Async-fork: Mitigating Query Latency Spikes Incurred by the Fork-based Snapshot Mechanism from the OS Level
Summary: Async-fork offloads fork-based snapshots from the query path by asynchronous OS-level forking, preserving IMKVS persistence. Linux kernel implementation, deployed on Redis in public clouds, yields substantially lower tail latency during snapshots. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Pu Pang (Alibaba; Shanghai Jiao Tong University)
- 2. Quan Chen (Shanghai Jiao Tong University)
- 3. Yu Xu (Alibaba)
- 4. Xiyu Wang (Alibaba)
- 5. Yong Yang (Alibaba)
- 6. Gang Deng (Alibaba)
- 7. Shixuan Sun (National University of Singapore)
- 8. Kaihao Bai (Alibaba; Shanghai Jiao Tong University)
- 9. Bo Liu (Shanghai Jiao Tong University)
- 10. Hongbo Yao (Alibaba)
- 11. Zhengheng Wang (Alibaba)
- 12. Zheng Liu (Alibaba)
- 13. Tao Ma (Alibaba)
- 14. Zhuo Song (Alibaba; Shanghai Jiao Tong University)
- 15. Minyi Guo (Shanghai Jiao Tong University)
BibTeX Citation
@article{pang_vldb23,
title = {{Async-fork: Mitigating Query Latency Spikes Incurred by the Fork-based Snapshot Mechanism from the OS Level}},
author = {Pang, Pu and Chen, Quan and Xu, Yu and Wang, Xiyu and Yang, Yong and Deng, Gang and Sun, Shixuan and Bai, Kaihao and Liu, Bo and Yao, Hongbo and Wang, Zhengheng and Liu, Zheng and Ma, Tao and Song, Zhuo and Guo, Minyi},
journal = {PVLDB},
series = {{VLDB} '23},
volume = {16},
number = {5},
pages = {1033--1045},
doi = {10.14778/3579075.3579079},
url = {https://doi.org/10.14778/3579075.3579079},
year = {2023}
}Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 9,345 | Practical DB-OS Co-Design with Privileged Kernel Bypass | 2025 | SIGMOD | 5.284891e-05 |
| 11,201 | Live Patching for Distributed In-Memory Key-Value Stores | 2024 | SIGMOD | 5.093636e-05 |
| 11,333 | The Case for DBMS Live Patching | 2024 | VLDB | 5.093636e-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 |
|---|---|---|---|---|
| 38 | Hekaton: SQL Server’s Memory-Optimized OLTP Engine | 2013 | SIGMOD | 0.00047648573 |
| 614 | Faster: A Concurrent Key-Value Store with In-Place Updates | 2018 | SIGMOD | 0.00015720975 |
| 1,255 | Fast Checkpoint Recovery Algorithms for Frequently Consistent Applications | 2011 | SIGMOD | 0.00011472627 |
| 4,675 | Accelerating Analytical Processing in MVCC using Fine-Granular High-Frequency Virtual Snapshotting | 2018 | SIGMOD | 6.5727393e-05 |
| 9,151 | Tair-PMem: A Fully Durable Non-Volatile Memory Database | 2022 | VLDB | 5.3133586e-05 |
Previous
Page 1 / 1
Next