Bespoke OLAP: Synthesizing Workload-Specific One-size-fits-one Database Engines

Johannes Wehrstein, Timo Eckmann, Matthias Jasny, Carsten Binnig · 2026

Read Bespoke OLAP: Synthesizing Workload-Specific One-size-fits-one Database Engines by Johannes Wehrstein, Timo Eckmann, Matthias Jasny, Carsten Binnig (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-02001.txt.

Description

Johannes Wehrstein, Timo Eckmann, Matthias Jasny, and Carsten Binnig synthesize a database engine fitted to one fixed workload. Their 2026 work argues that general analytical engines pay for their flexibility in overhead. Runtime schema reading, indirection layers, and abstraction walls persist even in highly tuned systems.

A bespoke engine can drop those costs and use data structures shaped to the actual queries. Hand-building one was never worth the effort, but language-model code synthesis changes that math. Naive prompting fails, so their pipeline adds performance feedback, structured refinement, and validation loops. The autonomous system builds an engine from storage to query execution in minutes to hours. Order-of-magnitude speedups over a modern general engine show what workload-specific synthesis can win.

Catalog record

CREATORJohannes Wehrstein, Timo Eckmann, Matthias Jasny, Carsten Binnig
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE73610
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.02001

Browse the full library · About the Tower of Babel