The Key to Effective UDF Optimization: Before Inlining, First Perform Outlining
Summary: Challenges whole-UDF inlining, showing it can yield worse plans. PRISM first outlines optimizer-unhelpful fragments, then selectively inlines the rest, improving unnesting and data skipping while avoiding joins (1.29× DuckDB; 298.73× SQL Server). (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Samuel Arch (Carnegie Mellon University)
- 2. Yuchen Liu (Carnegie Mellon University)
- 3. Todd C. Mowry (Carnegie Mellon University)
- 4. Jignesh M. Patel (Carnegie Mellon University)
- 5. Andrew Pavlo (Carnegie Mellon University)
BibTeX Citation
@article{arch_vldb25,
title = {{The Key to Effective UDF Optimization: Before Inlining, First Perform Outlining}},
author = {Arch, Samuel and Liu, Yuchen and Mowry, Todd C. and Patel, Jignesh M. and Pavlo, Andrew},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {1},
pages = {1--13},
doi = {10.14778/3696435.3696436},
url = {https://doi.org/10.14778/3696435.3696436},
year = {2025}
}
Incoming Citations (Sorted by Pagerank)
Showing 2 of 2 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 9,917 | The UDFBench Benchmark for General-purpose UDF Queries | 2025 | VLDB | 5.1955087e-05 |
| 10,541 | Scalable GPU Acceleration of Scalar Functions in Analytical Databases: Compilation, Benchmarking, and Optimization | 2026 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 22 of 22 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 12,510 | Fast and Dynamic OLAP Exploration Using UDFs | 2009 | SIGMOD |
| 2 | 3,945 | Exploiting Correlations for Expensive Predicate Evaluation | 2015 | SIGMOD |
| 3 | 6,553 | Containerized Execution of UDFs: An Experimental Evaluation | 2022 | VLDB |
| 4 | 5,791 | Accelerating Python UDFs in Vectorized Query Execution | 2022 | CIDR |
| 5 | 6,328 | Functional-Style SQL UDFs With a Capital 'F' | 2020 | SIGMOD |
| 6 | 10,740 | UDFBench: A Tool for Benchmarking UDF Queries on SQL Engines | 2025 | SIGMOD |
| 7 | 9,917 | The UDFBench Benchmark for General-purpose UDF Queries | 2025 | VLDB |
| 8 | 8,096 | Efficient Execution of User-Defined Functions in SQL Queries | 2023 | VLDB |
| 9 | 4,180 | BlackMagic: Automatic Inlining of Scalar UDFs into SQL Queries with Froid | 2019 | VLDB |
| 10 | 6,015 | Dear User-Defined Functions, Inlining isn't working out so great for us. Let's try batching to make our relationship work. Sincerely, SQL | 2024 | CIDR |