DBLab School of Computer and Electrical Engineering KDBSL NTUA
Tuesday, July 07, 2020

1. Introduction

Data warehousing & On-Line Analytical Processing (OLAP) are core database technologies that comprise the driving forces behind modern business intelligence. The need for the novel integration of many incompatible data sources, the efficient storage and querying of vast amounts of multidimensional hierarchical data and the expressive presentation of results to the end user pose many new problems.

2. Research Issues

-ETL (Extraction Transformation Loading)
T. Sellis, A. Simitsis, M. Terrovitis, P. Georgantas

The Extraction, Transformation and Loading of data comprise one of the most challenging processes in a data warehouse. Tools that perform the aforementioned processes are collectively named ETL tools. The main tasks that ETL tools perform are (a) discovering of required information at the source site, (b) extraction of selected information, (c) transformation and consolidation of information from multiple sources to a common format, (d) data cleansing according to database integrity constraints and business rules, (e) storage of the transformed data in the data warehouse and/or update commence of materialized views. Our research focuses on:

  • Issues on representing and modeling ETL processes
  • Description languages for ETL processes
  • ETL process optimization

-Data Warehousing - OLAP
T. Sellis, A. Tsois, N. Karayiannidis)

OLAP has caused a significant shift in the traditional database query paradigm. Queries have become more complex and entail the processing of large amounts of data. Our study efforts aim at:

  • Designing sophisticated data structures for efficient storage and querying of multidimensional data with hierarchies
  • Optimization of complex ad-hoc queries containing aggregates and blocking operators (e.g., GROUP BY)

-OLAP Visualization
Y. Vassiliou, A. Maniatis

OLAP deals with the analysis and suitable presentation to end users of multidimensional data retrieved from Data Warehouses. In that sense, the introduction of suitable presentation models, the adoption, adaptation and exploitation of relative display techniques (Information Visualization) and their integration into real case applications and environments cover a large part of this area. Our research activities include:

  • Presentation models for OLAP
  • Information Visualization Techniques for OLAP data
  • Presentation and Visualization Models for Mobile OLAP.
3. Selected Publications

N. Karayannidis, T. Sellis, Y. Kouvaras: CUBE File: A File Structure for Hierarchically Clustered OLAP Cubes. In Proc. of the EDBT'04 International Conference, Heraklion, Greece, 2004.

A. Tsois, T. Sellis: The Generalized Pre-Grouping Transformation: Aggregate-Query Optimization in the Presence of Dependencies. In Proc. of the VLDB'03 International Conference, Berlin, Germany, 2003.

P. Vassiliadis, A. Simitsis, P. Georgantas, M. Terrovitis: A Framework for the Design of ETL Scenarios. In Proc. of the CAiSE'03 International Conference, Klagenfurt/Velden, Austria, 2003.

N. Karayannidis et. al.: Processing Star Queries on Hierarchically-Clustered Fact Tables. In Proc. of the VLDB'02 International Conference, Hong Kong, China, 2002. 

4. Funding

EDITH - European Development of Indexing Techniques for Databases with Multidimensional Hierarchies, European Comission, IST, 2000-2002

5. Contacts

Prof. Timos Sellis                      
Phone: +30-1-772-1601, Fax: +30-1-772-1442
e-mail: timos@dblab.ece.ntua.gr

Prof. Yannis Vassiliou
Phone: +30-1-772-2526, Fax: +30-1-772-2527
e-mail: yv@dblab.ece.ntua.gr

School of Electr. and Comp. Engineering
Computer Science Division
National Techn. Univ. of Athens
Zographou, 157 73 Athens, Greece

[ Back ]