Showing papers by "Ralph Deters published in 2003"

The costs of using JXTA

Abstract: Project JXTA is an open-source effort to specify the standard protocols for peer-to-peer communication and collaboration. We propose a JXTA performance model and present results obtained by benchmarking the JXTA 1.0 reference implementation in Java. We focus on the performance evaluation of typical peer operations and consequences for the peer network, the user and the developer. The important trade-off between peer startup latency and the maintenance of the local cache is shown and discussed. The throughput limits of pipes, the core JXTA communication concept, are also measured in a LAN environment for smooth and bursty traffic. The results indicate that the limiting factor for reliable throughput is the number of messages rather than size in bytes, as well as that small JXTA messages carry an excessive overhead of control data. Important performance issues and trade-offs are identified and explored, as a basis for the formulation of guidelines for system designers and simulation-based research of JXTA networks.

3LS - a peer-to-peer network simulator

Abstract: Peer-to-peer (p2p) networks are the latest addition to the already large distributed systems family. With a strong emphasis on self-organization, decentralization and autonomy of the participating nodes, p2p-networks tend to be more scalable, robust and adaptive than other forms of distributed systems. The much-publicized success of p2p-networks for file-sharing and cycle-sharing have resulted in an increased awareness and interest into the p2p protocols and applications. However, p2p-networks are difficult to study due to their size and the complex interdependencies between users, application, protocol and network. We present a 3-level simulator designed to study complex p2p networks.

JXTA performance study

Abstract: Project JXTA is the first peer-to-peer application development infrastructure that includes standard protocols and multi-language implementations. Performance and scalability of JXTA are not well understood, despite its widespread usage in research and increasing popularity in the industry. This paper reports on a performance evaluation of the JXTA protocol implementations in Java. The peer startup and pipe performance are evaluated for the JXTA protocol versions 1.0 and 2.0. The results show that the performance of the latter JXTA version has improved for secure message transfer and multicast propagation, but degraded for several other operations.

Using dynamic proxy agent replicate groups to improve fault-tolerance in multi-agent systems

Abstract: Despite the considerable effort spent researching and developing multi-agent systems the actual number of deployed systems is still surprisingly small. One of the reasons for the significant gap between developed and deployed systems is their brittleness. Multi-agent systems are susceptible to all of the same faults as any distributed system, they lack centralized control components, which makes it difficult to detect and treat failures of individual agents, and the agents making up the system are social, thus risking fault-propagation These faults can impact system performance and lead to overall system failure. Multi-agent systems must be made more dependable before they will be deployed on a large scale. Using redundancy by replication of individual agents within a multi-agent system is one possible approach for improving fault-tolerance, and hence improving reliability and availability-two key components of dependability. Having a group of agents, a replicate group, act like an individual agent leads to increased complexity and system load, and it introduces new challenges to system construction. Using a message proxy, to handle communication for the group, and passive replication strategies effectively deals with the complexity and overhead issues. This paper presents an architecture for implementing agent replicate groups using a message proxy and passive replicate group management. Experimentation and application testing using an implementation of the architecture is presented. The architecture is demonstrated to be a viable technique for increasing dependability in multi-agent systems.

Improving fault-tolerance in MAS with dyanmic proxy replicate groups

Abstract: This paper presents a technique for replicating agents in a multi-agent system (MAS) with a goal of improving the fault-tolerance of the system. Replicating agents, or forming a replicate group, will always add complexity and overhead to a system. To mitigate these effects passive replication and a proxy agent is used to represent and manage the group. So that a new single point of failure (the proxy) is not created, the proxy is dynamic and can move to any agent in the replicate group. An implementation and experimentation is presented which shows that the technique is viable.