In
Internet to identify a system or a host, the IP address is used. It is very
difficult for people to memorize all IP addresses and people prefer to use
names instead of numeric addresses. To deal with this situation we need a
system that can map between name and address. When the Internet was small,
mapping between IP address and host name can be done by using a single file.
The file consist two columns, one for host name and other store corresponding
IP address. Today it is difficult to
store all addresses in a single file for mapping. The solution is, divide this
large amount of data into smaller parts and store each part on a different
computer. Each host that needs a mapping
can contact the nearest system which
store the needed information. This method is used in the DNS.
One
essential requirement is that the name must be unique. A name space is used to
assign unique names. In the case of flat namespace, there is no predefined
structure for assigning names, but hierarchical namespace assign name space
based on a structure. In the case of hierarchical namespace the name is divide
into smaller
parts, each parts have a specific meaning. The domain name space
is an example of hierarchical namespace. In this case, names are representing
by using an inverted-tree with root at top. The tree can have maximum of 128 levels. |
| Domain name space |
The label
for each node in the tree has at most 63
characters. The label of the root node is null. A full domain name is a sequence of labels separated by dots. A Fully Qualified Domain Name (FQDN)
always ends in a null label. If a label
is not terminated by a null string, it is called a Partially Qualified Domain Name (PQDN). A domain is a
sub-tree of the domain name space.
FQDN:
air.ac.in.
PQDN:
air.ac
The
domain name hierarchy is divided among many servers. What a server is
responsible in domain space is called a zone. The zone of the root server
consist the whole tree. DNS defines two types of servers: primary and secondary.
A primary server stores a file about the zone for which it is responsible. It
use local storage to store the zone file and it is responsible for creating, maintaining,
and updating the zone file. A secondary server store information about zone
from another server. The secondary
server is used to create redundancy for the information so that if one server
fails, the other can serve answer for client requests. The process of
downloading information from primary to secondary is called zone transfer.
The
domain name space is divided into three different regions in Internet: generic
domains, country domains, and the inverse domain. The generic domains define hosts according to their generic
behavior. For example mil for military groups and org for nonprofit organizations.
The country domains use two-character to
represent a specific country. For example, in for India ,
us for United States .
The inverse domain is used to map an
address to a name. Mapping an address to a name
or a name to an address is called name-address
resolution.
DNS has two types of messages: query and response. The
query message contains a header and question records, the response message
contains a header, question records, answer records, authoritative records, and
additional records.
 |
| DNS Header format |
DNS
can use either TCP or UDP. DNS server use port 53. UDP is used when the size of
the response message is less than 512 bytes and TCP used if the size of the response
message is more than 512 bytes.
Previous : Email Architecture
No comments:
Post a Comment