Beyond Compression: A Comprehensive Evaluation of Lossless Floating-Point Compression
Summary: Comprehensive empirical evaluation of lossless floating-point compressors for columnar engines, measuring compression ratio, in-situ DB query performance, and ML queries (distance, k-NN/RAG); implemented in Rust and released as a library. Finds clear trade-offs: no single method dominates both space and query speed—some compressors yield higher compression but slower DB/ML query performance. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Kaisei Hishida (Keio University)
- 2. Chunwei Liu (Massachusetts Institute of Technology)
- 3. John Paparrizos (Aristotle University of Thessaloniki; Ohio State University)
- 4. Aaron J. Elmore (University of Chicago)
BibTeX Citation
@article{hishida_vldb25,
title = {{Beyond Compression: A Comprehensive Evaluation of Lossless Floating-Point Compression}},
author = {Hishida, Kaisei and Liu, Chunwei and Paparrizos, John and Elmore, Aaron J.},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {11},
pages = {4396--4409},
doi = {10.14778/3749646.3749701},
url = {https://doi.org/10.14778/3749646.3749701},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 5 of 5 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 7,096 | Time-Series Clustering: A Comprehensive Study of Data Mining, Machine Learning, and Deep Learning Methods | 2025 | VLDB | 5.5991152e-05 |
| 10,480 | HYDRA: A Multi-Level Hierarchy-Driven Approach for Robust Anomaly Detection in Time Series | 2026 | SIGMOD | 4.9769913e-05 |
| 10,494 | MUFASA: Fast and Accurate Multivariate Time-Series Clustering | 2026 | SIGMOD | 4.9769913e-05 |
| 10,521 | The Power of Anomaly Detection in Predictive Maintenance: [Experiments & Analysis] | 2026 | SIGMOD | 4.9769913e-05 |
| 10,858 | Error-bounded Point Cloud Compression Using Truncated Octahedron Quantization | 2026 | VLDB | 4.9769913e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 36 of 36 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 | 10,662 | Improving LZ4 for Effective Compression and Efficient Query | 2026 | SIGMOD |
| 2 | 2,061 | Similarity search in the blink of an eye with compressed indices | 2023 | VLDB |
| 3 | 898 | Query Optimization In Compressed Database Systems | 2001 | SIGMOD |
| 4 | 9,295 | Revisiting B-tree Compression: An Experimental Study | 2024 | SIGMOD |
| 5 | 4,015 | A New Compression Method with Fast Searching on Large Databases | 1987 | VLDB |
| 6 | 6,736 | CompressDB: Enabling Efficient Compressed Data Direct Processing for Various Databases | 2022 | SIGMOD |
| 7 | 9,705 | Experimental Analysis of Large-scale Learnable Vector Storage Compression | 2024 | VLDB |
| 8 | 1,974 | Decomposed Bounded Floats for Fast Compression and Queries | 2021 | VLDB |
| 9 | 8,040 | FCBench: Cross-Domain Benchmarking of Lossless Compression for Floating-Point Data | 2024 | VLDB |
| 10 | 12,078 | An Evaluation of Methods of Compressing Doubles | 2020 | SIGMOD |