Data Warehouse Decentralization Strategy

A Data Warehouse (DW) is characterized by a huge amount of data centralized in a single database. This centralization has negative impacts on its performance especially for distributed companies; such us elevated storage cost, high query execution time and hard maintenance task. Since a relational data warehouse has the same physical structure as a classical database, it can enjoy all the benefits realized during the past in distributed databases like data availability, simplicity, rapid local data access and transparent access to remote sites. Then, it would be interesting to manage and test its decentralization on several distant sites. However, very few studies have considered this topic. In this paper, we present a relational DW decentralization strategy. We show experimentally with the use of the APB-1 release II benchmark that DW decentralization gives better performance than the centralized context. Global queries execution time is reduced by 80%.

[1]  Anne Tchounikine,et al.  A model for distributing and querying a data warehouse on a computing grid , 2005, 11th International Conference on Parallel and Distributed Systems (ICPADS'05).

[2]  Ken Barker,et al.  A horizontal fragmentation algorithm for the fact relation in a distributed data warehouse , 1999, CIKM '99.

[3]  Anca Georgiana Fodor.  Aspects of Data Allocation in Distributed Database Systems , 2007 .

[4]  Henrique Madeira,et al.  Handling big dimensions in distributed data warehouses using the DWS technique , 2004, DOLAP '04.

[5]  Patrick Valduriez,et al.  Principles of Distributed Database Systems , 1990 .

[6]  Christie I. Ezeife,et al.  Dynamic Database Object Horizontal Fragmentation , 1999 .

[7]  Matteo Golfarelli,et al.  Vertical Fragmentation of Views in Relational Data Warehouses , 1999, SEBD.

[8]  Krithi Ramamritham,et al.  Curio: A Novel Solution for Efficient Storage and Indexing in Data Warehouses , 1999, VLDB.

[9]  Shamkant B. Navathe,et al.  An objective function for vertically partitioning relations in distributed databases and its analysis , 2005, Distributed and Parallel Databases.

[10]  Ladjel Bellatreche,et al.  An Evolutionary Approach to Schema Partitioning Selection in a Data Warehouse , 2005, DaWaK.

[11]  Pedro Furtado.  Workload-Based Placement and Join Processing in Node-Partitioned Data Warehouses , 2004, DaWaK.

[12]  Alejandro P. Buchmann,et al.  Research Issues in Data Warehousing , 1997, BTW.

[13]  Cristina Dutra de Aguiar Ciferri,et al.  Horizontal fragmentation as a technique to improve the performance of drill-down and roll-up queries , 2007, SAC '07.

[14]  Habib Ounalli,et al.  Experimental Evidence on Data Warehouse Fragmentation and Allocation in a Distributed Context , 2010, KMIS.

[15]  Christie I. Ezeife,et al.  Incremental Horizontal Fragmentation of Database Class Objects , 2003, ICEIS.