Distribution of dns - application layer , Computer Networking

Distribution of DNS

1.By centralized data base

The information contained in the domain  name space must be stored. A simple design  or DNS would have one DNS server that contains  all the mappings. In this centralized design. Client simply direct all  queries to the single  DNS server, and the DNS  server responds  directly  to the querying  clients. The problems  with a  centralized  design  include.

a.Traffic Volume: A single DNS server would have handle  all DNS queries (for all  the HTTP request and  e mail message  generated from  hundreds  of millions  of hosts.

b. A single  point of failure. It the DNS server  crashes so does  the entire internet.

c. Distant Centralized Database: A single  DNS  server cannot  be close  to all  querying client. If  we put  the single  DNs  server in new York city  then all quarters  form  Australia must travel  to the  other  side of the globe perhaps over slow and  congested links. This can  lead to  significant delays.

d.aintenance: This single  DNS  server would  have to keep records for all internet hosts.  Not only  would  this centralized database be huge,, but it would  have to be updated frequently to account  for every new host to those  servers. In  the internet there are 13 root DNS serves ( Ladled a through M)  most of which  are located  in north  America.

In summary  single  DNS with centralized data base has lot  of problems. To  overcome  these problems  DNS should b presented in a particular  hierarchy. Show as below.

Posted Date: 3/8/2013 6:36:05 AM | Location : United States







Related Discussions:- Distribution of dns - application layer , Assignment Help, Ask Question on Distribution of dns - application layer , Get Answer, Expert's Help, Distribution of dns - application layer Discussions

Write discussion on Distribution of dns - application layer
Your posts are moderated
Related Questions
IPv6 The  internet  engineering task force begin  the effort to develop a successor  protocols  to IPv4 in the early  1990 is an internet  layer protocols  for packet  switched

XYZ company's executives have approached you to architect, design, and recommend if they should use an IP network infrastructure for all their applications. Here is some informa

After studying this part the students will be able to understand about the following: Analysis of Parallel Algorithms; Different Models of Computation; o   Interco

What is the basic difference between latches and flip flops? Ans) flip flops have clock signal whereas latches dont require anyclock pulse..

Q. Explain TCP-IP Routing Architecture? The TCP/IP routing architecture looks extremely much like the OSI routing architecture. Hosts utilize a discovery protocol to o

What are Triggers and Rules? Triggers are special user defined actions generally in the form of stored processes that are automatically invoked by the server based on data rel

network diagram which must include at least 2router,2bridge and 4 switches and must have different protocols and different cable...?

Parallel Balance Point In order to implement a parallel algorithm on a parallel computer, K number of processors is needed. It may be noted that the given input is assigned to

DVMRP • Distance vector multicast routing protocol: Distance vector multicast routing protocol are abbreviated as DVMRP. These protocols are suitable for small networks which

Q. Illustrate about Sliding Window Protocols? - Sliding window - Keeps the unacknowledged outstanding frames in sender - Keeps the expected frames in receiver - Sequen