The term internet is simply an abbreviation of internetworking, which means the interconnection of networks. The name Internet is the internetworking of thousands, or even millions, of sub networks on a global scale that has been growing at a prodigious rate since the early 1990s. The sub networks can be LANs, or WANs. The Internet can also be viewed as an enormous global WAN. The objective here is to explain the relationship between this huge computer network and the telecommunications network. It is probably fair to say that without the telecommunications network infrastructure the Internet would not exist. The biggest problem facing telecommunications network the fact that telephone networks were designed for telephone voice connections but are now required to send large amounts of data down those same lines. Adjusting to the incompatibilities of the two requirements is not a simple matter. Voice telephone traffic (POTS) still accounts for more than 50 percent of all telecommunications revenues worldwide. Telecommunications networks are nevertheless being used for networking, but a serious gap is already opening up between today's throughput requirements and the throughput that telecommunications companies can actually deliver. Furthermore, all this seemed to happen overnight without any single body being able to plan its evolution. Fortunately, some of the "rules of the game" for Internet information flow are remarkably simple. The two protocols- transmission control protocol (TCP) and Internet protocol (IP), usually written as TCP/IP are the key to allowing one sub network to communicate with another. They correspond to the OSI model, where the TCP is the transport layer and the IP part is the network layer. Because their functions are transparent to the physical and data link layers they can be applied to Ethernet, FDDI, or token ring networks.

The TCP/IP was created by the U.S. Defense and Advanced Research Projects Agency (DARPA), with the original intention of connecting research centers and universities to promote rapid information flow and dialogue. The costs of the interconnections through the telephone network were originally paid for by research grants or direct government funds. The commercial benefits are changing the Internet landscape and no doubt eventually the costs will all be passed on to Internet users.UP

The initial purpose of file transfer and remote computer log-on have now become a minor activity compared to information location by "browsing." Telecommunications service providers are finding it difficult to dimension their telephone networks adequately. A large part of Internet use is by individuals who pay monthly subscriptions to an Internet service provider for dial-up connection. Whereas the average phone call used to be 3 to 5 min, it is rising significantly as many browsers stay on the line for hours. Serious data flow bottlenecks are developing that will probably only be alleviated by the eventual deployment of all-optical networks.

The TCP/IP is used by the Internet to transfer data. Every Internet node has its own unique network address known as its IP address. A node is also known as a TCP/IP host and is the start or end point for communications over the Internet. Hosts are computers themselves which run software applications. TCP/IP gateway nodes connect different types of network together within the Internet, such as token ring and Ethernet networks. Both hardware and software are needed to perform this function. Gateways route TCP/IP packed between networks. A router is a computer that connects similar types of networks together, such as two or more token rings.

The Internet uses a connectionless packet-based protocol, which means there is no established route for the series of packets to take. Every packet has its own addressing information to find its way to the destination independently of other packets. This method has pros and cons. A major advantage is that if one or more routers in the system fail, it does not affect the overall flow of packets because they simply find another path through the maze of routers. This benefit is offset by the fact that each router has to do a route determination for every packet it receives. The IP has no flow control mechanism, so when a router becomes congested it deletes excess packets until the congestion is reduced.

The Internet datagram. Routing through the network is done by an addressing scheme written into each packet. All packets are constructed in the Internet datagram format which is composed of a header followed by information data. A maximum of 65,536 bytes can be sent in one datagram and larger blocks of data are split into two or more datagrams. The header contains all the routing information. First, the TCP/IP version number allows nodes and gateways to interpret the datagram correctly as newer versions become available. The type of service 8 bits take the form PPPDTRXX and characterize the datagram as follows. PPP is a number from 0 to 7 to define the priority, D defines a low delay, T defines high throughput, R defines high reliability, and XX is for future use. The header length describes the length of the header to follow in multiples of 4 bytes normally up to 20 bytes (without options). The D-bit is to inform gateways that the data must not be fragmented into smaller data blocks. Gateways can split data into fragments and the M-bit denotes that data is fragmented. The fragment offset indicates the number of the fragment (not the fragment quantity). Depending on traffic circumstances, a datagram can be delayed anywhere from small fractions of a second to never arriving at all. The time-to-live 8-bit words place a maximum datagram transfer time value in seconds. At each gateway, the value is reduced automatically. On reaching zero, the datagram is deleted. This time-to-live word is very useful because it also gives the destination node a maximum time it should wait for the next datagram fragment. Other Internet protocol variants can be used in the datagram, and the 8-bit protocol word indicates which one is used. The header checksum is a special 16-bit word that is used for error detection. The source IP address and the destination IP address are both unique 32-bit words to allow the datagram to reach the desired location. Finally, the options field provides routing information and other items such as error control.

 wpe95622.gif (86330 bytes)
 The internet datagram.

The IP address.

Every device on the Internet has a unique identifying number, called an IP Address Currently there are two types of Internet Protocol (IP) addresses in active use: IP version 4 (IPv4) and IP version 6 (IPv6). IPv4 was initially deployed on 1 January 1983 and is still the most commonly used version. IPv4 addresses are 32-bit numbers often expressed as 4 octets in "dotted decimal" notation typical IP address looks like this:192.165.110.115. An Internet Service Provider (ISP), or network administrator, permanently [fixed or static] or dynamically [for the duration of a single session] assigns an IP address to a network device. An IPv4 address is 32 bits wide, it is composed of two parts: the network number, and the host number [1, 2, and 3]. By convention, it is expressed as four decimal numbers separated by periods, such as "200.1.2.3" representing the decimal value of each of the four bytes. Valid addresses thus range from 0.0.0.0 to 255.255.255.255, a total of about 4.3 billion addresses. The first few bits of the address indicate the Class that the address belongs to:

Class Prefix

Network Number

Host Number

A

0

Bits 0-7

Bits 8-31

B

10

Bits 1-15

Bits 16-31

C

110

Bits 2-24

Bits 25-31

D

1110

N/A

E

1111

N/A

The bits are labeled in network order, so that the first bit is bit 0 and the last is bit 31, reading from left to right. Class D addresses are multicast, and Class E is reserved. The range of network numbers and host numbers may then be derived:

Class

Range of Net Numbers

Range of Host Numbers

A

0 to 126

0.0.1 to 255.255.254

B

128.0 to 191.255

0.1 to 255.254

C

192.0.0 to 254.255.255

1 to 254

Any address starting with 127 is a loop back address and should never be used for addressing outside the host. A host number of all binary 1's indicates a directed broadcast over the specific network. For example, 200.1.2.255 would indicate a broadcast over the 200.1.2 network. If the host number is 0, it indicates "this host". If the network number is 0, it indicates "this network" [2]. All the reserved bits and reserved addresses severely reduce the available IP addresses from the 4.3 billion theoretical maximum. Most users connected to the Internet will be assigned addresses within Class C, as space is becoming very limited. This is the primary reason for the development of IPv6, which have 128 bits of address space. Deployment of the IPv6 protocol began in 1999. IPv6 addresses are 128-bit numbers and are conventionally expressed using hexadecimal strings (for example, 1080:0:0:0:8:800:200C:417A).

IN The IP address format a 32-bit word. 232 - 4,294,967,296, which means over 4 billion hosts or physical networks can be addressed with this length of digital word. No doubt 32 bits will be inadequate a few years from now and will have to be increased in length. The 32-bit word is split into two parts, to identify a particular network number and a specific host within that network. Three address formats have been defined, depending on the quantity of networks and the quantity of hosts to be internet worked. The IP address can be specified as four decimal numbers, designated by the letters W, X, Y, and Z, where each decimal number represents 1 byte (8 bits) of IP address. Because 28 = 256, the maximum decimal number for each of the four letters W, X, Y, or Z is 255. The network information center (NIC) assigns IP addresses. This address format allows quite a lot of flexibility and, for example, sub networks can be defined within a large network.

The 32-bit IP address is difficult to remember so user names are usually given to colleagues and friends, and application programs are used to convert between the two. For example WWW.IEEE.COM denotes a World Wide Web server for the Institute of Electrical and Electronics Engineers, based in the United States, and the corresponding Internet IP address is 140.98.1.1.

Each institution within the Internet has its host which incorporates a domain name server (DNS) that has a directory of all hosts within the network. In this manner, the Internet addressing scheme is the equivalent of a telephone numbering plan. Interestingly, the telephone network is used as part of the interconnection of networks, gateways, and hosts. The main difference in connectivity of addressing is created by the leased line. The computer community uses the term being on-line. That simply means there is a permanent connection between a user's computer and the others in the network. The network can be just one other computer in the next office with an Ethernet LAN connection or it can be a computer across the other side of the world within, say, a large university campus LAN. Telephone companies have set up numerous leased lines to allow, for example, a permanent connection between a gateway in San Francisco and a gateway in New York. New Internet users get the impression that such long distance connections are free. They are not. With speech over the Internet improving in quality and becoming ever more popular, telecom service providers are taking a serious look at the definition of both a long-distance and a local call.