Print Email Facebook Twitter EAVS: Edge-assisted Adaptive Video Streaming with Fine-grained Serverless Pipelines Title EAVS: Edge-assisted Adaptive Video Streaming with Fine-grained Serverless Pipelines Author Hou, Biao (Beijing Institute of Technology) Yang, Song (Beijing Institute of Technology) Kuipers, F.A. (TU Delft Networked Systems) Jiao, Lei (University of Oregon) Fu, Xiaoming (University of Göttingen) Date 2023 Abstract Recent years have witnessed video streaming grad- ually evolve into one of the most popular Internet applications. With the rapidly growing personalized demand for real-time video streaming services, maximizing their Quality of Experience (QoE) is a long-standing challenge. The emergence of the server- less computing paradigm has potential to meet this challenge through its fine-grained management and highly parallel comput- ing structures. However, it is still ambiguous how to implement and configure serverless components to optimize video streaming services. In this paper, we propose EAVS, an Edge-assisted Adaptive Video streaming system with Serverless pipelines, which facilitates fine-grained management for multiple concurrent video transmission pipelines. Then, we design a chunk-level optimiza- tion scheme to address video bitrate adaptation. We propose a Deep Reinforcement Learning (DRL) algorithm based on Proximal Policy Optimization (PPO) with a trinal-clip mechanism to make bitrate decisions efficiently for better QoE. Finally, we implement the serverless video streaming system prototype and evaluate the performance of EAVS on various real-world network traces. Our results show that EAVS significantly improves QoE and reduces the video stall rate, achieving over 9.1% QoE improvement and 60.2% latency reduction compared to state- of-the-art solutions. Subject Video streamingServerless computingDeep reinforcement learningQuality of Experience To reference this document use: http://resolver.tudelft.nl/uuid:75c07e94-6cc7-46a6-b68d-ad3c8edb921e DOI https://doi.org/10.1109/INFOCOM53939.2023.10229102 Publisher IEEE, Danvers Embargo date 2024-02-29 ISBN 979-8-3503-3415-9 Source Proceedings of the INFOCOM 2023 - IEEE International Conference on Computer Communications Event IEEE INFOCOM 2023 - IEEE Conference on Computer Communications, 2023-05-17 → 2023-05-20, New York City, United States Bibliographical note Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. Part of collection Institutional Repository Document type conference paper Rights © 2023 Biao Hou, Song Yang, F.A. Kuipers, Lei Jiao, Xiaoming Fu Files PDF EAVS_Edge_assisted_Adapti ... elines.pdf 2.42 MB Close viewer /islandora/object/uuid:75c07e94-6cc7-46a6-b68d-ad3c8edb921e/datastream/OBJ/view