Haitian Pang
ABSTRACT
Recent years have witnessed a rapid increase of crowdsourced live streaming (CLS): applications like Twitch.tv have attracted millions of daily active users. Content delivery in such crowdsourced live streaming involves two phases: 1) Video stream (i.e., a live channel) is generated and uploaded by a broadcaster user, and 2) The video stream is then delivered to many viewers choosing to watch the channel. Today's crowdsourced live streaming service usually employs conventional content delivery network (CDN) solutions to address the above content delivery problem, i.e., letting the broadcaster upload the video to a sinking CDN that then distributes the content to viewers. This solution causes a large delay and bandwidth insufficiency in the first mile between the broadcasters and the sinking CDN servers - our measurement study shows that the first-mile upload network quality causes a large portion of viewer rebuffers in the whole channel.
In this paper, we propose a content harvest network (CHN) solution to address the first-mile problem. In particular, the content harvest network employs relays at the edge of the network, to receive the content uploaded by broadcasters and then forward it to the CDN servers. Though the idea seems straightforward, it faces the following challenges: i) How to determine which channels need relay assistance? ii) How to choose the right relays to provide good first-mile QoS? iii) How to dynamically adjust the relay assignment in different channels?
In order to provide global optimal and real-time assignment, we use a hybrid solution, i.e., centralized assignment and distributed assignment. Specifically, we formulate global relay assignment as an optimization problem and develop an approximation algorithm using rounding technique. We use a multi-armed bandit (MAB) based method to perform the distributed assignment. Experiment results on a large-scale trace show that our solution can reduce the overall viewer latency by 40%, as compared to state-of-the-art solutions.
- Cisco visual networking index: Global mobile data traffic forecast update 20170209Google Scholar
- http://www.inke.tv/Google Scholar
- Zhang, C., Liu, J. On crowdsourced interactive live streaming: a twitch. tv-based measurement study, ACM Workshop on Network and Operating Systems Support for Digital Audio and Video, 2015. Google ScholarDigital Library
- Shi, W., Cao, J., Zhang, Q., et al. Edge computing: Vision and challenges, IEEE Internet of Things Journal, 2016.Google Scholar
- Chiang, M. Fog networking: An overview on research opportunities. arXiv preprint arXiv:1601.00835, 2016.Google Scholar
- Chen, L., Zhou, Y., Jing, M., et al. Thunder crystal: a novel crowdsourcing-based content distribution platform, ACM Workshop on Network and Operating Systems Support for Digital Audio and Video, 2015. Google ScholarDigital Library
- Wang, B., Zhang, X., Wang, G., et al. Anatomy of a personalized livestreaming system, ACM Internet Measurement Conference, 2016. Google ScholarDigital Library
- Siekkinen, M., Masala, E., Kämäräinen, T. A First Look at Quality of Mobile Live Streaming Experience: the Case of Periscope, ACM on Internet Measurement Conference, 2016. Google ScholarDigital Library
- Mukerjee, M. K., Naylor, D., Jiang, J., et al. Practical, real-time centralized control for cdn-based live video delivery, ACM SIGCOMM Computer Communication Review, 2015. Google ScholarDigital Library
- Chen, F., Zhang, C., Wang, F., et al. Cloud-Assisted Live Streaming for Crowdsourced Multimedia Content, IEEE Transactions on Multimedia, 2015.Google Scholar
- Chen, F., Zhang, C., Wang, F., et al. Crowdsourced live streaming over the cloud, IEEE Conference on Computer Communications (INFOCOM), 2015.Google ScholarCross Ref
- Wang, F., Liu, J., Chen, M., et al. Migration towards cloud-assisted live media streaming, IEEE/ACM Transactions on networking, 2016. Google ScholarDigital Library
- He, Q., Liu, J., Wang, C., et al. Coping with heterogeneous video contributors and viewers in crowdsourced live streaming: A cloud-based approach, IEEE Transactions on Multimedia, 2016. Google ScholarDigital Library
- Savage, S., Collins, A., Hoffman, E., et al. The end-to-end effects of Internet path selection, ACM SIGCOMM Computer Communication Review, 1999. Google ScholarDigital Library
- Jiang, J., Das, R., Ananthanarayanan, G., et al. Via: Improving internet telephony call quality using predictive relay selection, ACM SIGCOMM, 2016. Google ScholarDigital Library
- Zhang, X., Liu, J., Li, B., et al. CoolStreaming/DONet: A data-driven overlay network for peer-to-peer live media streaming, IEEE Conference on Computer Communications (INFOCOM), 2005.Google Scholar
- Qiu, T., Ge, Z., Lee, S., et al. Modeling channel popularity dynamics in a large IPTV system, ACM SIGMETRICS Performance Evaluation Review, 2009. Google ScholarDigital Library
- Madhyastha, H. V., Isdal, T., Piatek, M., et al. iPlane: An information plane for distributed services, Symposium on Operating systems design and implementation. USENIX Association, 2006. Google ScholarDigital Library
- Mok, R. K. P., Chan, E. W. W., Chang, R. K. C. Measuring the quality of experience of HTTP video streaming, IFIP/IEEE International Symposium on Integrated Network Management (IM), 2011.Google ScholarCross Ref
- Egilmez, H. E., Civanlar, S., Tekalp, A. M. An optimization framework for QoS-enabled adaptive video streaming over OpenFlow networks, IEEE Transactions on Multimedia, 2013. Google ScholarDigital Library
- He, Q., Zhang, C., Liu, J. Utilizing Massive Viewers for Video Transcoding in Crowdsourced Live Streaming, IEEE International Conference on Cloud Computing (CLOUD), 2016.Google ScholarCross Ref
- Wang, Z., Sun, L., Wu, C., et al. Joint online transcoding and geo-distributed delivery for dynamic adaptive streaming, IEEE INFOCOM, 2014.Google ScholarCross Ref
- Raghavan, P., Tompson, C. D. Randomized rounding: a technique for provably good algorithms and algorithmic proofs, Combinatorica, 1987. Google ScholarDigital Library
- Klee, V., Minty, G. J. How good is the simplex algorithm, WASHINGTON UNIV SEATTLE DEPT OF MATHEMATICS, 1970.Google Scholar
- Auer, P., Cesa-Bianchi, N., Fischer, P. Finite-time analysis of the multi-armed bandit problem, Machine learning, 2002. Google ScholarDigital Library
- https://en.wikipedia.org/wiki/Knapsack_problemGoogle Scholar
Index Terms
- First Mile in Crowdsourced Live Streaming: A Content Harvest Network Approach
Recommendations
On crowdsourced interactive live streaming: a Twitch.tv-based measurement study
NOSSDAV '15: Proceedings of the 25th ACM Workshop on Network and Operating Systems Support for Digital Audio and VideoEmpowered by today's rich tools for media generation and collaborative production, the multimedia service paradigm is shifting from the conventional single source, to multi-source, to many sources, and now toward crowdsource. Such crowdsourced live ...
Enhancing the crowdsourced live streaming: a deep reinforcement learning approach
NOSSDAV '19: Proceedings of the 29th ACM Workshop on Network and Operating Systems Support for Digital Audio and VideoWith the growing demand for crowdsourced live streaming (CLS), how to schedule the large-scale dynamic viewers effectively among different Content Delivery Network (CDN) providers has become one of the most significant challenges for CLS platforms. ...
On HTTP live streaming in large enterprises
SIGCOMM '13: Proceedings of the ACM SIGCOMM 2013 conference on SIGCOMMIn this work, we present a distributed caching solution which addresses the problem of efficient delivery of HTTP live streams in large private networks. With our system, we have conducted tests on a number of pilot deployments. The largest of them, ...
Comments