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.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Agrawal, S., Chaudhuri, S., Narasayya, V.R.: Automated selection of materialized views and indexes in SQL databases. In: VLDB (2000)
Agrawal, S., Narasayya, V., Yang, B.: Integrating vertical and horizontal partitioning into automated physical database design. In: SIGMOD (2004)
Azefack, S., Aouiche, K., Darmont, J.: Dynamic index selection in data warehouses. CoRR abs/0809.1965 (2008). http://arxiv.org/abs/0809.1965
Benedikt, M., Bohannon, P., Bruns, G.: Data cleaning for decision support. In: CleanDB (2006)
Borodin, A., El-Yaniv, R.: Online Computation and Competitive Analysis. Cambridge University Press, Cambridge (1998)
Bruno, N., Chaudhuri, S.: An online approach to physical design tuning. In: ICDE (2007)
Chaudhuri, S., Narasayya, V.: Autoadmin: What-if index analysis utility. In: SIGMOD (1998)
Difallah, D.E., Pavlo, A., Curino, C., Cudre-Mauroux, P.: OLTP-Bench: an extensible testbed for benchmarking relational databases. Proc. VLDB Endow. 7(4), 277–288 (2013)
Lagoudakis, M.G., Parr, R.: Least-squares policy iteration. J. Mach. Learn. Res. 4, 1107–1149 (2003)
Lohman, G.M.: Is query optimization a “solved” problem? (2014). http://wp.sigmod.org/?p=1075
Papadomanolakis, S., Dash, D., Ailamaki, A.: Efficient use of the query optimizer for automated physical design. In: VLDB (2007)
Powell, W.B.: Approximate Dynamic Programming: Solving the Curses of Dimensionality. Wiley-Interscience, New York (2007)
Puterman, M.L.: Markov Decision Processes: Discrete Stochastic Dynamic Programming. Wiley, New York (2009)
Raab, F.: TPC-C - the standard benchmark for online transaction processing (OLTP). In: Gray, J. (ed.) The Benchmark Handbook. Morgan Kaufmann, San Francisco (1993)
Ramakrishnan, R., Gehrke, J., Gehrke, J.: Database Management Systems, vol. 3. McGraw-Hill, New York (2003)
Rasin, A., Zdonik, S.: An automatic physical design tool for clustered column-stores. In: EDBT (2013)
Schnaitter, K., Abiteboul, S., Milo, T., Polyzotis, N.: On-line index selection for shifting workloads. In: SMDB (2007)
Schnaitter, K., Polyzotis, N.: Semi-automatic index tuning: keeping DBAs in the loop. Proc. VLDB Endow. 5(5), 478–489 (2012)
Stillger, M., Lohman, G.M., Markl, V., Kandil, M.: LEO - DB2’s LEarning Optimizer. In: VLDB (2001)
Sutton, R.S., Barto, A.G.: Reinforcement Learning. MIT Press, Cambridge (1998)
White, D.J.: Markov Decision Processes. Wiley, New York (1993)
Young, P.: Recursive least squares estimation. In: Young, P. (ed.) Recursive Estimation and Time-Series Analysis, pp. 29–46. Springer, Berlin Heidelberg (2011)
Acknowledgement
This research is funded by the National Research Foundation Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE) programme with the SP2 project of the Energy and Environmental Sustainability Solutions for Megacities - E2S2 programme.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Basu, D. et al. (2015). Cost-Model Oblivious Database Tuning with Reinforcement Learning. In: Chen, Q., Hameurlain, A., Toumani, F., Wagner, R., Decker, H. (eds) Database and Expert Systems Applications. Globe DEXA 2015 2015. Lecture Notes in Computer Science(), vol 9261. Springer, Cham. https://doi.org/10.1007/978-3-319-22849-5_18
Download citation
DOI: https://doi.org/10.1007/978-3-319-22849-5_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-22848-8
Online ISBN: 978-3-319-22849-5
eBook Packages: Computer ScienceComputer Science (R0)