Alan Southall

A Framework for Resource Management in Peer-to-Peer Networks

The huge amount of recent activities in peer-to-peer (P2P) computing led to the development of a number of applications and protocols all dealing with similar problems and providing different noninteroperable implementations. In this paper, we present a framework for resource management and information storage in P2P networks. Our framework aims at providing an easy to use and flexible model for P2P computing encouraging more modularized application development and permitting reuse of components. To provide efficient localization of resources and representation of information, our resource management framework (RMF) forms a distributed information space enabling not only flat storage of information but forming graph structures that can be used to build new kinds of applications enabling the users to browse the contents of a P2P network.

The resource management framework: a system for managing metadata in decentralized networks using peer-to-peer technology

Given the current explosion in peer-to-peer based protocol implementations, application developers require a means of abstracting the individual characteristics of specific peer-to-peer implementations away from their application logic. The Resource Management Framework provides a unified model of peer-to-peer computing which is independent of the underlying protocols. To this end, a data model and APIs for both application and peer-to-peer protocol developers are provided which hide many of the intricate details, such as database management and redundancy, inherent in all peer-to-peer systems. A first application of the Resource Management Framework has been demonstrated via the implementation of a FIPA compliant, distributed, ad-hoc Directory Facilitator for the Java Agent DEvelopment Framework (JADE).

Adaptive and fault-tolerant service composition in peer-to-peer systems

Service-orientation enables dynamic interoperation of distributed services and facilitates seamless service provision or runtime creation of new applications. This dynamic service composition is particularly powerful in peer-to-peer (P2P) systems which offer scalability through self-management and autonomy. However, P2P service composition is nontrivial due to permanent peer churn and lack of central control. Existing approaches reduce composite service initialization to an NP-hard path finding problem. Thus, peer failure adaptation is costly and runtime consideration of peer logons or load changes is not practicable. This paper introduces logical peer groups for service composition. They enable runtime composite service reconfiguration including the migration of services to other peers. A prototype implementation is presented and the algorithms are evaluated through both formal and empirical analysis. The evaluation shows that the approach results in significant reduction of computational complexity, improves fault-tolerance and enables adaptation of logons and load changes which has not been possible so far.

Mobile network self-configuration based on distributed adaptive transactions

ABSTRACT The assurance of network-wide configuration data consistency in mobile networks is a complex task due to the degree of distribution of the network elements (NEs), the properties of the O&M network and the configuration dependencies between the network elements. To cope with these characteristics, an approach is presented, where configuration changes are communicated as transactions between the NEs and their element management system (EMS). A master-replica paradigm is adopted where the replica (=NE) are allowed to autonomously commit configuration data as “tentative” and the master (EMS) follows the state of the replicas, but may force rollbacks on them. Configuration dependencies between different NEs are expressed as transaction groups. The execution of these groups is controlled by a centralized transaction manager located at the EMS. To provide for a defined duration of the transaction group execution in the presence of transaction failures, either a complete rollback or a partial commit of the transaction group (thus re-assessing the configuration dependencies) is possible based on operator policy. The presented scheme allows to improve the level of configuration data consistency and the degree of automation in the configuration operations with benefits for both the manufacturer and the Mobile Network Operator. It can be used in parallel to legacy configuration management protocols.