Uncertainty processing is a core task in many applications of distributed systems. Typical distributed systems have local processing nodes to collect information, which usually contain uncertainties, and do computational work. The nodes can interact with each other, and they evolve with time. A promising way of modelling and processing uncertainties in these systems is to use graphical models to form beliefs about the required information. Dynamic probabilistic networks for distributed uncertainty processing are presented in this paper. Two approaches are given, and comparison shows that the model with the more state-of-art approach performs better. Since it is not possible to obtain enough knowledge to construct an exact model at the beginning, the model needs to adjust itself when evolving. Therefore we have developed a parameter update algorithm to make the model adapt to the changing environment. Experiments are presented to show the effectiveness of the models and the algorithms.

This paper focuses on the design and prototype of a version-enabled peer-to-peer content distribution system based on distributed hash table. The system is different from the CVS system; it manages the version at item level, not project level. The system provides the user version management functionality, which is absence in most of the current peer-to-peer content distribution network. In addition, the version management functionality can be used as a concurrent version system in team collaboration system. In such application, it provides more features such as serverless, robustness and reliability. The peer-to-peer nature of our system is more suitable to the human collaboration model than the client-server model. The system is now being prototyped using OpenDHT on Java2 platform

Abstract The research of transaction processing in Web Services and Grid Services is active in academic and engineering areas. However, the formal method of transaction processing has not been fully investigated in the literature. This paper proposes a preliminary theoretical model called Membrane Calculus based on Membrane Computing and Petri nets to formalize Grid transactions. Five kinds of transition rules in Membrane Calculus (including object rules and membrane rules) are introduced and the operational semantics of transition rules are defined. Then, a typical long‐running transaction example is presented to demonstrate the use of Membrane Calculus. Finally, the rewriting logic tool Maude is adopted to specify and execute the specification of this example. Copyright © 2006 John Wiley & Sons, Ltd.

We study two key problems arising from mobile peer-to-peer media streaming: the stability of interconnection between supplying peers and requesting peers in the mobile peer-to-peer streaming system; and fast capacity amplification of the entire mobile peer-to-peer streaming system. Many algorithms for the fast capacity amplification of the peer-to-peer streaming system have been proposed. However, these algorithms are inefficient in the mobile ad-hoc networks. Topological dynamics of mobile peer-to-peer are exacerbated by the changes of mobile user's location and interest. We observe that the mobile users exhibit correlated mobility patterns in practical ad-hoc networks. We use the stable group algorithm to characterize user mobility in mobile ad-hoc networks. Based on the stable group, we then propose a distributed stable-group differentiated admission control algorithm (SGDAC/sub p2p/), which leads to fast amplifying of the system's total streaming capacity using self-growing. Finally, extensive simulation results are presented to compare between the SGDAC/sub p2p/ and traditional methods to prove the superiority of the algorithm.

This paper studies the problem of categorical data clustering, especially for transactional data characterized by high dimensionality and large volume. Starting from a heuristic method of increasing the height-to-width ratio of the cluster histogram, we develop a novel algorithm -- CLOPE, which is very fast and scalable, while being quite effective. We demonstrate the performance of our algorithm on two real world datasets, and compare CLOPE with the state-of-art algorithms.