Bullion: A Column Store for Machine Learning
Summary: Bullion is a columnar store tailored for ML: modular cascading encodings, in‑storage feature quantization, optimized long-sequence sparse‑feature encodings, and quality‑aware sequential reads for wide-table/multimodal training. Yields lower I/O for deletion compliance, substantial storage savings on sparse features, and faster metadata parsing than existing column stores. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Gang Liao (University of Maryland)
- 2. Ye Liu (ByteDance)
- 3. Jianjun Chen (ByteDance)
- 4. Daniel J. Abadi (University of Maryland)
BibTeX Citation
@inproceedings{liao_cidr25,
address = {Amsterdam, Netherlands},
series = {{CIDR} '25},
title = {{Bullion: A Column Store for Machine Learning}},
booktitle = {Proceedings of the {Conference} on {Innovative} {Data} {Systems} {Research}},
author = {Liao, Gang and Liu, Ye and Chen, Jianjun and Abadi, Daniel J.},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 4 of 4 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 9,038 | The FastLanes File Format | 2025 | VLDB | 5.3270943e-05 |
| 9,277 | F3: The Open-Source Data File Format for the Future | 2026 | SIGMOD | 5.2937431e-05 |
| 9,836 | Towards Functional Decomposition of Storage Formats | 2025 | CIDR | 5.2110542e-05 |
| 11,006 | Magnus: A Holistic Approach to Data Management for Large-Scale Machine Learning Workloads | 2025 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 27 of 27 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 | 11,006 | Magnus: A Holistic Approach to Data Management for Large-Scale Machine Learning Workloads | 2025 | VLDB |
| 2 | 2,023 | Decomposed Bounded Floats for Fast Compression and Queries | 2021 | VLDB |
| 3 | 1,395 | Column-oriented Database Systems | 2009 | VLDB |
| 4 | 10,509 | FlatStor: An Efficient Embedded-Index Based Columnar Data Layout for Multimodal Data Workloads | 2026 | VLDB |
| 5 | 9,385 | The Hopsworks Feature Store for Machine Learning | 2024 | SIGMOD |
| 6 | 4,592 | Data Platform for Machine Learning | 2019 | SIGMOD |
| 7 | 6,047 | Column Stores For Wide and Sparse Data | 2007 | CIDR |
| 8 | 9,827 | Frequency-Store: Scaling Image AI by A Column-Store for Images | 2025 | CIDR |
| 9 | 3,877 | An Empirical Evaluation of Columnar Storage Formats | 2024 | VLDB |
| 10 | 6,143 | ColumnML: Column-Store Machine Learning with On-The-Fly Data Transformation | 2019 | VLDB |