Physical Design Refinement: The "Merge-Reduce" Approach

Abstract. Physical database design tools rely on a DBA-provided workload to pick an "optimal" set of indexes and materialized views. Such an approach fails to capture scenarios where DBAs are unable to produce a succinct workload for an automated tool but still able to suggest an ideal physical design based on their broad knowledge of the database usage. Unfortunately, in many cases such an ideal design violates important constraints (e.g., space) and needs to be refined. In this paper, we focus on the important problem of physical design refinement, which addresses the above and other related scenarios. We propose to solve the physical refinement problem by using a transformational architecture that is based upon two novel primitive operations, called merging and reduction. These operators help refine a configuration, treating indexes and materialized views in a unified way, as well as succinctly explain the refinement process to DBAs.

• Type: Conference
• Year: 2006
• Where: EDBT
• Authors: Nicolas Bruno, Surajit Chaudhuri
• Keywords: Database, EDBT 2006, Ideal Physical Design, Physical Design Refinement, Physical Refinement Problem