No Provisioned Concurrency: Fast RDMA-codesigned Remote Fork for Serverless Computing
Xingda Wei, Fangming Lu, Tianxia Wang, Jinyu Gu, Yuhan Yang, Rong Chen, Haibo Chen
摘要
Serverless platforms essentially face a tradeoff between container startup time and provisioned concurrency (i.e., cached instances), which is further exaggerated by the frequent need for remote container initialization. This paper presents MITO-SIS, an operating system primitive that provides fast remote fork, which exploits a deep codesign of the OS kernel with RDMA. By leveraging the fast remote read capability of RDMA and partial state transfer across serverless containers, MITOSIS bridges the performance gap between local and remote container initialization. MITOSIS is the first to fork over 10,000 new containers from one instance across multiple machines within a second, while allowing the new containers to efficiently transfer the pre-materialized states of the forked one. We have implemented MITOSIS on Linux and integrated it with FN, a popular serverless platform. Under load spikes in real-world serverless workloads, MITOSIS reduces the function tail latency by 89% with orders of magnitude lower memory usage. For serverless workflow that requires state transfer, MITOSIS improves its execution time by 86%.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper31
- ServerlessLLM: Low-Latency Serverless Inference for Large Language ModelsYao Fu, Leyang Xue, Yeqi Huang, Andrei-Octavian Brabete 等OSDI 2024 · 被引用 125 次
- Characterizing Off-path SmartNIC for Accelerating Distributed SystemsXingda Wei, Rongxin Cheng, Yuhan Yang, Rong Chen 等OSDI 2023 · 被引用 68 次
- Serialization/Deserialization-free State Transfer in Serverless WorkflowsFangming Lu, Xingda Wei, Zhuobin Huang, Rong Chen 等EuroSys 2024 · 被引用 31 次
- Serverless Cold Starts and Where to Find ThemArtjom Joosen, Ahmed Hassan, Martin Asenov, Rajkarn Singh 等EuroSys 2025 · 被引用 30 次
- BlitzScale: Fast and Live Large Model Autoscaling with O(1) Host CachingDingyan Zhang, Haotian Wang, Yang Liu, Xingda Wei 等OSDI 2025 · 被引用 29 次
它引用的顶会 Paper31
- Serverless in the Wild: Characterizing and Optimizing the Serverless Workload at a Large Cloud ProviderMohammad Shahrad, Rodrigo Fonseca, Iñigo Goiri, Gohar Irfan Chaudhry 等USENIX ATC 2020 · 被引用 946 次
- Faasm: Lightweight Isolation for Efficient Stateful Serverless ComputingSimon Shillaker, Peter R. PietzuchUSENIX ATC 2020 · 被引用 382 次
- FIRM: An Intelligent Fine-grained Resource Management Framework for SLO-Oriented MicroservicesHaoran Qiu, Subho S. Banerjee, Saurabh Jha, Zbigniew T. Kalbarczyk 等OSDI 2020 · 被引用 350 次
- Autopilot: workload autoscaling at GoogleKrzysztof Rzadca, Pawel Findeisen, Jacek Swiderski, Przemyslaw Zych 等EuroSys 2020 · 被引用 299 次
- When Cloud Storage Meets RDMAYixiao Gao, Qiang Li, Lingbo Tang, Yongqing Xi 等NSDI 2021 · 被引用 228 次
相关 Paper
- TrEnv: Transparently Share Serverless Execution Environments Across Different Functions and NodesJialiang Huang, Mingxing Zhang, Teng Ma, Zheng Liu 等SOSP 2024 · 被引用 14 次
- Catalyzer: Sub-millisecond Startup for Serverless Computing with Initialization-less BootingDong Du, Tianyi Yu, Yubin Xia, Binyu Zang 等ASPLOS 2020 · 被引用 280 次
- MXFaaS: Resource Sharing in Serverless Environments for Parallelism and EfficiencyJovan Stojkovic, Tianyin Xu, Hubertus Franke, Josep TorrellasISCA 2023 · 被引用 39 次
- Memory deduplication for serverless computing with MedesDivyanshu Saxena, Tao Ji, Arjun Singhvi, Junaid Khalid 等EuroSys 2022 · 被引用 54 次
- KRCORE: A Microsecond-scale RDMA Control Plane for Elastic ComputingXingda Wei, Fangming Lu, Rong Chen, Haibo ChenUSENIX ATC 2022 · 被引用 15 次
