Autonomous query-driven index tuning

Abstract

Abstract

Adaptive Segmentation for Scientific Databases

“…Our approach aligns with and extends the recently proposed on-line tuning techniques [5], [6], [7], which focus on an adaptive solution for the index selection problem by monitoring the query load and changing the auxiliary access structures (indices and materialized views) on-the-fly. The adaptive segmentation reorganizes the data itself and partially shares the overhead with the query execution.…”

Adaptive Physical Design for Curated Archives

“…However, this approach requires further tuning by a DBA to ensure that the tool does not react too quickly or too slowly to changes in the workload and result in poor design choices. Recent research [6][7][8] on the online physical design problem focuses on one aspect of physical design: index design. Bruno and Chaudhari [8] infer costs and plan properties during the query optimization phase to efficiently decide between various index configurations in an online fashion.…”

Self-organizing strategies for a column-store database

“…On-line tools for physical design tuning have been a hot area of research in the last years [12,13,5]. They monitor the query load, keep statistics on the existing and virtual alternatives and change the auxiliary access structures (indices and materialized views).…”

“…In a row-store system reorganization typically affects the indices [12,13,5] and materialized views. Both are expensive to construct and their contribution strongly depends on the stability of the workload characteristics and database volatility.…”