Back to papers
Optimizing Data Pipelines for Machine Learning in Feature Stores
Summary: Introduces DB-style optimizations for feature stores targeting point-in-time joins to reduce resource use and speed up ML data pipelines. Implemented in Feathr and evaluated on TPCx-AI and real retail workloads, achieving up to 3× pipeline acceleration.
(summarized by gpt-5-mini on Feb 09 2026)
- Paper ID
- 13293
- Venue
- VLDB
- Year
- 2023
- Pagerank
- 5.4253204e-05
- Overall Rank
- 5,575 | 61.26%
- DOI
-
10.14778/3625054.3625060
Incoming Non-self Citations Over Time
Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
Outgoing Citations (Sorted by Pagerank)
Showing 21 of 21 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank |
Cited Paper |
Year |
Venue |
Pagerank |
| 97 |
Maintaining Views Incrementally |
1993 |
SIGMOD |
0.00050863077 |
| 160 |
Automated Selection of Materialized Views and Indexes for SQL Databases |
2000 |
VLDB |
0.00040053897 |
| 482 |
Incremental Maintenance of Views with Duplicates |
1995 |
SIGMOD |
0.00022145976 |
| 728 |
Optimizing Queries Using Materialized Views: A Practical, Scalable Solution |
2001 |
SIGMOD |
0.00017459654 |
| 758 |
Materialization Optimizations for Feature Selection Workloads |
2014 |
SIGMOD |
0.00017053915 |
| 1,059 |
Answering Complex SQL Queries Using Automatic Summary Tables |
2000 |
SIGMOD |
0.00014370009 |
| 1,155 |
A Scalable Algorithm for Answering Queries Using Views |
2000 |
VLDB |
0.00013606507 |
| 1,911 |
Algorithms for Materialized View Design in Data Warehousing Environment |
1997 |
VLDB |
0.00010117173 |
| 1,921 |
Selecting Subexpressions to Materialize at Datacenter Scale |
2018 |
VLDB |
0.00010085899 |
| 2,398 |
Physical Data Independence, Constraints, and Optimization with Universal Plans |
1999 |
VLDB |
8.8871067e-05 |
| 3,866 |
Cloudy with High Chance of DBMS: A 10-year Prediction for Enterprise-Grade ML |
2020 |
CIDR |
6.6795497e-05 |
| 4,171 |
Computation Reuse in Analytics Job Service at Microsoft |
2018 |
SIGMOD |
6.3800823e-05 |
| 4,969 |
Relative Error Streaming Quantiles |
2021 |
PODS |
5.790405e-05 |
| 5,615 |
TPCx-AI - An Industry Standard Benchmark for Artificial Intelligence and Machine Learning Systems |
2023 |
VLDB |
5.409e-05 |
| 5,636 |
KLL± Approximate Quantile Sketches over Dynamic Datasets |
2021 |
VLDB |
5.3985928e-05 |
| 6,225 |
Managing ML Pipelines: Feature Stores and the Coming Wave of Embedding Ecosystems |
2021 |
VLDB |
5.1425119e-05 |
| 6,242 |
Optimizing In-memory Database Engine for AI-powered On-line Decision Augmentation Using Persistent Memory |
2021 |
VLDB |
5.1351431e-05 |
| 6,464 |
Materialization and Reuse Optimizations for Production Data Science Pipelines |
2022 |
SIGMOD |
5.0471003e-05 |
| 8,515 |
UPLIFT: Parallelization Strategies for Feature Transformations in Machine Learning Workloads |
2022 |
VLDB |
4.4901466e-05 |
| 8,826 |
Delta: Scalable Data Dissemination under Capacity Constraints |
2014 |
VLDB |
4.4371065e-05 |
| 9,349 |
Hippo: Sharing Computations in Hyper-Parameter Optimization |
2022 |
VLDB |
4.3497707e-05 |
Semantically Similar Papers