Paper review: IP Next Generation Overview

Reviewer: Hanlin D. Qian

The current Internet Protocol has many problems, among which the most crucial is the scarcity of available IP addresses. With the birth of new markets that demand the use of more IP addresses, it is necessary to create a new IP scheme that supports more available Internet addresses.

This paper presents a solution to the problem by creating a new version of Internet Protocol - Ipv6, or IP Next Generation (IPng). IPng includes a 128-bit address space, which is significantly larger that Ipv4's 32-bit address space. Note that 2^128 is a value that is larger than the total number of particles in the Universe. So even if every particle gets an address, we still won't run out. This large address space can easily accommodate the growth of networked nomadic personal computing devices, of different types of network entertainment, and of networked electronic devices.

In addition to solving the address space problem, IP Next Generation also made other improvements:

  1. A new type of address called an "anycast address" is defined. The use of anycast address allows nodes to control the path through which their traffic flows.
  2. The Header information for IP packets is simplified.
  3. Improved support of header options creates more flexibility and scalability.
  4. Traffic Flows can now be identified with what's called a "flow label." This label allows senders to request special handling of the packets by routers.
  5. IPng has two types of new security headers that promote authentication, integrity, and confidentiality.

The most import issue of the paper is that of transition from IP to IPng. After all, like any software version upgrades, IPng is useless if it is not backward compatible with the existing Internet. This paper presents a way to upgrade from IP to IPng. The idea is to embed IPv4 addresses within IPv6 addresses. Also, The author encourages what he calls incremental upgrades and deployment. Thus the transition can be made in a smooth and gradual manner, and the operation of the Internet will not be disrupted.

It is difficult to give an objective review of the paper, especially because it was written six years ago, and one may be tempted to judge the success of the paper based on how successful IPv6 was implemented in the past six years. I give the paper a rating of 4 (significant contribution) because it identified and attempted to solve some major problems with the current Internet Protocol. These problems are still present and need to be addressed, especially in the 21st century, when the Internet continues to grow in size, in the types of services, and in the importance of its services. Although one can argue that the transition from IP to IPng has not been so successful in the past six year, further research can produce better solutions. Because of its inherent problems, IPv4 cannot last forever. Therefore the research presented in this paper is still valid in solving some fundamental problems of the current Internet.

The author has presented a solution. Now research must to done to implement and test this solution, especially in the area of transition from IPv4 to IPv6. This area is where research is most needed.