Parallel k-skyband computation on multicore architecture

Abstract

Given a set of elements D in a d-dimensional space, a k-skyband query returns all elements which are worse than at most k other elements. The k-skyband query is a fundamental analysis query. It can offer minimal candidate set for other queries such as top-k ranking query where the ranking functions are monotonic. With the development of multicore processors, it has been a trend that more algorithms focus on parallel execution to exploit the computation resources. In this paper, rather than use traditional methods to compute k-skyband, we employ parallel techniques to get the result efficiently based on multicore architecture. Extensive experiments show the efficiency and effectiveness of our proposed algorithm. © 2013 Springer-Verlag.