Computer Communication Laboratory
Dept. of computing, Soonsil University
 

DNS with IPv6

Over the last 20 years, the Internet has been fast growing accompanied with tremendous popularity of WWW. It is beyond doubt DNS has had absolute necessity on this stage, and up to now. Additionally, rapid distension of Internet leads us resource exhaustion, especially exhaustion of address. When we mention address, it is a 32-bit-length identifier of IPv4 for internet protocol in TCP/IP stack. Immediately, we know there are 2^32 addresses to identify individual host in a network, and it becomes a critical problem to support rapidly increased hosts. Several new mechanisms (e.g., CIDR, NAT, DHCP and VPN) have proposed and used, but any of them does not become a radical solution. The last expedient was a new protocol, IPv6 that uses 128-bit-length address and accommodates us with copious address resources, but it is not the end of story. Transition to IPv6 faces many difficulties. Firstly, most of implementations and hosts are based on IPv4, and IPv6 should coexist with IPv4, so moving on IPv6 is not built in one day. Secondly, some people do not want to IPv6, because they think IPv4 with NAT or DHCP is enough. Besides, name resolution for IPv6 is not supported from any root or intermediate name servers, even if hosts is based on dual-stack or on IPv6, when the host wants communicate other end point (host), he should remember the 128-bit-length address of other. Looking at another angle, it is a quietly tedious job to type IPv6 address.

For more than 10 years, the usage of IPv6 itself has been controversial, and absence of DNS name lookup service for IPv6 and hesitation of transition from IPv4 to IPv6 retard reformation. To clear the way for IPv6, we have to enable name lookup service (including reverse name lookup) for users, and moreover, speed up development of IPv6 based applications. We conduct researches with DNS for IPv6 as the central figure.


Copyright. Computer Networking and Mobile Computing Team All right reserved.