Cost-Model Oblivious Database Tuning with Reinforcement Learning

Debabrota Basu 1 Qian Lin 1 Weidong Chen 1, 2, 3 Hoang Tam Vo 1 Zihong Yuan 1 Pierre Senellart 4, 3 Stéphane Bressan 1
2 RMS - Réseaux, Mobilité et Services
LTCI - Laboratoire Traitement et Communication de l'Information
4 DIG - Data, Intelligence and Graphs
LTCI - Laboratoire Traitement et Communication de l'Information
Abstract :

In this paper, we propose a learning approach to adaptive performance tuning of database applications. The objective is to validate the opportunity to devise a tuning strategy that does not need prior knowledge of a cost model. Instead, the cost model is learned through reinforcement learning. We instantiate our approach to the use case of index tuning. We model the execution of queries and updates as a Markov decision process whose states are database configurations, actions are configuration changes, and rewards are functions of the cost of configuration change and query and update evaluation. During the reinforcement learning process, we face two important challenges: not only the unavailability of a cost model, but also the size of the state space. To address the latter, we devise strategies to prune the state space, both in the general case and for the use case of index tuning. We empirically and comparatively evaluate our approach on a standard OLTP dataset. We show that our approach is competitive with state-of-the-art adaptive index tuning, which is dependent on a cost model.

Document type :
Conference papers
Complete list of metadatas

https://hal-imt.archives-ouvertes.fr/hal-01261945
Contributor : Admin Télécom Paristech <>
Submitted on : Tuesday, January 26, 2016 - 8:47:43 AM
Last modification on : Thursday, October 17, 2019 - 12:37:01 PM

Identifiers

  • HAL Id : hal-01261945, version 1

Collections

Citation

Debabrota Basu, Qian Lin, Weidong Chen, Hoang Tam Vo, Zihong Yuan, et al.. Cost-Model Oblivious Database Tuning with Reinforcement Learning. DEXA, Sep 2015, Valencia, Spain. pp.253-268. ⟨hal-01261945⟩

Share

Metrics

Record views

278