[ Pobierz całość w formacie PDF ]
.The protocolsuite was integrated into the University of California at Berkeley's UNIXoperating system and became available to the public for a nominal fee.Fromthat point TCP/IP became widely used.Its spread to other operating systemsresulted in increasing use in both local area network (LAN) and wide areanetwork (WAN) environments.Today, TCP/IP enables corporations to merge differing physical networkswhile giving users a common suite of functions.It allows interoperabilitybetween equipment supplied by multiple vendors on multiple platforms, and itprovides access to the Internet.In fact, the Internet, which has become thelargest computer network in the world, is based on TCP/IP.So why has the use of TCP/IP grown at such a rate? The reasons include theavailability of common application functions across differing platforms and theability to access the Internet, but the primary reason is that of interoperability.The open standards of TCP/IP allow corporations to interconnect or mergedifferent platforms.An example is the simple case of allowing file transfer78 Getting Started with Data Warehouse and Business Intelligence capability between an MVS/ESA host and, perhaps, a Hewlett Packardworkstation.TCP/IP also provides for the routing of multiple protocols to and from diversenetworks.For example, a requirement to connect isolated networks usingIPX, AppleTalk, and TCP/IP protocols using a single physical connection canbe accomplished with routers using TCP/IP protocols.One further reason for the growth of TCP/IP is the popularity of the socketprogramming interface between the TCP/IP transport protocol layer andTCP/IP applications.A large number of applications today have been writtenfor the TCP/IP socket interface.4.1.1.1 TCP/IP ArchitectureTCP/IP, as a set of communications protocols, is based on layers.UnlikeSystem Network Architecture (SNA) or Open Systems Interconnection (OSI),which distinguish seven layers of communication, TCP/IP has only fourlayers.See Figure 17.The layers enable heterogeneous systems tocommunicate by performing network-related processing such as messagerouting, network control, and error detection and correction." Application LayerThe application layer is provided by the program that uses TCP/IP forcommunication.Examples of applications are Telnet, File TransferProtocol (FTP), e-mail, Gopher and SMTP.The interface between theapplication and transport layers is defined by port numbers and sockets." Transport LayerThe transport layer provides communication between applicationprograms.The applications may be on the same host or on differenthosts.Multiple applications can be supported simultaneously.Thetransport layer is responsible for providing a reliable exchange ofinformation.The main transport layer protocol is TCP.Another is UserDatagram Protocol (UDP), which provides a connectionless service incomparison to TCP, which provides a connection-oriented service.Thatmeans that applications using UDP as the transport protocol have toprovide their own end-to-end flow control.Usually, UDP is used byapplications that need a fast transport mechanism.Data Communication 79 Figure 17.TCP/IP Architecture Model: Layers and Protocols" Internet LayerThe Internet layer provides communication between computers.Part ofcommunicating messages between computers is a routing function thatensures that messages will be correctly delivered to their destination.TheInternet Protocol (IP) provides this routing function.Examples of Internetlayer protocols are IP, ICMP, IGMP, ARP and RARP." Network Interface LayerThe network interface layer, sometimes also referred to as link layer, datalink layer or network layer, is implemented by the physical network thatconnects the computers.Examples are LAN (IEEE 802.x standards),Ethernet, X.25, ISDN, ATM, Frame Relay, or asynch.Note that the RFCs actually do not describe or standardize any networklayer protocols by themselves, they only standardize ways of accessingthose protocols from the Internet layer.4.1.1.2 IP AddressingIP uses IP addresses to specify source and target hosts on the Internet.(For example, we can contrast an IP address in TCP/IP with a fully qualifiedNETID.LUNAME in SNA).An IP address consists of 32 bits and is usually80 Getting Started with Data Warehouse and Business Intelligence represented in the form of four decimal numbers, one decimal number foreach byte (or octet).For example:00001001 01000011 00100110 00000001 a 32-bit address9 67 38 1 decimal notationAn IP address consists of two logical parts: a network address and a hostaddress.An IP address belongs to one of four classes.The class to whichan IP address belongs is determined by the value of its first four bits.SeeFigure 18.Figure 18.Assigned Classes of IP Addresses" Class A addresses use 7 bits for the and 24 bits for the hostportion of the IP address.That allows for 126 (2**7-2) networks with16777214 (2**24-2) hosts each; a total of more than 2 billion addresses." Class B addresses use 14 bits for the network and 16 bits for the hostportion of the IP address.That allows for 16382 (2**14) networks with65534 (2**16-2) hosts each; a total of more than 1 billion addresses.Data Communication 81 " Class C addresses use 21 bits for the network and 8 bits for the hostportion of the IP address.That allows for 2097150 (2**21) networks with254 (2**8-2) hosts each; a total of more than half a billion addresses." Class D addresses are reserved for multicasting (a sort of broadcasting,but in a limited area, and only to hosts using the same class D address)." Class E addresses are reserved for future use.Some values for these host IDs and network IDs are preassigned and cannotbe used for actual network or host addressing:" All bits 0Stands for this: this host (IP address with =0) or thisnetwork (IP address with =0).When a host wants tocommunicate over a network, but does not know the network IP address, itmay send packets with =0.Other hosts on the networkinterpret the address as meaning this network.Their reply contains thefully qualified network address, which the sender records for future use." All bits 1Stands for all: all networks or all hosts.For example:128.2.255.255Means all hosts on network 128.2 (class B address).This is called a directed broadcast address because it contains both avalid and a broadcast." LoopbackClass A network 127.0 is defined as the loopback network.Addressesfrom that network are assigned to interfaces that process data inside thelocal system and never access a physical network (loopback interfaces).4.1.1.3 SubnetsBecause of the explosive growth of the Internet, the principle of assigned IPaddresses became too inflexible to facilitate changes to local networkconfigurations.Such changes might occur when:" A new type of physical network is installed at a location." Growth of the number of hosts requires splitting the local network into twoor more separate networks." Growing distances require splitting a network into smaller networks, withgateways between them.82 Getting Started with Data Warehouse and Business Intelligence To avoid having to request additional IP network addresses in these cases,the concept of subnets was introduced.The assignment of subnets can bedone locally, as the whole network still appears to be one IP network to theoutside world.Recall that an IP address consists of a pair and.For example, let us take a class A network; the address format isshown in Figure 19.Figure 19
[ Pobierz całość w formacie PDF ]