HOW MANY CLASS A, B, C NETWORKS AND HOSTS ARE POSSIBLE

<< WHAT IS A NETWORK

NETWORKING DEVICES >>

E-COMMERCE IT430

Lesson 03

HOW MANY CLASS A, B, C NETWORKS AND HOSTS ARE POSSIBLE?

Note that 126 class A networks are possible (0 and 127 are not used for general purposes). Each class A

network can accommodate a maximum of 16 million hosts on it. There are 16384 class B networks possible

and each can host upto 65000 host machines. There are 2 million class C networks possible each having the

capacity to accommodate upto 256 host machines on it.

Subnet mask

An IP address is meaningless for the computers unless it is accompanied by a subnet mask. It tells to the

computer machines that what part of the accompanying IP address corresponds to the network address and

what part corresponds to the host machine on that network. A subnet mask is also represented by four

decimal numbers separated by a period. Default subnet masks for class A, B and C types of IP addresses are

as under:

Class A: 255.0.0.0

Class B: 255.255.0.0

Class C: 255.255.255.0

IP version

The existing IP scheme (version 4) is likely to be replaced by IP version 6. It would provide 128 bits IP

addresses in hexadecimal format. According to an estimate a total of 3.4 x 10 addresses would then be

available.

Domain name system

A Domain Name is a user friendly name used to locate a web site on the internet. For example, vu.edu,

bestcomputers.com etc. Domain Name System (DNS) provides the structure and the strategy that is used to

refer to computers on the internet by these user friendly names. Domain Names are Unique. They are

assigned as one has to pay and register for them. With the help of DNS a domain name is translated into its

corresponding IP address (see Fig. 1 below). A fully qualified domain name is processed from right to left

for its translation into the corresponding IP address. A fully qualified domain name can be made up of a top

level domain (TLD), second level domain (SLD) and subdomains, as shown in Fig. 2 below. Seven

popular TLDs are ".com,.edu,.org,.gov,.net,.countryname,.int". SLDs represent the name of a

company/institution/entity. Subdomains represent the geographical or functional units of a

company/institution etc.

E-COMMERCE IT430

Fig. 1

Fig. 2

Name resolution

The translation or resolution of a fully qualified domain name into its IP address takes place using the

hierarchy of special computer machines called Domain Name Servers. A DNS server is a server on the

network that maintains a database/table that contains the list of domain names and their corresponding IP

addresses. The name servers for TLDs are also called Root Name Servers. There is a program called

`Resolver' built into the browser, which sends the request for resolution of the domain name to the machine

called Local Name Server. Local name server then contacts the root name server, which looks into its

database and where possible provides the IP address of the name server below in the Hierarchy. Hence in a

similar number of steps the fully qualified domain name is completely translated into the IP address of the

web server where the requested web page is stored. The process of name resolution has been shown in an

example in Fig. 3 below:

E-COMMERCE IT430

Name Resolution

Root

Name

Server

ecom.cs.vu.edu

vu.edu, 128.196.128.233

Local

ecom.cs.vu.edu

Name

Name

ecom.cs.vu.edu

Client

Server

server

cs.vu.edu, 192.12.69.5

192.12.69.60

ecom.cs.vu.edu

Name

Server

Getting domain names and IP addresses

Domain names are administered in a hierarchy. At the global level the task of registration/administration of

domain names is supervised by the organization called Internet Corporation for Assigned Names and

Numbers (ICAAN). There are organizations working under ICAAN in different regions. For example,

APNIC is for Asia and Pacific Rim, ARIN for America and South Africa, RIPE-NCC for Europe and

North Africa etc. Each of the above are further connected to Internet Service Providers (ISPs) at the local

level. One can register one's domain name through an ISP also. Similarly, one can lease IP addresses from a

local ISP apart from the direct sources of IP addresses, that is, APNIC, ARIN or RIPE-NCC. ISPs can use

a server to dynamically supply the IP addresses to their clients for a session. Such a server is called DHCP

(Dynamic Host Configuration Protocol) server.

Media access control (MAC) address

Data to be delivered on a network has to be converted first into serial transmission. This is done through a

device called Network Interface Card (NIC) that must be installed in a computer on the network. NIC has

also got printed on it a 48 bits unique address called the MAC or hardware address of the computer

machine. In other words, it consists of 12 hexadecimal characters and can have different format as shown

below:

090017A9B2EF

09:00:17:A9:B2:EF

09-00-17-A9-B2-EF

Whereas the IP address of a computer may change from time to time, its MAC address would remain the

same unless the existing NIC is replaced with another. It can easily be inferred that there are 248 unique

MAC addresses possible (which is an unimaginably high number). An organization called Institute of

Electrical and Electronic Engineers (IEEE) administers the allocation of MAC addresses, worldwide. It

ensures that there is no duplication of MAC addresses by the manufacturers of NICs.

Table of Contents: