Catálogo de publicaciones - libros

Many data grid applications manage and process huge datasets distributed across multiple grid nodes and stored into heterogeneous databases; e-Science projects need to access widespread databases within a computational grid environment, through a set of secure, interoperable and efficient data grid services. In the data grid management area several projects aims at addressing these issues providing different solutions and proposing a set of data access and integration/federation services.

In this paper we present the GRelC Data Access, a WS-I based data grid access service developed by SPACI Consortium and the University of Salento. Its architectural design is discussed and experimental results related to an European testbed are also reported.

The monitor construct has been implemented in several concurrent and/or parallel programming languages for shared-memory system environments, Extensions of the monitor to support process synchronization in distributed systems have also been proposed. But, most existing work only provides the architecture design of the distributed monitor. There is no discussion about the algorithmic and implementation issues. Also, none of them consider how to implement conditional variables. In this paper, we present the design and implementation of a distributed monitor construct, named DisMoniC, for programming process synchronization in distributed systems. DisMoniC is generic in the sense that it can be used with any distributed mutual exclusion (DME) algorithm to implement exclusive access to the monitor operations. Time-efficient algorithms are proposed to implement conditional process synchronization in the distributed monitor. We also present performance evaluation of the proposed construct.

Recently, information communication market is reorganized focused on the digital convergence service used for integrating the communication, broadcasting and internet media by connecting the communication equipment, broadcasting equipment and computer via a network. The mobility support protocol enabling the effective handover between heterogeneous networks is emerged as the hot issue in relation with these next generation communication networks. In this paper, we are to suggest the MIH Vertical handover procedure required for the implementation of MIH over MIPv4 Low-Latency vertical handover by analyzing the handover performance in the overlap area between wireless networks using the vertical handover function of MIPv4 Mobile IP Technology of IETF and considering the expected problems. Also, we are to suggest the MIH L2-Trigger generation algorithm required for deciding the optimal handoff. We evaluate the vertical handover performance of MIPv4 low-latency in the overlap area between wireless networks by using the NS-2 and applying the IEEE 802.21 MIH (Media Independent Handover) based on the suggested technique and algorithm.

A category of wireless sensor networks consists of lots of autonomous sensor nodes with limited power and few base stations with theoretically unlimited power. A number of redundant nodes are usually deployed densely in these types of networks in order to provide redundancy for sensing and communications. There is a challenge though of which nodes must be active and which ones must be asleep, without compromising the coverage and network connectivity. To get round this challenge, each node should somehow know the position of its immediate neighbors. Previous researches have impractically assumed the existence of a GPS module in each node, which is in direct contradiction with the main constraints of low cost and size of sensor nodes. This paper proposes an energy saving solution without requiring the nodes to possess any physical GPS. The goal is to minimize the number of active sensors with respect to coverage and connectivity. Each node decides locally by itself whether to be active or not. There is no need for any global synchronization between nodes. Simulation results show that the higher density of nodes in our proposed solution leads to better coverage, higher energy saving and longer network lifetime.

Recently superpeers are introduced to improve the performance of P2P systems. A superpeer is a node in a P2P system that operates as a server for a set of clients. By exploiting heterogeneity, the superpeer paradigm allows P2P systems to run more efficiently. This paper proposes a hierarchy-adaptive P2P topology DAHP2P and a hierarchical routing algorithm Hroute. Peers are grouped into clusters according to proximity and super peers form the upper-level overlay, the number of hierarchy is self-adaptively changed according to the number of nodes in the system, a hierarchical routing algorithm is designed to reduce the routing hops. Simulation results show that Hroute can significantly reduce the expected number of hops and latency of message routing, and loads of peers at different layers are relatively balanceable.

One of the key challenges in the optimal control of adaptive multimedia services transmission in mobile cellular network is to balance user satisfaction and fairness among all users, while at the same time ensuring that the scarce bandwidth be utilized efficiently. We propose a novel Bandwidth Degradation Policy () for Adaptive Multimedia Services which could solve the conflicting relationship between user satisfaction and fairness. The employs specific degradation policy to accommodate more users and compensation mechanism to maintain fairness among all users, and meanwhile increasing the user satisfactions and the whole network revenue. Specifically, we introduce four new metrics to evaluate the performances of adaptive multimedia services in mobile cellular networks. Finally we compare our with other strategies in the literature through both of the analytical results and simulation results. The presented extensive simulation results with concerns of mobility validate our analysis.

We demonstrate in the paper that user movement has great influences on the system performance of mobile cellular networks. We also develop a system model to study to what extent the influences will be, with focusing on the user movement simulation. We concern the parameters that determine the system performance of mobile cellular networks. We show partially in this paper that these parameters vary while user movement changes, but the effects are interactive. Since handoff calls are given a higher priority over new calls and mechanism is employed, our analysis shows that a mature mobile cellular network should make use of user movement characteristics and dynamically adjust its reservation threshold of the handoff call requests. Extensive simulation results validate our analysis.

In wireless mesh network, by equipped mesh router with multiple radios tuned into orthogonal channels, throughput improvement problem can be alleviated. Efficient channel assignment and link scheduling is essential for throughput improvement. Effective channel assignment schemes can greatly relieve the interference effect between nearby transmissions. However, not only the links in wireless mesh network using different channels can be activated at a time, but some links in the same channel also can be activated concurrently if the SNIR (Signal-to-Noise and Interference Ratio) at their receiver endpoints is not lower than the threshold. In this paper, we investigate the problem of how to schedule a set of feasible transmission under physical interference model by using the Spatial TDMA access scheme and channel assignment in wireless mesh networks. We also consider the fairness enhancement to prevent some border nodes of the network from starvation. By using Minimum Spanning Tree as network subgraph constructed from original network graph, we propose centralized algorithms for scheduling and channel assignment to maximize the aggregate throughput and to provide the fairness of the network. We also evaluate the throughput improvement and fairness enhancement of our algorithms through extensive simulations and the results show that our algorithm can achieve significant aggregate throughput and fairness performance.

The use of MANET’s (or Mobile Ad hoc NETworks) is becoming very popular. Power efficiency is a key issue in this type of network, as mobile devices usually rely on limited power supplies. One essential service, the routing protocol, employed to discover routes between nodes in the network, can greatly affect power consumption.

Furthermore, many distributed applications require an additional membership service to keep track of the nodes that make up the system at any moment. In general, this information is not provided by routing services with the exception of the Optimised Link State Routing protocol (OLSR).

The two services, routing and membership estimation, form a basic support to build other higher–level distributed services on ad hoc networks. To decrease the over-all power consumption these services should be optimized for the intended use of the network. In particular the degree of mobility can have an impact on the power consumption and performance of different approaches to routing and membership estimation.

In this paper we present a study of two different approaches that combine a routing service with membership estimation. We compare the proactive OLSR with our own approach. Our approach consists of integrating a gossip-style failure detector with the reactive Dynamic Source Routing protocol (DSR).

We present an analysis of the effects of mobility on the global performance and power consumption of the two approaches. We identify scenarios for which each approach is best suited.

By means of analysis and generalization of the hypercube and its variations of the same topological properties and network parameters, a family of interconnection networks, referred to as binary recursive networks, is introduced in this paper. This kind of networks not only provides a powerful method to investigate the hypercube and its variations on the whole, but also puts forth an effective tool to explore new network structures. A constructive proof is presented to show that binary recursive networks are Hamiltonian based on their recursive structures, and an approach to prove 4-pancyclicity of a subfamily of binary recursive networks is outlined.