Design and Implementation of Traditional DNS Protocol

Hosts need to identify their location on the Internet. One way is to use an IP address. In IPV4 networks, each computer is identified by a unique 4-byte number, which is generally written as a four-point format, such as 199.1.35.90. When data are transmitted via the network, the header of the packet will contain the IP address of the destination host. Another method is to use the host name to identify themselves, such as www.yahoo.com, www.google.com. Compared with the IP address, it is more memorable. However, the host name provides little or no information about the host's location in the network. In addition, since host names are usually composed of alphanumeric characters of varying lengths, routers are better at handling fixed-length, hierarchical IP addresses. Therefore, domain name system (DNS) has been developed for converting host names that are easy to be remembered by humans into digital Internet addresses. But the use of DNS sometimes reduces the efficiency of accessing the network because the network environment varies all the time, which results in the transmission rate of information between the client and the DNS server. If the package losses due to timeout, the respond of the DNS server cannot be received. As a result, this paper designs a model that can decrease the probability of these problems by introducing a local DNS database.