Angelica Garcia 05 12 2006

From EECS

OLAP Pre-Aggregation for Raster Image Databases

Angelica Garcia Gutierrez

Abstract:

Data warehouses using a multidimensional view of data have become very popular in both business and scientific applications. These application domains pose several challenges to existing data warehouse technology, particularly to ensure fast query performance for data volumes in the order of Terabytes or higher. Pre-aggregation, that is, the pre-computation and storage of data, constitutes the strength of the On-Line Analytical Processing (OLAP) technology and has proved to speed up query response times by factors of 100 and in some cases up to 1000 [Kimball, Winter]. However, the appliance of OLAP pre-aggregation in raster image databases has been surprisingly neglected.

A comparative study between raster image and business databases has shown similarities on their operations and data structures. Therefore, it becomes natural to think in exploiting the knowledge gained with OLAP pre-aggregation for more over a decade, and adapt it to raster image databases. This is the core topic of my research project, and in this talk I will introduce it formally. I will also show the current state of the project as well as the next goals to be reached