RAIN: always on data warehousing

  • Authors:

  • Jorge Vieira;Marco Vieira;Marco Costa;Henrique Madeira

  • Affiliations:

  • Critical Software SA, Coimbra, Portugal;CISUC, Department of Informatics Engineering, University of Coimbra, Coimbra, Portugal;Critical Software SA, Coimbra, Portugal;CISUC, Department of Informatics Engineering, University of Coimbra, Coimbra, Portugal

  • Venue:
  • DASFAA'08 Proceedings of the 13th international conference on Database systems for advanced applications
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Redundant Arrays of Inexpensive DWS Nodes (RAIN) technique is a node-level data replication approach that introduces failover capabilities to DWS (Data Warehouse Striping) clusters. RAIN is based on the selective replication of fact tables' data across the cluster nodes and endows DWS clusters with the capability of providing query answers even when one or more nodes are unavailable. Two distinct replication modes are supported: simple redundancy (RAIN-0) and stripped redundancy (RAIN-S). In this demo we are going to show a DWS cluster using the RAIN technique, focusing on the execution of queries in the presence of nodes failures and on the process of recovering failed nodes.