Data Bases • Custom Term Papers • Free Term Papers • Free Research Papers • Free Essays • Free Book Reports • Plagiarism? Links • Top 100 Term Paper Sites • Top 25 Essay Sites • Top 50 Essay Sites • Search 97,000 Papers @ DirectEssays.com • Search 101,000 Papers @ ExampleEssays.com • Search 90,000 Papers @ MegaEssays.com Free Essays • Term Paper Sites • Chuck III's Free Essays • Free College Essays • TermPaperSites.com • Free Essays • My Term Papers • Essay World • Planet Papers	Search Lots of Essays Back to Subjects - Computers TCPIP TCPIP Until recently, computer systems were isolated workstations that communicated with each other only with difficulty. Either it was a stand-alone unit, or if it was bought to be part of a system, you had to buy a network communication system to go with it. In the 1970’s and 1980’s, more sophisticated networks began to become more common. Dozens of architectures populated the networks of this time, with companies such as IBM, Digital, Sperry, Burroughs, and Honeywell being largely involved in this development. Problems arose when these proprietary networks needed to communicate to another workstation outside of their network. Later in the 1980’s, LANs became more commonly used to connect workstations, but they tended to utilize proprietary rules of communication called protocols. Protocols such as Novell’s IPX/SPX, AppleTalk, and IBM and Microsoft’s NetBEUI were better for communication, but they still presented problems. Getting a PC to talk to a mainframe required specialized technologies that would turn a “smart” PC into a dumb terminal that could be permitted within the mainframe’s sphere of influence. Near the end of the 1980’s, network isolation became unacceptable. Businesses realized that LANs served vital business needs, they were important databases and sharing the stored information was critical. TCP/IP, the network protocol, was invented by the Department of Defense due to its complex network requirements. TCP/IP provided a common protocol suite that could be specified for all networks, would operate with equipment from various vendors, was easy to configure, high performing, and reliable. TCP/IP is a protocol suite, or group of protocols, which allows communication over the Internet of in private networks ranging from home offices to enterprise networks. TCP/IP is constructed to cover the seven layers of the OSI model. The application, presentation and session layers are made up of the Telnet, FTP, TFTP, SMPT, LPD, NFS, SNMP, X Window, SQL, RPC, DNA, and SCP protocols. The transport layer is composed of the TCP and UDP protocols. The protocols included in the network layer are IP, ICMP, ARP, RARP, and BootP. The data link layer is made of LAN protocols, Ethernet: 100 BaseFx, 100 Base4, 100BaseTX, Token Ring FDDI and WAN protocols ISDN, PPP, SLIP, X.25, LAPB, Frame Relay, HDLC, SDLC, and other protocols. Finally, the physical layer is made of EIA/TAI-232, EIA/TIA-449, V.24, V.35, X.21, G.703, EIA-530, HSSI, and others. The information is transferred by packet switching, a process by which voice of data is converted into a stream of bytes. It is then broken down into packets that contain information defining the sender and receiver, and then is stored in RAM of the network devices along the path from the sender to the receiver. The Internet layer is responsible for delivering data through an internetwork, and the primary Internet layer protocol is the Internet Protocol (IP). IP uses other protocols for special tasks, such as ICMP, used to deliver messages to the host-to-host layer, and routing protocols, implemented to improve the IP’s routing efficiency. IP is a required protocol for Internet communication which primary functions include unique addressing, packet fragmentation and reassembly, and packet delivery. Upper layer TCP/IP protocols do not directly use network hardware address. Instead, they use a system of logical addresses for identifying hosts. These IP addresses provide several benefits, including making routing simple, and making TCP/IP resistant to changes in network hardware. For example, if a NIC card is exchanged, the IP address can remain the same. A TCP/IP host is very stable because a host configuration includes information about several other hosts, all expressed in IP addresses. If a default gateway is changed, IP makes the manual address change much simpler. This is important for Internet use because users and applications frequently access other hosts on the network. If host addresses were to change frequently, disseminating those changes to the network would be difficult. DNS significantly reduces the severity of this problem by allowing users to use names rather than numbers to identify hosts. IP addresses are 32 bits in length and are divided into a netid field to identify the network to which the host is attached, and a hostid field to assign each host on a given network a unique identifier. All TCP/IP hosts that occupy the same network must be assigned the same netid, and hosts with different netids must communicate through a router. Each network on an internetwork must be assigned a unique netid. There are five classes of IP address: Class A-beginning with a high-order bit of 0, with the first octet comprising the netid, the remaining three being the hostid. Class B-beginning with a high-order bits of 10, with the first two octets comprising the netid, the remaining two being the hostid. Class C-beginning with a high-order bits of 110, with the first three octets comprising the netid, the remaining one being the hostid. Class D-beginning with a high-order bits of 1110, are used to support multicasts. Class E-beginning with a high-order bits of 11110, are used for experimental purposes. All networks that connect to the Internet must be configured with the InterNIC assigned IP addresses. The host to host layer has two primary areas of responsibility, providing upper-layer processes and applications with a convenient network interface, and delivering upper-layer messages between hosts. Transmission Control Protocol (TCP) is a reliable protocol, which makes a concerted effort to deliver data to its destination, test for errors, resound if required, and to report errors to upper layers if a successful transmission is not achieved. TCP is a robust protocol with high network overhead, conversely, User Datagram Protocol (UDP) is used when reliability is not an issue, but speed is.TCP it is not concerned with routing data through the internetwork, because the network infrastructure is IP’s responsibility. At the host-to-host layer, TCP on one host communicates directly with TCP on another host, regardless of whether they are on the same network or remote from each other. TCP is not implemented on routers unless the router function is performed on a host that runs upper-layer processes, and in fact, it is oblivious to the network. It accommodates a variety of technologies, including circuit switching and packet switching on LAN’s and WAN’s. TCP is used to identify hosts using IP addresses and does not concern itself with physical addresses. The interface between TCP and a local process is a port, a mechanism that enables the process to call TCP and in turn enables TCP to deliver data streams to the appropriate process. Ports are identified by a port number, which can be either assigned manually, or assigned by the Internet Assigned Numbers Authority (IANA). To fully specify a connection, the host IP address is appended to the port number. This combination of IP address and port number is a socket, which is a unique number on an internetwork. A bi-directional communication path between two hosts is fully described by the sockets assigned to each end of the connection. Sockets provide an application-program interface (API) between TCP and processes and applications, which provide programmers with a clean interface between applications and TCP. There are two socket types used in TCP/IP, stream sockets, reliable, sequential, bi-directional socket used with TCP, and datagram sockets, which are used with UDP to provide unreliable, bi-directional data transfer. Data is processed as it travels down the protocol stack of the sender, through the network, and up the protocol stack of the receiver. First, TCP receives a stream of data from the upper-layer process, and it may fragment the data stream into segments that meet the maximum datagram size of IP, or to conform to restrictions of the network. Then network protocols transmit the datagram in the form of bits, and protocols at the receiving host reconstruct datagrams from the bits they receive. Then, IP receives datagrams, and where necessary, reassembles them to reconstruct the original segment. Finally, TCP presents the data in segments to upper-layer protocols in the form of data streams. Bibliography: Word Count: 1318
Copyright © 1998-2008 College Term Papers, INC All Rights Reserved. DMCA Notifications and Requests