Salvatore Riccobene

An analytical comparison of the client-server, remote evaluation and mobile agents paradigms

We deal with the study of the actual convenience of using the agent programming paradigm for accessing distributed service. We point out the benefits of such a communication paradigm by providing an analytical study of its basic features in comparison with the client-server approach and remote evaluation. The results that we have obtained show how agents must not always be considered the only solution to any communication issue, since in several cases their use might even reveal a drawback. We present several models of non-Markovian Petri nets, which have been solved through the WebSPN tool, and we provide a close comparison between the agent techniques (client-server and remote evaluation). We also focus our attention on providing some practical remarks which can help the developer during the design, in order to select the communication paradigm which best suits the features of the application that has to be developed.

Which paradigm should I use? An analytical comparison of the client–server, remote evaluation and mobile agent paradigms

In this paper we deal with the study of the actual convenience of using the agent programming paradigm for accesssing distributed service. We try to point out the benefits of such a communication paradigm, by providing an analytical study of its basic features in comparison with the client–server approach and remote evaluation. The aim of the paper is to show how the Petri net analysis technique can be used for deciding whether to use traditional client/server, remote evaluation or mobile agents paradigm in designing a particular evaluation. So, we present several models of non‐Markovian Petri nets, which have been solved through the WebSPN tool, and we provide a close comparison between the agents technique, the client‐server and the remote evaluation communication paradigm. The results that we have obtained show how agents must not always be considered the only solution to any communication issue, since in several cases their use might even reveal a drawback. We also focus out attention on providing some practical remarks, which can help the developer during the design in order to select the communication paradigm which best suits the features of the application that has to be developed. Copyright

Research: Monitoring performance in distributed systems

Continuous monitoring of computer network performance is probably the only solution which allows prompt identification of anomalous functioning conditions and provides knowledge of parameters on which to base recovery interventions. A number of tools for performance management already exist, but their effectiveness is limited as they are essentially inserted inside owner network management solutions. In this paper we describe the realization of PMt, a platform for the development of performance management applications for the control and real-time management of a heterogeneous computer network. PMt uses an object-oriented approach in managing the various elements in the distributed system, and defines a methodology for the design of applications to monitor performance. The methodology is mainly destined for use on distributed systems, and provides the user with a set of tools and services that can be used to develop management applications.

Managing Reputation over MANETs

The use of small portables and mobile devices has made MANETs (mobile ad hoc networks) very popular. A MANET is a network composed by a group of mobile nodes without any fixed device or a central coordination. They work in an open net and their collaboration is the sole means to allow communications and the survival of the MANET itself. A critical issue is to assess the behaviour of the nodes that participate in the network, possibly identifying selfish conduct that can compromise the functioning of the system. This paper shows a method to evaluate the behaviour of all nodes by establishing a reputation value that represents the trustworthiness of each node. A protocol is presented to calculate the reputation of a node by locally observing the node from another one, and then tuning this intermediate value with additional observations from other participants. When the reputation value of a node is available, it is circulated and distributed uniformly over the network. This reputation protocol is viable. Each node can efficiently calculate the reputation values of its neighbours and then of all network nodes. A variety of simulations conducted using the network simulator NS-2 strongly support these claims.

Enforcing privacy in e-commerce by balancing anonymity and trust

Privacy is a major concern in e-commerce. There exist two main paradigms to protect the customers privacy: one relies on the customers trust that the network will conform to his privacy policy, the other one insists on the customers anonymity. A new paradigm is advanced here as a natural balance between these two. It sees the customer act using his real identity but only circulate cover data that conceal the resources he requires. Privacy enforcement is thus shifted from the customers identity to his purchase preferences. The new paradigm is suitable for scenarios such as eBay purchases where trust that a network sticks to a privacy policy is problematic, while anonymity is either forbidden or impossible. The computation of cover data is done by a node other than the customer in order to minimize impact on the customer. That node will therefore see the customers private data that are used to compute the cover. This demands some technology to prevent the node from exposing private data. An existing protocol developed for self-enforcing privacy in the area of e-polls is thoroughly analysed and found somewhat weak in terms of fairness among its participants. A stronger version is designed and adopted, together with an innovative differential-privacy preserving function, in the new privacy paradigm. The strengthened e-poll protocol and the new differential-privacy preserving function, which strictly speaking only are side contributions of this paper, each appear as important as the new e-commerce privacy paradigm.

Design and evaluation of a multimedia storage server for mixed traffic

Abstract. The relative simplicity of access to digital communications nowadays and the simultaneous increase in the available bandwidth are leading to the definition of new telematic services, mainly oriented towards multimedia applications and interactivity with the user. In the near future, a decisive role will be played in this scenario by the providers of interactive multimedia services of the on-demand type, which will guarantee the end user a high degree of flexibility, speed and efficiency. In this paper, some of the technical aspects regarding these service providers are dealt with, paying particular attention to the problems of storing information and managing service requests. More specifically, the paper presents and evaluates a new storage technique based on the use of disk array technology, which can manage both typical multimedia connections and traditional requests. The proposed architecture is based on the joint use of the partial dynamic declustering and the information dispersal algorithm, which are employed for the allocation and retrieval of the data stored on the disk array. We also define efficient strategies for request management in such a way as to meet the time constraints imposed by multimedia sessions and guarantee good response times for the rest of the traffic. The system proposed is then analyzed using a simulation approach.

Java-based and secure learning agents for information retrieval in distributed systems

Intelligent agents represent a powerful tool to implement distributed functions in a network environment. This paper describes the current status of a research project whose target consists in using this new technology for automatic information retrieval, network management and security purposes within a distributed system. Specifically, we describe the software agents dedicated to information retrieval and to security aspects, focusing on architectural aspects and implementation issues. The system has been implemented using the Java language.

Design and performance analysis of a disk array system

We concentrate on the architectural issues of parallelizing I/O access in a disk array system by means of definition of a new, particularly flexible architecture, called partial dynamic declustering, which is fault-tolerant and offers higher levels of performance and reliability than the solutions normally used. A simulation analysis highlights the efficiency of the proposed solution in balancing the file system workload and demonstrates its validity in both cases of unbalanced loads and expansion of the system. Particular attention is also paid to the definition of analytical models, based on stochastic reward nets, in order to analyze the performance and reliability of the system. The response time distribution function is evaluated and a specific performance analysis with varying degrees of declustering and workload is carried out. >

Distributed Backup through Information Dispersal

The formal aspects underlying a novel distributed backup service are discussed. Strength and originality of the service lie in the combined adoption of an established information dispersal algorithm with a simplified version of an existing location service. Information dispersal makes our service threshold-secure in that the backup owner only needs participation of a pre-established threshold number of nodes to recompose a distributed backup. This means that the service is highly available as it tolerates a number of node breakdowns. Even the right threshold number of nodes cannot retrieve the backup on their own initiative. The location service adopted allows our service to work over non-organized, flat networks. Indirect advantages are the optimization of the total redundancy of data and the efficient management of resources. Our service has reached the stage of a proof-of-concept implementation.

Evaluation of Performability Parameters in Client–Server Environments

The recent development of high-speed communication networks has led to great interest in client-server environments. In this context it is becoming increasingly important to have adequate tools to support the study of the behaviour and size of the various parts of these systems. This paper provides a systematic approach to modelling and assessing a particular class of system-fault-tolerant parallel/distributed system-which feature a capacity to reconfigure dynamically on the occurrence of partial faults, so as to guarantee a certain level of performance, albeit a reduced one. Using the formalism offered by Petri Nets, the paper presents combined measures of dependability and performance which identify the behaviour of the system in as complete a way as possible, above all in relation to fault/repair phenomena.