Reproduce “CATune: Structural Constraint-Aware Bayesian Optimization for DBMS Configuration Tuning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Fangping Lan, Qi Zhang, Eduard Dragut
CATune constrains Bayesian optimization for database tuning by encoding verified orderings from DBMS manuals into the feasible search space, so proposed settings obey those rules. In PostgreSQL TPC-C with default ranges, it reached baseline SMAC’s best performance after 200 iterations 12.5 times faster in iteration count and achieved 33.39% higher throughput; this evidence is limited to tested settings and constraints.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.