Analytics Are Heavy. The DBMS Is Busy. When Will My Mission-Critical Transaction Start Running?
Summary: PreemptDB shows preemptive scheduling is practical in modern memory-optimized DBMSs by combining optimistic CC with new x86 userspace interrupts for low-overhead, user-interrupt-driven context switches. Demo: prioritized scheduling cuts latency of short, mission-critical transactions while preserving throughput and fairness versus non‑preemptive systems. (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Jiatang Zhou (Simon Fraser University)
- 2. Kaisong Huang (Simon Fraser University)
- 3. Zhuoyue Zhao (State University of New York at Buffalo)
- 4. Dong Xie (Pennsylvania State University)
- 5. Tianzhen Wang (Simon Fraser University)
BibTeX Citation
@article{zhou_vldb25,
title = {{Analytics Are Heavy. The DBMS Is Busy. When Will My Mission-Critical Transaction Start Running?}},
author = {Zhou, Jiatang and Huang, Kaisong and Zhao, Zhuoyue and Xie, Dong and Wang, Tianzhen},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {12},
pages = {5299--5302},
doi = {10.14778/3750601.3750656},
url = {https://doi.org/10.14778/3750601.3750656},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 1 of 1 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,857 | Interoperable ACID Transactions for Open Table Formats | 2026 | VLDB | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 5 of 5 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 2,850 | Priority in DBMS Resource Scheduling | 1989 | VLDB | 7.9421331e-05 |
| 2,925 | Interleaving with Coroutines: A Practical Approach for Robust Index Joins | 2018 | VLDB | 7.8482706e-05 |
| 3,777 | HTAP Databases: What is New and What is Next | 2022 | SIGMOD | 7.0268163e-05 |
| 4,243 | The Art of Latency Hiding in Modern Database Engines | 2024 | VLDB | 6.7069789e-05 |
| 9,008 | Low-Latency Transaction Scheduling via Userspace Interrupts: Why Wait or Yield When You Can Preempt? | 2025 | SIGMOD | 5.2387493e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 7,003 | Towards Optimal Transaction Scheduling | 2024 | VLDB |
| 2 | 3,519 | Towards Predicting Query Execution Time for Concurrent and Dynamic Database Workloads | 2013 | VLDB |
| 3 | 6,870 | An Almost-Serial Protocol for Transaction Execution in Main-Memory Database Systems | 2002 | VLDB |
| 4 | 5,535 | Real-Time Transaction Scheduling: A Cost Conscious Approach | 1993 | SIGMOD |
| 5 | 9,465 | Executing Long-Running Transactions in Synchronization-Free Main Memory Database Systems | 2013 | CIDR |
| 6 | 4,388 | Self-Tuning Query Scheduling for Analytical Workloads | 2021 | SIGMOD |
| 7 | 110 | DBMSs On A Modern Processor: Where Does Time Go? | 1999 | VLDB |
| 8 | 10,148 | Improving DBMS Scheduling Decisions with Accurate Performance Prediction on Concurrent Queries | 2025 | VLDB |
| 9 | 2,850 | Priority in DBMS Resource Scheduling | 1989 | VLDB |
| 10 | 9,008 | Low-Latency Transaction Scheduling via Userspace Interrupts: Why Wait or Yield When You Can Preempt? | 2025 | SIGMOD |