Primal-dual parallel algorithm for continuous aggregate query dissemination

Abstract

Minimum weight set cover problem is an NP-hard problem and many approaches are there to handle the problem including greedy algorithm and primal-dual approach. In data dissemination networks the division of client issued continuous aggregate queries into sub-queries and assignment of these sub-queries optimally to different data aggregators of data dissemination networks is a minimum weight set cover problem. In this paper we propose a primal-dual parallel algorithm for continuous aggregate query dissemination. A dynamic data dissemination network is a content delivery network implemented with a hierarchical network of data aggregators which is used in online applications like stock applications and meteorological applications. Optimal dissemination of continuous aggregate queries and designing algorithm for optimization is a very important issue in such applications. Our algorithm is best suited for such distributed applications.