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 project, we learn how to partition a distributed DBMS for OLAP-style workloads using Deep Reinforcement Learning (DRL). 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.
|Benjamin Hilprecht M.Sc.|
|S2|02 D110||+49 6151 firstname.lastname@example.org-...|
Hilprecht, Benjamin ; Binnig, Carsten ; Röhm, Uwe Bordawekar, Rajesh ; Shmueli, Oded (Hrsg.) (2019):
Towards learning a partitioning advisor with deep reinforcement learning.
In: Proceedings of the Second International Workshop on Exploiting Artificial Intelligence Techniques for Data Management, aiDM@SIGMOD 2019, S. 6:1-6:4,
Amsterdam, The Netherlands, ACM, Second International Workshop on Exploiting Artificial Intelligence Techniques for Data Management, aiDM@SIGMOD 2019, Amsterdam, The Netherlands, July 5, 2019, DOI: 10.1145/3329859.3329876,