Search

During the last decade, data management and storage have become increasingly distributed. Advanced query operators, such as skyline queries, are necessary in order to help users to manipulate the huge amount of available data by identifying a set of interesting data objects. The optimization of those queries becomes an open research area especially in distributed contexts. In this paper, we tackle the problem of efficient skycube computation in structured P2P systems. We introduce a top-down algorithm called Distributed-Top-Sky based on the recently introduced Top-Sky. Furthermore, we introduce two types of nodes namely the Scheduling-Node where the network is organized by assigning the computation of each cuboid to a Data-Node and the Data-Node which holds a part of the dataset used to compute the assigned cuboids. In order to evaluate the effectiveness of our approach, we have conducted extensive experiments on three real datasets over a simulated CAN (content addressable network) network.

It is a big pleasure for the user to find the result of its skyline query without waiting for its processing; this can be realized by getting a saved version of this query. This paper focuses on the pre-computation of the skylines of all possible nonempty subsets of a given set of dimensions, what we call the skycube. We develop an efficient top-down approach called TOP-SKY for the skycube computation which derives the skyline objects from one of the subspace's parents adopting some techniques that help to achieve a better performance. In order to evaluate the effectiveness of the approach, TOP-SKY has been compared with the best algorithm in our knowledge Orion and with computing the cuboids of the skycube individually using BNL algorithm.

The materialisation of views has been one of the most successful techniques in optimising OLAP queries. In this context, many works have been conducted aiming either at studying the relationships between the views or at solving the materialised views selection problem. For skyline queries, researchers investigated only the first phase of the materialisation whereas the second phase which is the materialised skyline views selection problem has never been studied because of the difficulty of creating a cost model. In this paper, we propose a genetic method for the materialised skyline views selection problem. In order to reduce the high cost produced by evaluating the skycube all over the selection process, we propose a time reducing heuristic called OnceVisited which avoids the computation of the same query from the same views many times. The conducted experiments have proven the efficiency of the genetic approach since high quality solutions have been found. In addition, the proposed time reducing heuristic OnceVisited has been shown to be very efficient.