Catálogo de publicaciones - libros

Range queries, retrieving all keys within a given range, is an important add-on for Distributed Hash Tables (DHTs), as they rely only on exact key matching lookup. In this paper we support range queries through a balanced tree algorithm, Decentralized Balanced Tree, that runs over any DHT system.

Our algorithm is based on the -tree  design that efficiently stores clustered data while maintaining a balanced load on hosts. The internal structure of the balanced tree is suited for range queries operations over many data distributions since it easily handles clustered data without losing performance.

We analyzed, and evaluated our algorithm under a simulated environment, to show it’s operation scalability for both insertions and queries. We will show that the system design imposes a fixed penalty over the DHT access cost, and thus inherits the scalability properties of the chosen underlying DHT.

In a decentralized network system, an authenticated node is referred to as a Byzantine node, if it is fully controlled by a traitor or an adversary, and can perform destructive behavior to disrupt the system. Typically, Byzantine nodes together or individually attack point-to-point information propagation by denying or faking messages. In this paper, we assume that Byzantine nodes can protect themselves from being identified by authentication mechanisms. We present an authentication-free, gossip-based application-level propagation mechanism called LASIRC, in which “healthy” nodes utilize Byzantine features to defend against Byzantine attacks. We show that LASIRC is robust against message-denying and message-faking attacks. Our experimental studies verify LASIRC’s effectiveness.

When several heterogeneous database systems are combined to a federation it is necessary to deal with the problem of maintaining the consistency of semantically interrelated data. The integrity constraints that are defined on the schemas of the federation’s members must be monitored and ensured. In this paper we introduce the COMICS system, a global nstraint anager for nteractive omponent Database ystems, that is able to ensure strict consistency in a collection of autonomous and heterogeneous databases. The system uses the functionality of enhanced active databases to trigger a global integrity check from a local update operation.

In this paper, we provide an evaluation model for marketable quality and profitability. We define the marketable quality as a qualitative aspect of profitability. We apply the real values of some leading manufacturing corporations in Japan to our proposed model to analyze its accuracy. From the analysis, we found that theoretical and real standard values of the marketable quality indicator were very close. This shows that the proposed model has a good approximation. From the fair relation of network service providers and users, we present the network pricing guidelines and extend our proposed network service pricing model considering network externalities.

Information systems have to be consistent and secure in presence of multiple conflicting transactions. The role-based access control model is widely used to keep information systems secure. Here, a role is a set of access rights, i.e. permissions. A subject is granted a family of roles, i.e. one or more than one role. A subject is allowed to issue a method to an object only if an access right is included in the roles granted to the subject . In the access control models, even if every access request satisfies the access rules, illegal information flow might occur as well known confinement problem. In this paper, we define a legal information flow relation () among a pair of role families and . This means, no illegal information flow occur if a transaction with a role family is performed prior to another transaction with . In addition, we define which role families are more significant than others in terms of types of methods and security classes of objects. Conflicting methods from different transactions are totally ordered in the significancy of roles of the transactions. We discuss how to synchronize transactions so as to prevent illegal information flow and how to serialize conflicting methods from multiple transactions in terms of significancy and information flow relation of roles families.

Multi-agent technology and constraint satisfaction techniques can be effectively combined and utilized to solve many real-world problems. This paper presents a multi-agent system based on Probabilistic Interval Algebra (PIA) networks to solve distributed scheduling problems. An IA network is a graph where nodes represent intervals and directed edges are labelled with temporal interval relations. A PIA network has probabilities associated with the relations on the edges that are used to capture preferences. The proposed multi-agent system consists of PIA-Agents that are connected via edges to form a network. Each PIA-Agent has ownership and control over a PIA network. A prototype is implemented using JXTA and demonstrated on a university domain to show how the PIA-Agents work together and coordinate their activities to recommend a temporal schedule which is a globally consistent solution which attempts to maximize the desires of each individual PIA-Agent.

On-demand broadcast is an effective approach to disseminating real-time data to mobile clients in mobile environments. Recent studies mostly assume that clients request only a single data item. This assumption may not sufficiently support the increasingly sophisticated real-time applications in mobile environments. In this paper, we present three on-demand broadcast algorithms to cater for multi-item requests with timing constraints. We argue for the case of aggregation of data requests to minimize the broadcast bandwidth and deadline miss rate. The three algorithms try to exploit different properties of system characteristics for aggregation. Our simulation results show that a good aggregation algorithm should consider both data overlapping of client requests and their urgency.

Grid technologies congregate numerous computers to provide powerful computing and massive storage. In data-intense applications, Data Grids are developed to cope with the efficiency of data access. Replication is one of the methods to elevate the access performance. When a file is replicated, it can be downloaded from all the nodes with that file in parallel, thus reducing the access latency. Therefore, a fair and adaptive replication strategy for high speed transmission is important. In this paper, we design such a strategy to duplicate popular files to beneficial nodes in the grid networks. All the users deserve the same quality of transmission. The contributions of our mechanism are to average the transmission cost and evenly distribute the workload for download. Simulation results are also given to demonstrate the performance of our replication strategy.

In this paper we tackle the issues of exploiting the concepts of social networking in processing similarity queries in the environment of a P2P network. The processed similarity queries are laying the base on which the relationships among peers are created. Consequently, the communities encompassing similar data emerge in the network. The architecture of the presented metric social network is formally defined using the acquaintance and friendship relations. Two version of the navigation algorithm are presented and thoroughly experimentally evaluated. Finally, learning ability of the metric social network is presented and discussed.

Flooding is an obligatory message dissemination technique for network-wide broadcast within mobile ad hoc networks (MANETs). The conventional blind flooding algorithm causes broadcast storm problem, a high number of unnecessary packet rebroadcasts – thus resulting in high contention and packet collisions. This leads to significant network performance degradation. Because of the highly dynamic and mobile characteristics of MANETs, an appropriate probabilistic broadcast protocol can attain higher throughput, significant reduction in the number of rebroadcast messages without sacrificing the reachability. This paper proposes a new probabilistic approach that dynamically fine-tunes the rebroadcasting probability for routing request packets (RREQs). We assess the performance of the proposed approach by evaluating it against the ad-hoc on demand distance vector (AODV) routing protocol (which follows blind flooding approach), fixed probabilistic approach and the existing dynamic probabilistic approaches. The simulation results reveal thatthe proposed approach demonstrates better performance than the existing approaches.