Self-CAN: Structured P2P with Self-* Properties
This page is under construction.
Self-CAN is a P2P system that inherits the ability of CAN-like structured systems for the construction and maintenance of an overlay of peers, but features enhanced functionalities deriving from ant-inspired algorithms, such as autonomy behavior, self-organization and capacity to adapt to a changing environment. As opposed to the structured P2P systems deployed so far, resource indexing and placement is uncorrelated with network structure and topology, and resource keys are organized and managed by self-organizing mobile agents through simple local operations driven by probabilistic choices.
This site gives a quick introduction to Self-CAN and offers a Self-CAN simulator.
Some questions...
1) Can Structured P2P Systems be Self-Organizing?
So far, self-organizing and adaptive properties have always been associated to unstructured P2P systems. Structured systems are considered too rigid to present self-* behaviours
Self-CAN tries to confute this belief. It exploits the structured topology of CAN, but features self-organizing mechanisms for key ordering and discovery
2) How are self-organizing properties obtained in Self-CAN?
Thanks to the use of mobile agents, which travel the CAN multi-dimensional grid and order resource keys
Mobile agents operations are inspired by the behavior of ants, in accordance with the swarm intelligence paradigm
3) What are the benefits vs. classical structured systems like CAN?
Essentially three:
A better load balance among the peers can be obtained
Dynamic properties (e.g., management of joining nodes) are improved. For example, it is not necessary to reassign keys when new peers or resources are added to the system: the mobile agents will spontaneously reorganize the keys
Finally, and most notably, by decoupling the space of the peers from the one of the keys, Self-CAN allows a semantic meaning to be associated with the coordinates of the keys, so that each dimension of the key space can be used for a different attribute of the resources. In this fashion, multi-dimensional queries can be solved easily and efficiently, which makes Self-CAN particularly well suited to Grid and Cloud environments.
The Self-CAN simulator is available here.
Also, a Self-CAN prototype is available here.