Comparison of two approaches to processing long aggregates lists in spatial data warehouses

Abstract

In this paper we present a comparison of two approaches for storing and processing of long aggregates lists in a spatial data warehouse. An aggregates list contains aggregates, calculated from the data stored in the database. Our comparative criteria are: the efficiency of retrieving the aggregates and the consumed memory. The first approach assumes using a modified Java list supported with materialization mechanism. In the second approach we utilize a table divided into pages. For this approach we present three different multi-thread page-filling algorithms used when the list is browsed. When filled with aggregates, the pages are materialized. We also present test results comparing the efficiency of the two approaches.