By Melanie McCormick
Web Research Term Paper
October 7, 2008
OMGT 5823: Computer Applications
Instructor: Marita Ellixson and Nancy Sloan
Improving a networks availability is the process of improving an existing network for better performance and reliability. Network improvement has become crucial to our pursuit of life and happiness in this modern generation. However, police departments, hospitals, businesses, individuals and virtually anything we depend on runs on their networked computer systems. The more we depend on these networked computer systems, the more it affects us when they stop working to our satisfaction. Those of us that are involved in the planning, designing, building and operation of these networks, must also be able to predict problems in advance and also allowing rooms for scalability. Predicting problems in advance allows us to reduce the impact of these problems. With predictions of improved network availability, we can make sure our networks are going to service people satisfactorily before we build them. However, users of networks are expecting more from networks when compared to the early days when a network was meant for e-mail and resources sharing e.g. printers, files etc. Today, businesses depends mostly on networks for the purpose of transacting their business such as video conferencing with partners abroad, web-based order entry, payment processing and customer relationship management, to name a few. When these mission critical services are unavailable or performing poorly, they can bring business activities to a halt, affecting end user productivity, revenue, and customer satisfaction – hence, network availability should not go down but improved. Internet Protocol
The Internet Protocol (IP) is the most widely-used method for transporting data within and between communications networks. It is as useful for the growing field of intranets (networks internal to an enterprise or organization and not connected to the outside world--e.g., a network used for classified processing) as it is for the geographically distributed, highly heterogeneous Internet. IP combines the functions of internodes linking with those of links between physical networks to provide communications paths between nodes on different networks.
IP is called connectionless because it resembles the Postal Service or Western Union more than it does the telephone system. When a node using IP wishes to send a message to another such node, it simply sends the packet, properly addressed, analogous to mailing a letter or sending a telegram. (In fact, another name for an IP packet is a datagram.) The telephone system, on the other hand, creates a connection between two users which is maintained for the duration of the information exchange. Unlike the Postal Service, however, the services of IP can be used to create a connection-oriented operation mode, but this is the job of higher-level protocols and applications (such as TCP, File Transfer Protocol [FTP], Telnet, and others). In IP's connectionless design, every packet is treated completely independently from all others.
The IP Packet
IP centers around the concept of a packet. A packet is also known as a datagram, although that term is also used in the context of a number of different protocols at different levels of communications architecture; the term datagram specifically implies that the protocol operates in connectionless mode, as described above. Pursuing the Postal Service analogy used previously, an IP packet is comparable to a letter, although it is different in some important ways. An IP datagram consists of a header and a payload. The header is analogous to the envelope handled by the Postal Service, while the payload is analogous to the contents of that envelope. Note that the post office is not permitted to open envelopes or peek inside; its entire job can be...