Online index selection using deep reinforcement learning for a cluster database

Zahra Sadri, Le Gruenwald, Eleazar Leal

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Index selection plays a substantial role in database performance by reducing the I/O cost. Existing index advisors apply different heuristic methods to search the large search space of possible attributes for indexing. These heuristic approaches do not have a mechanism to learn about the goodness of the recommended index set. Thus, they might choose the same index set with a low impact on I/O cost reduction. Learning from their decisions can improve the quality of the recommended index set. We believe that Deep Reinforcement Learning (DRL) is a solution to tackle this issue. Using DRL, an index advisor can improve its decision using the feedbacks of its decisions. In this paper, we propose a DRL-index advisor for a cluster database. We describe the major components such as agent, environment, set of actions, the reward function, and other modules. We conclude the paper with open challenges and possible future work.

Original languageEnglish (US)
Title of host publicationProceedings - 2020 IEEE 36th International Conference on Data Engineering Workshops, ICDEW 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages158-161
Number of pages4
ISBN (Electronic)9781728142661
DOIs
StatePublished - Apr 2020
Event36th IEEE International Conference on Data Engineering Workshops, ICDEW 2020 - Dallas, United States
Duration: Apr 20 2020Apr 24 2020

Publication series

NameProceedings - 2020 IEEE 36th International Conference on Data Engineering Workshops, ICDEW 2020

Conference

Conference36th IEEE International Conference on Data Engineering Workshops, ICDEW 2020
CountryUnited States
CityDallas
Period4/20/204/24/20

Keywords

  • Cluster computer
  • Deep reinforcement learning
  • Index selection

Fingerprint Dive into the research topics of 'Online index selection using deep reinforcement learning for a cluster database'. Together they form a unique fingerprint.

  • Cite this

    Sadri, Z., Gruenwald, L., & Leal, E. (2020). Online index selection using deep reinforcement learning for a cluster database. In Proceedings - 2020 IEEE 36th International Conference on Data Engineering Workshops, ICDEW 2020 (pp. 158-161). [9094124] (Proceedings - 2020 IEEE 36th International Conference on Data Engineering Workshops, ICDEW 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDEW49219.2020.00035