Applied Data Communications and Networks by Buchanan - Book Report/Review Example

CHAPTER SUMMARY Student name: Institution: Date: Chapter 1 and 2: Computer Network Topology A computer network is the connection of a group of computers together with the motive of data and information sharing. Computer network topology, therefore, is the manner in which individual computers connect, including the devices and the nodes used. There are three common types of network topologies which include; the fully connected or mesh, bus, star and hub topologies. The mesh topology allows the interconnection of computer and other devices with each other. Every node is therefore capable of sending its own signal as well as disseminating data from other nodes. It is the most difficult to build and the number of connection at each node equals the total number of nodes minus one. Bus networks are characterised by competition for possession by all nodes on a common bus. Data is transmitted when an idle bus is detected. Collision occurs when two or more nodes transmit data simultaneously, and this is handled by the Carrier Sense Multiple Access with Collision Detection, which causes nodes to stop and wait before data retransmission. In star networks, all the nodes connect to a central communication hub known as the concentrator. Small networks require only a single hub with4, 8, 16, 32 or more connections available. Large networks require multiple hubs thus increasing hardware and cabling costs. Fig 1.1: Topologies for a five node network In ring networks, each node is connected to two other nodes and the data circulates in a ring. It consumes time since information travels around a ring and link failure can cause network failure. Network access points offer access to national, as well as global network traffic. Companies can install communication networks by connecting to one or more NAPs or acting as NAPs themselves. NAP connected companies enter into agreements which allow traffic exchange. Chapter 3: Networking Hardware The basic networking hardware components comprise of cabling, network interface cards, repeaters, transceivers, hubs, switches, router and firewalls. The major cable types that are used are the Ethernet cables and the specialised cables. Ethernet cables enhance fast connections between two devices like computers and routers. The three main varieties of Ethernet cables are the RG-58 coaxial cables or thin wire such as the 10 base 2 operation, the RG-11 coaxial cable or thick wire such as the 10 base 5 operation and the unshielded twisted pair (UTP) which comprises of 10 base T, 100 base T and 1000 base T operations. The UTP cable is used with hubs switches and other 10 or 100 base T equipment. The twisted pair wires are advantageous because they reduce noise and cross talk and allow higher speed data rates. Fibre optic cables on the other hand use light pulses in the transmission of information down the fibre lines. The construction of these cables require plastic or glass with different properties. The light beam used reflects off the boundary between the core and the cladding. The expression of the fibre optic sizes is usually given in terms of the core size followed by the cladding size. There are two major classes of the fibre optics cables. These are the single mode and the multi- mode allowed. The fibre optic cables are advantageous since they eliminate the problems of copper wires, but are very fragile and should be handled carefully. Another hardware device used is the network interface card (NIC) that is the connection between a networked device and a physical network connection. It connects to a coaxial cables, UTP cables and the fibre segment. It allows data transfer to another computers well as controlling data flow between the computer and the cabling system. The token ring allows the connection of computer with a bit passing technology that prevents data collision between two computers transmitting data concurrently. The token ring technology dates back to the 1980s and uses a multi station access unit (MAU) to establish a ring connection. Connections are often made using the STP cables and has data transfer rates of between 4 and 16 megabits per second. Repeaters allow connection between two network segments, thus broadcasting segments within the packets and amplifying the signals. The common Ethernet rule is that four repeaters can join up to a maximum of five segments. Repeaters work at the open systems interconnection (OSI) physical layer. Transceivers on the other hand help in converting data transmissions from one media to another. Installations commonly use more than one media type hence availability of different transceivers. They also operate at OSI model physical layer and are especially important in upgrading networks. Hubs expand an Ethernet connection into many connections. They transmit the data received by any port to all the other ports on the hub. They however contain some little intelligence and can check for integrity through examination of the received files. They also operate at the physical layer and the number of hubs that can be connected in a series is limited. Cable modems are high speed network devices connected to a local television provider. They use a pair of channels for data transmission on the cable system. Subscribers to the modem services use a splitter to create two cable wires. Chapter 4: The Ethernet Technology The Ethernet frame format allows data transmission from the less significant byte to the most significant byte. The minimum and maximum sizes are 72 and 1526 bytes respectively. Compared to the Ethernet II, it has a 2 byte type field instead of a length field and a data field longer than 1500 bytes. The inter frame refers to the idle time appended to the end of every frame. The inter frame allows the stabilization of the network media for network components to process frame. The larger the frame size, the mere the utilization of the available bandwidth and the less the significance of the inter frame gap. The Carrier Sense Multiple Access with collision detection (CSMA/CD) enables band with sharing. Collision occurs when two or more transmitting stations collide with each other. Collision during the frame is alleviated since the station waits for an idle period, that can be equal or longer than the inter frame gap before beginning the transmission of frame bitwise. In case of collisions during the frame however, the Ethernet transceivers detect signal and energy distortions, the stations then begin a random waiting period before retransmission is done. Error detection is done by the transceiver, by listening to itself as it transmits the signals. The irregularity of signals due to collision causes network malfunction. This is followed by a jam sequence generated by stations detecting collision. The 32 bit pattern is used to propagate collisions throughout the network. The random waiting period then occurs before retransmission. Ethernet controllers are network interface card digital connections that transmit and receive Ethernet frames. This task is performed by single dedicated ASIC (Application Specific Integrated Circuit) Ethernet controller. The ASIC contains a bus interface logic that connects directly to the PCI. An example is the Realtek RTL8130. Among the available Ethernet technologies are 10-Mbps Ethernet, comprising of 10 base 5, 10 base 2 and 10 base T. The 100- Mbps Ethernet has a smaller network diameter than the 10-Mbps Ethernet and is necessary in maintenance of CSMA/CD parameters. Gigabit Ethernet are used for bandwidth demands exceeding100-Mbps. They are natural extensions of earlier Ethernet versions. They however have decreased network diameters and require solutions to maintain a reasonable network diameter. The latter differs from the 10 and 100 Mbps Ethernet since they have a new 8 bit wide transmit and receive path and full duplex operation. Examples include 1000 base T, 1000 base CX, 1000 base SX and 1000 base LX. The 10 GbE increases the network speed between servers. Offers solutions for streaming audio and video through provision of fast backups and standard specification of different types of fibres as well as copper cabling requirements. Another important type of Ethernet is the wireless Ethernet, which is an Ethernet over radio frequency or infrared. It is composed of one or more fixed stations. In wireless Ethernets, the transceiver cannot listen to other transmissions while disseminating and stations avoid collision through random back off delays. Propagation of radio based signals is through objects and with the use of ISM band. In infrared based signals, the signals bounce off walls, ceilings and floors and the data rate is limited by the multipath effect. Point to point infrared propagation involves the use of line- of -sight IR lasers which offer faster data rates than diffused IR and works over larger distances. The problems encountered include cabling errors such as lost termination and excess utilization, alignment error where the last bit received is not the last one on the frame final byte and frame bits not being multiples of 8. Others are jabber where the frame is longer than 1526 bytes and runt in which the transmitted frame has a shorter length than the minimum frame size. Chapter 8: TCP/IP It refers to the Transmission Control Protocol or the Internet protocol and it is the fundamental communication language in the internet. The publication of the official internet standards are done in the Request for Comment (RFCs) which are electronic documents. The Internet Protocol suite based layer is used at the TCP/IP internet layers. The data in datagrams which inserted into hardware frames such as Ethernet frames for transmission over the local area network. The datagram transmission involves transfer of the datagram over LAN to a router where the destination address determines its transfer to another network. The datagrams are treated separately from and they can also take different routes. The IP is however disadvantageous since data delivery is not an obligation. Datagrams use local network Maximum Transmission Units of 1500 bytes and may sometimes reach lower MTU networks thus requiring fragmentation. For successful datagram transmission, the IP addresses are significant. The IP addresses uses decimals to rout the datagrams across a network. These addresses are either dynamically or statistically assigned by the software. There are different versions of the IP. These include; the Internet protocol version 6 which offers location and identification for computers on networks it accommodates the future network growth. The version allows addition of new features thus does not require updating. The IP version 4 on the other hand consists of 32 bits divided into four sections each containing 8 bits. TCPs enhance communication of computers over interconnected networks. They also serve as communication links between the application programs and the internet protocol. The TCP guarantees data delivery and focuses on data transfer, reliability, flow control, multiplexing and message precedence and security. Other applications however do not require TCP complexity. These are the Domain Name Service for host name conversion to IP address, the Dynamic Host Configuration protocol which allocates an IP address to network devices and Simple Mail Transfer Protocol that routes emails using TCP/IP. Chapter 10: Switching and routing The hubs perform an important role of transmitting the received frames to other ports while switches just forward the received frames to specified ports. Fully switched networks, therefore, are LAN stations which are connected to switched ports. Switches are composed of input port logics with Ethernet receiving logic and buffers for frame reception, and output logics with Ethernet transmitters and output frame buffers. They also have the switching fabrics for direction of input port frames to output ports and handling transmissions to the output ports. Different switching methods are used for different connections. Characterised by latency, the store and forward switches were used initially for the storage of entire frames. Validity of the frame check sequence allowed forwarding to the next relevant port. Cut- through switching reduces latency since the forwarding process begins with the reception of the address of the incoming frame. It however allows error propagation as the frame cannot be checked before receiving is completed. Routing protocols operate at the network layer of the OSI and carry out different forms of packet forwarding. The creation and maintenance of routing table is done through static routing involving creation of many predefined routes, the routers cannot discover new routes. Dynamic routing involves discovery of new routes and updating the old ones. The routing protocols used in private networks are Autonomous System (AS) number with a single and well defined routing policy, interior gateway protocols (IGP) for use inside the autonomous systems and exterior gateway protocols (EGP), which allow information exchange between different systems. Other classes are link state routing, policy routing and distant layer routing. The evolution of information technology is vital since it allows data and information sharing over both the public and local networks. The evolution of the various devices used for connectivity have made landmark changes in this field. Globalization has been made a success with the different communication medium. The communication technologies have therefore played a vital role in the comfort we enjoy today as far as communication is concern. References Buchanan, W. (2012). Applied Data Communications and Networks. Boston, MA: Springer US. Read More