MB2: Decomposed Behavior Modeling for Self-Driving Database Management Systems
Summary: MB2's ModelBot2 presents a decomposed, end-to-end ML framework for self-driving DBMSs, using fine-grained units to predict behavior for unseen configurations. It provides offline data generation and in-memory deployment, delivering up to 25x accuracy against state-of-the-art models for OLTP/OLAP in dynamic workloads. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Lin Ma (Carnegie Mellon University)
- 2. William Zhang (Carnegie Mellon University)
- 3. Jie Jiao (Carnegie Mellon University)
- 4. Wuwen Wang (Carnegie Mellon University)
- 5. Matthew Butrovich (Carnegie Mellon University)
- 6. Wan Shen Lim (Carnegie Mellon University)
- 7. Prashanth Menon (Carnegie Mellon University)
- 8. Andrew Pavlo (Carnegie Mellon University)
BibTeX Citation
@inproceedings{ma_sigmod21,
title = {{MB2: Decomposed Behavior Modeling for Self-Driving Database Management Systems}},
author = {Ma, Lin and Zhang, William and Jiao, Jie and Wang, Wuwen and Butrovich, Matthew and Lim, Wan Shen and Menon, Prashanth and Pavlo, Andrew},
series = {{SIGMOD} '21},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3448016.3457276},
url = {https://dl.acm.org/doi/10.1145/3448016.3457276},
year = {2021}
}Incoming Citations (Sorted by Pagerank)
Showing 22 of 22 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 33 of 33 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next