Extending Relational Query Processing with ML Inference
Summary: Raven integrates ML runtimes (e.g., ONNX Runtime) natively into SQL Server and uses a unified IR to enable cross-optimizations between model inference and relational operators. Tight DB/compiler/ML co-optimization yields up to 5.5x native gains and up to 24x with cross-optimizations, enabling enterprise-grade in‑RDBMS scoring. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Konstantinos Karanasos (Microsoft)
- 2. Matteo Interlandi (Microsoft)
- 3. Doris Xin (University of California Berkeley)
- 4. Fotis Psallidas (Microsoft)
- 5. Rathijit Sen (Microsoft)
- 6. Kwanghyun Park (Microsoft)
- 7. Ivan Popivanov (Microsoft)
- 8. Supun Nakandal (University of California San Diego)
- 9. Subru Krishnan (Microsoft)
- 10. Markus Weimer (Microsoft)
- 11. Yuan Yu (Microsoft)
- 12. Raghu Ramakrishnan (Microsoft)
- 13. Carlo Curino (Microsoft)
BibTeX Citation
@inproceedings{karanasos_cidr20,
address = {Amsterdam, Netherlands},
series = {{CIDR} '20},
title = {{Extending Relational Query Processing with ML Inference}},
booktitle = {Proceedings of the {Conference} on {Innovative} {Data} {Systems} {Research}},
author = {Karanasos, Konstantinos and Interlandi, Matteo and Xin, Doris and Psallidas, Fotis and Sen, Rathijit and Park, Kwanghyun and Popivanov, Ivan and Nakandal, Supun and Krishnan, Subru and Weimer, Markus and Yu, Yuan and Ramakrishnan, Raghu and Curino, Carlo},
year = {2020}
}Incoming Citations (Sorted by Pagerank)
Showing 28 of 28 citing papers.
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 9 of 9 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 106 | The MADlib Analytics Library or MAD Skills, the SQL | 2012 | VLDB | 0.00033539462 |
| 344 | On the Power of Magic | 1987 | PODS | 0.00020659405 |
| 415 | SystemML: Declarative Machine Learning on Spark | 2016 | VLDB | 0.0001888524 |
| 894 | Froid: Optimization of Imperative Programs in a Relational Database | 2018 | VLDB | 0.00013367658 |
| 2,239 | An Intermediate Representation for Optimizing Machine Learning Pipelines | 2019 | VLDB | 8.8875753e-05 |
| 2,370 | Declarative Recursive Computation on an RDBMS or, Why You Should Use a Database For Distributed Machine Learning | 2019 | VLDB | 8.6795925e-05 |
| 3,438 | A Demonstration of Willump: A Statistically-Aware End-to-end Optimizer for Machine Learning Inference | 2020 | VLDB | 7.415647e-05 |
| 3,614 | Cloudy with High Chance of DBMS: A 10-year Prediction for Enterprise-Grade ML | 2020 | CIDR | 7.2568185e-05 |
| 4,599 | Automatically Leveraging MapReduce Frameworks for Data-Intensive Applications | 2018 | SIGMOD | 6.6122875e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 4,643 | Learned Index Benefits: Machine Learning Based Index Performance Estimation | 2022 | VLDB |
| 2 | 9,962 | Structure-Aware Machine Learning over Multi-Relational Databases | 2021 | SIGMOD |
| 3 | 295 | Accelerating Machine Learning Inference with Probabilistic Predicates | 2018 | SIGMOD |
| 4 | 4,053 | Database-Agnostic Workload Management | 2019 | CIDR |
| 5 | 6,357 | A Unified Transferable Model for ML-Enhanced DBMS | 2022 | CIDR |
| 6 | 5,132 | Facilitating SQL Query Composition and Analysis | 2020 | SIGMOD |
| 7 | 10,139 | Leveraging Query Optimizers to Verify the Soundness of LLM-based Query Rewrites for Real-World Workloads, and More! | 2026 | CIDR |
| 8 | 5,456 | InferDB: In-Database Machine Learning Inference Using Indexes | 2024 | VLDB |
| 9 | 7,066 | Aero: Adaptive Query Processing of ML Queries | 2025 | SIGMOD |
| 10 | 2,865 | End-to-end Optimization of Machine Learning Prediction Queries | 2022 | SIGMOD |