SKQ: Event Scheduling for Optimizing Tail Latency in a Traditional OS Kernel
Siyao Zhao, Haoyu Gu, Ali José Mashtizadeh
Abstract
This paper presents Schedulable Kqueue (SKQ), a new design to FreeBSD Kqueue that improves application tail latency and low-latency throughput. SKQ introduces a new scalable architecture and event scheduling. We provide multiple scheduling policies that improve cache locality and reduce workload imbalance. SKQ also enables applications to prioritize processing latency-sensitive requests over regular requests.
In the RocksDB benchmark, SKQ reduces tail latency by up to 1022× and extends the low-latency throughput by 27.4×. SKQ also closes the gap between traditional OS kernel networking and a state-of-the-art kernel-bypass networking system by 83.7% for an imbalanced workload.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext e86f6856-090f-406e-aaa6-baa5275a8b87Cited by top-tier papers6
- Efficient Scheduling Policies for Microsecond-Scale TasksSarah McClure, Amy Ousterhout, Scott Shenker, Sylvia RatnasamyNSDI 2022 · 43 citations
- LibPreemptible: Enabling Fast, Adaptive, and Hardware-Assisted User-Space SchedulingYueying Li, Nikita Lazarev, David Koufaty, Tenny Yin et al.HPCA 2024 · 14 citations
- HA/TCP: A Reliable and Scalable Framework for TCP Network FunctionsHaoyu Gu, Ali José Mashtizadeh, Bernard WongNSDI 2025 · 1 citation
- Rakaia: Scalable In-Kernel Scheduling for TCP-Based RPCsRui Yang, Konstantinos Prasopoulos, Edouard BugnionOSDI 2026
- SBB: Eliminating Centralized Bottlenecks in Userspace Network RuntimeKang Hu, Shuqi Dong, Chuandong Li, Ran Yi et al.OSDI 2026
Builds on1
Related papers
- Rearchitecting Linux Storage Stack for µs Latency and High ThroughputJaehyun Hwang, Midhul Vuppalapati, Simon Peter, Rachit AgarwalOSDI 2021 · 63 citations
- Towards Optimal Rack-scale μs-level CPU Scheduling through In-Network Workload ShapingXudong Liao, Han Tian, Xinchen Wan, Chaoliang Zeng et al.USENIX ATC 2025 · 1 citation
- Understanding Host Network Stack LatencyTianyu Zuo, Jaehyun Hwang, Ao Tang, Rachit Agarwal et al.SIGCOMM 2026
- When Idling is Ideal: Optimizing Tail-Latency for Heavy-Tailed Datacenter Workloads with PerséphoneHenri Maxime Demoulin, Joshua Fried, Isaac Pedisich, Marios Kogias et al.SOSP 2021 · 39 citations
- Q-Zilla: A Scheduling Framework and Core Microarchitecture for Tail-Tolerant MicroservicesAmirhossein Mirhosseini, Brendan L. West, Geoffrey W. Blake, Thomas F. WenischHPCA 2020 · 30 citations
