General issues

The problems I've encountered and solutions I proposed only deal with a subset of the problems one can encounter when creating a P2P application. To create a usable and scalable system many other issues will also have to be resolved. I will briefly touch upon these issues to show the complexity of the field. It must be said, however, that this section is slightly outside the scope of our research.

These problems mainly deal with the infrastructure of P2P networks. While early networks used excessive amounts of bandwidth for network management alone this problem is slowly being dealt with (17). However, the current situation is still far from ideal and can seriously harm future growth (18).

Luckily, P2P technology has seen a new enthusiasm in the last few years from the academic society. If anywhere it is here that we can expect well-founded solutions to emerge. Counting not only on intuition, but increasingly on mathematical proofs, researchers have so far been looking into network topology and ways of exploiting this structure. Examples of this can be found in papers proposing slight adjustments to existing protocols () and others that detail new protocols based on other topologies (19).

A field of research I believe interesting solutions can be found is that of ad-hoc networking. Ad-hoc networks are wireless networks lacking a centralized backbone infrastructure. They therefore exhibit the same dynamic behaviour of P2P nets. Especially solutions dealing with node discovery and intelligent routing can be transferred to P2P applications. Routing protocols geared especially towards a dynamic environment that can be used in P2P applications are for instance the Ad-hoc On-Demand Distance Vector Routing protocol (AOVD) (20) and the Dynamic Source Routing Protocol (DSR) (21). A comparison of a number of protocols has also been undertaken (22).