{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,8]],"date-time":"2026-04-08T09:01:13Z","timestamp":1775638873918,"version":"3.50.1"},"reference-count":38,"publisher":"Association for Computing Machinery (ACM)","issue":"13","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2021,9]]},"abstract":"<jats:p>UDO is a versatile tool for offline tuning of database systems for specific workloads. UDO can consider a variety of tuning choices, reaching from picking transaction code variants over index selections up to database system parameter tuning. UDO uses reinforcement learning to converge to near-optimal configurations, creating and evaluating different configurations via actual query executions (instead of relying on simplifying cost models). To cater to different parameter types, UDO distinguishes heavy parameters (which are expensive to change, e.g. physical design parameters) from light parameters. Specifically for optimizing heavy parameters, UDO uses reinforcement learning algorithms that allow delaying the point at which the reward feedback becomes available. This gives us the freedom to optimize the point in time and the order in which different configurations are created and evaluated (by benchmarking a workload sample). UDO uses a cost-based planner to minimize reconfiguration overheads. For instance, it aims to amortize the creation of expensive data structures by consecutively evaluating configurations using them. We evaluate UDO on Postgres as well as MySQL and on TPC-H as well as TPC-C, optimizing a variety of light and heavy parameters concurrently.<\/jats:p>","DOI":"10.14778\/3484224.3484236","type":"journal-article","created":{"date-parts":[[2021,10,28]],"date-time":"2021-10-28T22:36:50Z","timestamp":1635460610000},"page":"3402-3414","source":"Crossref","is-referenced-by-count":41,"title":["UDO"],"prefix":"10.14778","volume":"14","author":[{"given":"Junxiong","family":"Wang","sequence":"first","affiliation":[{"name":"Cornell University"}]},{"given":"Immanuel","family":"Trummer","sequence":"additional","affiliation":[{"name":"Cornell University"}]},{"given":"Debabrota","family":"Basu","sequence":"additional","affiliation":[{"name":"Inria Lille-Nord Europe, Lille, France"}]}],"member":"320","published-online":{"date-parts":[[2021,10,28]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"2020. https:\/\/github.com\/jfcoz\/postgresqltuner.  2020. https:\/\/github.com\/jfcoz\/postgresqltuner."},{"key":"e_1_2_1_2_1","unstructured":"2020. https:\/\/github.com\/keras-rl\/keras-rl.  2020. https:\/\/github.com\/keras-rl\/keras-rl."},{"key":"e_1_2_1_3_1","unstructured":"2020. https:\/\/github.com\/major\/MySQLTuner-perl.  2020. https:\/\/github.com\/major\/MySQLTuner-perl."},{"key":"e_1_2_1_4_1","unstructured":"2021. https:\/\/github.com\/ankane\/dexter.  2021. https:\/\/github.com\/ankane\/dexter."},{"key":"e_1_2_1_5_1","unstructured":"2021. https:\/\/www.eversql.com\/.  2021. https:\/\/www.eversql.com\/."},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-75225-7_15"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1013689704352"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/2304510.2304522"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.5555\/1953048.2021053"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2004.75"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.14778\/1687627.1687739"},{"key":"e_1_2_1_12_1","unstructured":"CMU Database Group. 2020. https:\/\/github.com\/cmu-db\/ottertune.  CMU Database Group. 2020. https:\/\/github.com\/cmu-db\/ottertune."},{"key":"e_1_2_1_13_1","volume-title":"March","author":"Coquelin Pierre-Arnaud","year":"2007","unstructured":"Pierre-Arnaud Coquelin and R\u00e9mi Munos . 2007. Bandit Algorithms for Tree Search. Arxiv preprint cs0703062 23 , March ( 2007 ), 67--74. arXiv:0703062v1 [arXiv:cs] http:\/\/arxiv.org\/abs\/cs\/0703062 Pierre-Arnaud Coquelin and R\u00e9mi Munos. 2007. Bandit Algorithms for Tree Search. Arxiv preprint cs0703062 23, March (2007), 67--74. arXiv:0703062v1 [arXiv:cs] http:\/\/arxiv.org\/abs\/cs\/0703062"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3324957"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/1273496.1273531"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.14778\/2850583.2850594"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1511\/2009.77.126"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3329859.3329876"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389704"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.5555\/3042817.3043099"},{"key":"e_1_2_1_21_1","unstructured":"Andreas Kipf Thomas Kipf Bernhard Radke Viktor Leis Peter Boncz and Alfons Kemper. 2018. Learned cardinalities: estimating correlated joins with deep learning. In CIDR. arXiv:1809.00677 http:\/\/arxiv.org\/abs\/1809.00677  Andreas Kipf Thomas Kipf Bernhard Radke Viktor Leis Peter Boncz and Alfons Kemper. 2018. Learned cardinalities: estimating correlated joins with deep learning. In CIDR . arXiv:1809.00677 http:\/\/arxiv.org\/abs\/1809.00677"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1007\/11871842_29"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.14778\/3352063.3352129"},{"key":"e_1_2_1_24_1","volume-title":"4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings","author":"Lillicrap Timothy P.","year":"2016","unstructured":"Timothy P. Lillicrap , Jonathan J. Hunt , Alexander Pritzel , Nicolas Heess , Tom Erez , Yuval Tassa , David Silver , and Daan Wierstra . 2016 . Continuous control with deep reinforcement learning . 4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings (2016). arXiv:1509.02971 Timothy P. Lillicrap, Jonathan J. Hunt, Alexander Pritzel, Nicolas Heess, Tom Erez, Yuval Tassa, David Silver, and Daan Wierstra. 2016. Continuous control with deep reinforcement learning. 4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings (2016). arXiv:1509.02971"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389768"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389727"},{"key":"e_1_2_1_27_1","volume-title":"On-line Q-learning using connectionist systems","author":"Rummery Gavin A","unstructured":"Gavin A Rummery and Mahesan Niranjan . 1994. On-line Q-learning using connectionist systems . Vol. 37 . University of Cambridge , Department of Engineering Cambridge, UK. Gavin A Rummery and Mahesan Niranjan. 1994. On-line Q-learning using connectionist systems. Vol. 37. University of Cambridge, Department of Engineering Cambridge, UK."},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1145\/3410566.3410603"},{"key":"e_1_2_1_29_1","volume-title":"Felix Martin Schuhknecht, and Jens Dittrich","author":"Sharma Ankur","year":"2018","unstructured":"Ankur Sharma , Felix Martin Schuhknecht, and Jens Dittrich . 2018 . The case for automatic database administration using deep reinforcement learning. arXiv (2018), 1--9. arXiv:1801.05643 Ankur Sharma, Felix Martin Schuhknecht, and Jens Dittrich. 2018. The case for automatic database administration using deep reinforcement learning. arXiv (2018), 1--9. arXiv:1801.05643"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3300087"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3300088"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.14778\/3450980.3450992"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448016.3452754"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3329859.3329875"},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.14778\/2876473.2876479"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389770"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.14778\/3229863.3236222"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3300085"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3484224.3484236","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,28]],"date-time":"2022-12-28T09:40:16Z","timestamp":1672220416000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3484224.3484236"}},"subtitle":["universal database optimization using reinforcement learning"],"short-title":[],"issued":{"date-parts":[[2021,9]]},"references-count":38,"journal-issue":{"issue":"13","published-print":{"date-parts":[[2021,9]]}},"alternative-id":["10.14778\/3484224.3484236"],"URL":"https:\/\/doi.org\/10.14778\/3484224.3484236","relation":{},"ISSN":["2150-8097"],"issn-type":[{"value":"2150-8097","type":"print"}],"subject":[],"published":{"date-parts":[[2021,9]]}}}