Garen: Reliable Cluster Management with Atomic State Reconciliation
Mingi Kim, Ahnjae Shin, Jaewoo Maeng, Myeongjae Jeon, Byung-Gon Chun
Abstract
Modern cluster managers orchestrate large-scale services and resources through a set of controllers, each managing a specific part of the cluster by iteratively reconciling the cluster states into the desired states. However, controllers are prone to various state inconsistencies stemming from asynchrony, concurrency, and failures, posing significant challenges in reliable cluster operation. Our analysis of 51 consistency bugs in Kubernetes controllers reveals that many issues remain unresolved, and even proposed fixes are often rejected for backward incompatibility or performance loss.
Ask about this paper
Ask your agent about it.
Lune has read the top-tier papers around this one, so every answer names the papers it rests on.
Your agent calls
Lunesearch_papers
Free to start. No credit card required.
Terminal
Install the CLIlune papers get 0cd7ddd7-3c36-4e7a-ae74-a2d7eb9b9a95Related papers
- Automatic Reliability Testing For Cluster Management ControllersXudong Sun, Wenqing Luo, Jiawei Tyler Gu, Aishwarya Ganesan et al.OSDI 2022 · 44 citations
- An Empirical Study on Kubernetes Operator BugsQingxin Xu, Yu Gao, Jun WeiISSTA 2024 · 7 citations
- Who Watches the Watchers? On the Reliability of Softwarizing Cloud Application ManagementJiawei Tyler Gu, Zhen Tang, Yiming Su, Bogdan Alexandru Stoica et al.NSDI 2026 · 3 citations
- Anvil: Verifying Liveness of Cluster Management ControllersXudong Sun, Wenjie Ma, Jiawei Tyler Gu, Zicheng Ma et al.OSDI 2024 · 50 citations
- ZENITH: Towards A Formally Verified Highly-Available Control PlanePooria Namyar, Arvin Ghavidel, Mingyang Zhang, Harsha V. Madhyastha et al.SIGCOMM 2025 · 1 citation
