copy data files to the designated ports, files that are larger will take time and may congest the network. However, cost of the hub is low as compared to a switch; it may affect the performance of the network. On the other hand, switches can improve network performance significantly. As switches can learn the MAC address of the sender, they create a management information base for storing all the MAC addresses of network nodes. This will eliminate the time taken by the switch to identify and learn MAC address of the sender. Moreover, switches can provide the overall network usage along with statistics related to access policies via the network. Switches only share a disadvantage of a high cost as compared to hubs. A simple definition of regenerative repeater is available in network dictionary which states it as “a device which regenerates incoming digital signals and then retransmit these signals on an outgoing circuit” The regenerative repeaters are the network devices, installed within the network to regenerate signals that may become weak or unreachable to the destination, due to long distance. These devices are implemented in long distance wired networks for transferring data geographically from one network to the other. As signals, are evolved via network cables, they become weak and attenuated over a certain distance within the same network, resulting in unreachable signals to the destination. This is where regenerative repeaters are useful as they regenerate signals received on their ports and sends the regenerated signals to all the other ports connected on the other end. Moreover, regenerative repeaters do not maintain a database as compare to Ethernet switches and operates on Physical Layer of the OSI model. Furthermore, regenerative repeaters are not compatible for regenerating signals via two different networks. For instance, the regeneration of signals between a token ring network and Ethernet network is not supported. For example, there are two network locations, where the distance between the two is above 100 meters, repeaters are required to synchronize uninterruptible transmission of data from one network to the other. Otherwise, the attenuation may disturb the digital signals and may not reach the distance, which is above 100 meters. A simplest of definition for a network device known as bridge is as follows “Local Bridge is a bridge that directly interconnects networks in the same geographical area” (Local Bridge. 2007). Bridge has capabilities to develop network segments. The segments are developed to eliminate unnecessary broadcasting from some workstations or a department on the network. Consequently, the performance of the network will be efficient as bridges can segregate workstation from the network. As these devices operate on data link layer of the OSI model, it is implemented to amplify distance between network segments, allowing room for more workstations, to be attached within the same network. Moreover, the segregation of affected workstations from the network eliminates network congestions and traffic choking. Bridges do not deal with the information available in the higher levels of the OSI model due to their presence at the data link layer. Bridge identifies each node by its network address, it is the devices that recognize which protocol and data pertains to them. Moreover, there is a component installed inside the bridge device known as RAM. The bridge’s ram is used to store routing tables of source addresses of all the nodes associated with the network. For example, two neighbors want to share their home network with each other, installation of the bridge device between these two home networks will connect them to each other, representing a single network. A comprehensive definition is available in network dictionary that says “Ethernet Hub or repeater is a device that accepts numerous Ethernet connections from network devices and cross-connects them, making them act as a single segment. Data arriving via the receive pair of one connection is regenerated and sent out on the transmit pair to all connected devices except for the device who originated the transmission. Ethernet Hub works at the physical layer of the OSI model. If a signal comes into two ports at the same time when a collision occurs, so every attached device shares the same collision domain. Hubs support only half duplex Ethernet, providing bandwidth which is shared among all the connected devices”. Hubs are normally implemented for connecting workstations with network interface cards (NIC). It operates on the physical layer 1 of OSI model. Moreover, hubs are of two types named as active hubs and passive hubs. Active hubs are also knows as repeaters, is a device that amplifies data transmission on the network along with forwarding signals to the destination. On the other hand, passive hub only transmits the signals to all ports (Active Hub. 2007). For example, if two or more than two workstations require file sharing, print sharing, and Internet connection sharing, they can use a proxy server for sharing internet services with each workstation by installing a hub device. 3.1 Wide Area Network Connectivity A complete definition is available in network dictionary that defines router, as “Routers are physical devices or a software system within a switch operating a layer 3 of the OSI model. A router forwards and routes data packets along networks. A router connects at least two networks, commonly two LANs or WANs or a LAN and its ISP network. A router is located at any gateway where one network meets another, including each point-of-presence on the Internet”. Router is a processing device, consisting of various types including non-seed router, seed router, filtering router, level 1 router, level two routers, level four router, gigabit switch router, interior router and much more. Moreover, routers operate on the network layer 3 of OSI model. It takes data packets as input, analyzes them, and transmits them to their preferred location. Routers are more capable then hubs, repeaters, and switches as network engineers configure access control list in routers to restrict and allow access to specific users. If a home network is considered, router is directly connected to the WAN interface, and is the only device to allow or deny Internet access for one or more workstations in the home network The WAN is based on T1 links that are terminating on an external CSU / DSU. Bandwidth can be shared or committed information rate (CIR). Moreover, bursting is also supported up to 1.5 Mbps o CIR. Organizations spent massive funds to purchase advanced security appliances and hire security professionals, only to ensure the protection of the network. As over the last two decades, network security has gained significant importance due to increase in a number of threats and vulnerabilities. There is a requirement of monitoring the network on a continuous basis for possible threats and vulnerabilities. In order to establish a secure and robust WAN architecture, three factors need consideration. The second factor is the WAN architecture. If we suppose, WAN architecture is based on T1 that only supports 1.5 Mbps and is not cost effective as compared to other WAN technologies. Moreover, there is a requirement of physically installing the cable that may cause disruptions in business operations. Whereas, DSL can provide a significant advantage over a T1 as it operates on a telephone line along with higher bandwidth rates. Moreover, DSL is a ‘cost saving’ option due to its low rates. Likewise, DSL operates on two modes i.e. symmetric DSL and Asymmetric DSL. Some of the advantages of a DSL connection over a T1 are as follows: Higher Availability in rural areas Low bandwidth cost Due to its high speed, VoIP services eliminates lagging Provides Unlimited Downloads Splitter provides an extra telephone line as well as DSL service DSL routers can provide Wireless connectivity as well DSL routers have an internal Firewall DSL routers have a built in firmware supporting access control lists and filters. 3.2 Network Protocols The ‘TCP three way handshake’ is a method incorporated by the TCP to establish a data session or a TCP session (TCP three-way handshake.2007). Likewise, the session establishment is carried out between the client and the server that may initiate or accept the TCP connection. As the TCP is connection oriented protocol, a reliable connection is required to ensure reliable data connectivity (Three-way handshake.2007). Moreover, the ‘TCP three way handshake’ is considered to the standardized method of creating TCP data connections (Beardsley & Qian, 2010). However, in order to establish a data connection, a device must transmit a SYN and receives ACK from the other device. However, in a logical perspective, there are total four control messages that correspond between the two devices trying to establish connectivity. Likewise, SYN and ACK messages are not enough, as they are sending simultaneously by configuring relevant bits of both SYN+ACK. For this reason, there are total 3 messages and therefore called as a three way handshake procedure. The semantics of these messages passing between a client and a server as shown in fig 1.1, demonstrates the procedure and steps required for a three way handshake. Moreover, the fig 1.1 also demonstrates the traditional method of establishing a data connection between a server and a client. Likewise, the three messages transmitted for establishing a connection within each process along with the transition of devices that were in a ‘Closed’ state initially and changed to ‘Established’ status after the successful connection creation. Likewise, the circuit switch network is provisioned by the Internet service provider ISP, as the ISP provides two types of connection i.e. Analog home connection and Digital home Connection. The ‘analog connection’ comprises of Integrated Services Digital Networks ISDN that is comparatively a slow connection and does not support high bandwidth, as they only support 19 – 128 Kilobits per second. However, a research was conducted by using supercomputers, where an evaluation demonstrated dissimilarities in a TCP/IP computer network that was not able to handle overloading of bandwidth. On the other hand, a circuit switched network handled the bandwidth overload efficiently and justifies itself to be a reliable service (Kessler, 2009). Following protocols are required to establish DSL Internet Connectivity from ISP: TCP Transmission control protocol ensures the reliable delivery of data on time. If a data packet fails to reach destination, TCP recalls for the missing packet. UDP User datagram protocol is a connection less protocol that is configured to optimize ‘real time’ applications. However, UDP does not ensure reliable delivery of data. FTP File transfer protocol provides convenience for transferring files on the Local Area Network (LAN) as well as on the Wide Area Network (WAN). HTTP Hypertext transfer protocol (HTTP) displays information related to web pages on the web pages are demonstrated on the web browser by HTTP. HTTPS Secure Hypertext transfer protocol (HTTPS) also servers the same purpose except adding security features by incorporating Secure Socket layer (SSL). 3.3 Advantages and Disadvantages of Adopted Protocols Advantages Disadvantages TCP Ensures Delivery on time, Request for missing packets, built in debugging As TCP relies on operating system to operate, issues arise if an operating system becomes corrupted. UDP Makes more than one connection to receive data and ensures fast delivery of data Do not ensure reliable delivery of data due to no acknowledgment. Moreover, it also not ensures timely delivery of data packets. FTP Facilitates large file transfers and easy to organize file transfers. As FTP is not a secure protocol, it allows space for brute force attacks, packet sniffing and spoof attacks. HTTP It is easy to code in HTML as compared to other programming languages. As all the authentication details are available in the HTTP header, routers can easily manage authentication criteria. Authentication information available in the HTTP header can be easily retrieved by hackers as the information is nor encapsulated. HTTPS HTTPS operates on the same mechanism, except its integration with Secure socket layer. HTTPS encrypts all the information and transmits it in a secure tunnel. An SSL certificate expires in a small period. Moreover, integration of HTTPS in a website is a complex and time-consuming tasks. VoIP VoIP provides ‘low cost’ communication with exceptional voice quality. Moreover, trunks can be created on a DSL connection to divide them in a dedicated VoIP channels. Once the VoIP server goes down, there is no service for even emergency calls. Moreover, the quality of voice depends on Internet bandwidth. 3.4 Wireless Connectivity For Wireless connectivity, two security protocols will be used. If someone sends data from Wi-Fi network, the wireless router initialized Wi-Fi Protected Access (WPA). This protocol operates on wireless 802.11 networks from the Wi-Fi alliance established to enable migration from Wired Equivalent Privacy (WEP). The WPA provides strong security by implementing a sophisticated key hierarchy in order to generate new encryption keys each time a workstation connects from an access point. For home users, WPA operates in pre shared keys (PSK) mode, so that whoever connects to the wireless network, must provide a secret key in order to establish connectivity. References Active Hub. 2007. Network Dictionary, , pp. 21-21. Beardsley, T., & Qian, J. (2010). The TCP split handshake: Practical effects on modern network equipment. Network Protocols & Algorithms, 2(1), 197-217. Kessler, G. C. (2009). Cyber security and global information assurance: Threat analysis and response solutions. Journal of Digital Forensics, Security & Law, 4(3), 57-59. Local Bridge. 2007. Network Dictionary, , pp. 289-289. Regenerative Repeater. 2007. Network Dictionary, , pp. 405-405. Router. 2007. Network Dictionary, , pp. 419-419 TCP three-way handshake. (2007). Network Dictionary, , 479-479. Three-way handshake. (2007). Network Dictionary, , 487-487. Read More