Pchrome: indexing the distributed materialised views

Abstract

The distributed query processing suffered from identifying which views are materialised over various distributed sites and optimally communicating these queries for the materialised views from node to node. This paper focuses on using MOLAPs layered structure for indexing the materialised views in the distributed scenario. We propose a new approach Pchrome to index the materialised views in the distributed architecture. Unlike other indexing methods which indexes on the data of the data cube, Pchrome indexes on the lattice structure of the MOLAP data cube, making it more efficient for query processing. We also demonstrated how our method reduced distributed communication cost. We implemented our approach on data cube without concept hierarchies. As a future enhancement we are studying the method with concept hierarchies.