Starlight: Fast Container Provisioning on the Edge and over the WAN
Jun Lin Chen, Daniyal Liaqat, Moshe Gabel, Eyal de Lara
Abstract
Containers, originally designed for cloud environments, are increasingly popular for provisioning workers outside the cloud, for example in mobile and edge computing. These settings, however, bring new challenges: high latency links, limited bandwidth, and resource-constrained workers. The result is longer provisioning times when deploying new workers or updating existing ones, much of it due to network traffic.
Our analysis shows that current piecemeal approaches to reducing provisioning time are not always sufficient, and can even make things worse as round-trip times grow. Rather, we find that the very same layer-based structure that makes containers easy to develop and use also makes it more difficult to optimize deployment. Addressing this issue thus requires rethinking the container deployment pipeline as a whole.
Based on our findings, we present Starlight: an accelerator for container provisioning. Starlight decouples provisioning from development by redesigning the container deployment protocol, filesystem, and image storage format. Our evaluation using 21 popular containers shows that, on average, Starlight deploys and starts containers 3.0× faster than the current state-of-the-art implementation while incurring no runtime overhead and little (5%) storage overhead. Finally, it is backwards compatible with existing workers and uses standard container registries.
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.
Cited by top-tier papers6
- On-demand Container Loading in AWS LambdaMarc Brooker, Mike Danilov, Chris Greenwood, Phil PiwonkaUSENIX ATC 2023 · 73 citations
- Serverless Cold Starts and Where to Find ThemArtjom Joosen, Ahmed Hassan, Martin Asenov, Rajkarn Singh et al.EuroSys 2025 · 30 citations
- EdgeMove: Pipelining Device-Edge Model Training for Mobile IntelligenceZeqian Dong, Qiang He, Feifei Chen, Hai Jin et al.WWW 2023 · 12 citations
- BREAK: A Holistic Approach for Efficient Container Deployment among Edge CloudsYicheng Feng, Shihao Shen, Xiaofei Wang, Qiao Xiang et al.INFOCOM 2024 · 11 citations
- FlacIO: Flat and Collective I/O for Container Image ServiceYubo Liu, Hongbo Li, Mingrui Liu, Rui Jing et al.FAST 2025 · 7 citations
Builds on4
- Orbital Edge Computing: Nanosatellite Constellations as a New Class of Computer SystemBradley Denby, Brandon LuciaASPLOS 2020 · 272 citations
- FaaSNet: Scalable and Fast Provisioning of Custom Serverless Container Runtimes at Alibaba Cloud Function ComputeAo Wang, Shuai Chang, Huangshi Tian, Hongqi Wang et al.USENIX ATC 2021 · 171 citations
- DADI: Block-Level Image Service for Agile and Elastic Application DeploymentHuiba Li, Yifan Yuan, Rui Du, Kai Ma et al.USENIX ATC 2020 · 56 citations
- Surrounded by the Clouds: A Comprehensive Cloud Reachability StudyLorenzo Corneo, Maximilian Eder, Nitinder Mohan, Aleksandr Zavodovski et al.WWW 2021 · 54 citations
Related papers
- EDDE: Container Deployment Framework Beyond the CloudHao Fan, Zhuo Huang, Shadi Ibrahim, Lin Gu et al.SC 2025 · 1 citation
- ONCache: A Cache-Based Low-Overhead Container Overlay NetworkShengkai Lin, Shizhen Zhao, Peirui Cao, Xinchi Han et al.NSDI 2025
- Parallelizing packet processing in container overlay networksJiaxin Lei, Manish Munikar, Kun Suo, Hui Lu et al.EuroSys 2021 · 22 citations
- LOPO: An Out-of-order Layer Pulling Orchestration Strategy for Fast Microservice StartupLin Gu, Junhao Huang, Shaoxing Huang, Deze Zeng et al.INFOCOM 2023 · 8 citations
- Poby: SmartNIC-accelerated Image Provisioning for Coldstart in CloudsZihao Chang, Jiaqi Zhu, Haifeng Sun, Yunlong Xie et al.USENIX ATC 2025 · 5 citations
