When Database Meets New Storage Devices: Understanding and Exposing Performance Mismatches via Configurations
Summary: A controlled study across six DBMSs exposes widespread NVMe performance mismatches, tracing them to storage-unaware I/O size, parallelism, and sequentiality. A configuration-based detector finds mismatches 100× faster than exhaustive testing, validating 15 of 17 reports. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Haochen He (National University of Defense Technology)
- 2. Erci Xu (National University of Defense Technology)
- 3. Shanshan Li (National University of Defense Technology)
- 4. Zhouyang Jia (National University of Defense Technology)
- 5. Si Zheng (National University of Defense Technology)
- 6. Yue Yu (National University of Defense Technology)
- 7. Jun Ma (National University of Defense Technology)
- 8. Xiangke Liao (National University of Defense Technology)
BibTeX Citation
@article{he_vldb23,
title = {{When Database Meets New Storage Devices: Understanding and Exposing Performance Mismatches via Configurations}},
author = {He, Haochen and Xu, Erci and Li, Shanshan and Jia, Zhouyang and Zheng, Si and Yu, Yue and Ma, Jun and Liao, Xiangke},
journal = {PVLDB},
series = {{VLDB} '23},
volume = {16},
number = {7},
pages = {1712--1725},
doi = {10.14778/3587136.3587145},
url = {https://doi.org/10.14778/3587136.3587145},
year = {2023}
}
Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,543 | How to Write to SSDs | 2026 | VLDB | 5.093636e-05 |
| 10,770 | Moving on From Group Commit: Autonomous Commit Enables High Throughput and Low Latency on NVMe SSDs | 2025 | SIGMOD | 5.093636e-05 |
| 10,962 | Environmental Footprints of Query Processing: A Vision for Sustainable Database Architectures | 2025 | VLDB | 5.093636e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 8 of 8 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 86 | Automatic Database Management System Tuning Through Large-scale Machine Learning | 2017 | SIGMOD | 0.00035316107 |
| 334 | An End-to-End Automatic Cloud Database Tuning System Using Deep Reinforcement Learning | 2019 | SIGMOD | 0.00020875082 |
| 512 | Tree Indexing on Solid State Drives | 2010 | VLDB | 0.00017196795 |
| 1,822 | PALM: Parallel Architecture-Friendly Latch-Free Modifications to B+ Trees on Many-Core Processors | 2011 | VLDB | 9.6765747e-05 |
| 3,400 | A Demonstration of the OtterTune Automatic Database Management System Tuning Service | 2018 | VLDB | 7.4433294e-05 |
| 3,747 | B+-tree Index Optimization by Exploiting Internal Parallelism of Flash-based Solid State Drives | 2012 | VLDB | 7.1555585e-05 |
| 4,894 | Optimizing Databases by Learning Hidden Parameters of Solid State Drives | 2020 | VLDB | 6.4562316e-05 |
| 8,800 | Parallel I/O Aware Query Optimization | 2014 | SIGMOD | 5.3691168e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 9,409 | NV-SQL: Boosting OLTP Performance with Non-Volatile DIMMs | 2023 | VLDB |
| 2 | 10,142 | Flexible I/O for Database Management Systems with xNVMe | 2026 | CIDR |
| 3 | 5,841 | The Necessary Death of the Block Device Interface | 2013 | CIDR |
| 4 | 4,894 | Optimizing Databases by Learning Hidden Parameters of Solid State Drives | 2020 | VLDB |
| 5 | 6,000 | Matching Database Access Patterns to Storage Characteristics | 2003 | VLDB |
| 6 | 2,828 | Exploiting Directly-Attached NVMe Arrays in DBMS | 2020 | CIDR |
| 7 | 10,974 | SSD-iq: Uncovering the Hidden Side of SSD Performance | 2025 | VLDB |
| 8 | 10,770 | Moving on From Group Commit: Autonomous Commit Enables High Throughput and Low Latency on NVMe SSDs | 2025 | SIGMOD |
| 9 | 8,262 | It Takes Two: Instrumenting the Interaction between In-Memory Databases and Solid-State Drives | 2020 | CIDR |
| 10 | 3,001 | What Modern NVMe Storage Can Do, And How To Exploit It: High-Performance I/O for High-Performance Storage Engines | 2023 | VLDB |