Learning a Partitioning Advisor for Cloud Databases
Authors:
Open Access
- 31 May 2020
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 143-157
Abstract
Cloud vendors provide ready-to-use distributed DBMS solutions as a service. While the provisioning of a DBMS is usually fully automated, customers typically still have to make important design decisions which were traditionally made by the database administrator such as finding an optimal partitioning scheme for a given database schema and workload. In this paper, we introduce a new learned partitioning advisor based on Deep Reinforcement Learning (DRL) for OLAP-style workloads. The main idea is that a DRL agent learns the cost tradeoffs of different partitioning schemes and can thus automate the partitioning decision. In the evaluation, we show that our advisor is able to find non-trivial partitionings for a wide range of workloads and outperforms more classical approaches for automated partitioning design.Keywords
This publication has 21 references indexed in Scilit:
- Workload-driven adaptive data partitioning and distribution — The Cumulus approach2015
- Locality-aware Partitioning in Parallel Database Systems2015
- Human-level control through deep reinforcement learningNature, 2015
- SWORD2013
- Skew-aware automatic database partitioning in shared-nothing, parallel OLTP systems2012
- SchismProceedings of the VLDB Endowment, 2010
- Supporting table partitioning by reference in oracle2008
- Database tuning advisor for microsoft SQL server 20052005
- DB2 Design Advisor2004
- Automating physical database design in a parallel database2002