A Scalable and Efficient Live Video Streaming Over DHT Overlay Network
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Live video streaming becomes one of the most popular internet activities in the recent years. Along with its popularity, the problems such as scalability, availability and low-latency are still the maindrawbacks of the peer-to-peer (P2P) live video streaming. The DHT-aided chunk-driven overlay (DCO)[3] tried to overcome these problems by using the Distributed Hash Table (DHT) in live video streaming. Even the performance of DCO is better than other traditional works such as tree-based system, in termsof vulnerability to churn, and mesh-based system, in terms of high delay and message overhead, thereare some problems in DCO which need to be solved. The coordinator is easy to get overloaded by manyrequests because one coordinator is responsible for storing the chunk index of several segments. Moreover, the number of nodes that the video server is responsible for sending the video segment toare not clearly defined. To solve these problems, in this paper, a more efficient approach of using thecoordinator nodes is proposed. Furthermore, the threshold is carefully calculated in order to use theserver capacity efficiently. The server is only responsible for sending the video segment to the thresholdnodes. After that, the video segment will be shared among nodes in the network until every node receivesthe video segment. The performance evaluation results show that our work is more efficient than DCO,and the server capacity can be used efficiently if the threshold is careful defined.
Abstract
Live video streaming becomes one of the most popular internet activities in the recent years. Along with its popularity, the problems such as scalability, availability and low-latency are still the maindrawbacks of the peer-to-peer (P2P) live video streaming. The DHT-aided chunk-driven overlay (DCO)[3] tried to overcome these problems by using the Distributed Hash Table (DHT) in live video streaming. Even the performance of DCO is better than other traditional works such as tree-based system, in termsof vulnerability to churn, and mesh-based system, in terms of high delay and message overhead, thereare some problems in DCO which need to be solved. The coordinator is easy to get overloaded by manyrequests because one coordinator is responsible for storing the chunk index of several segments. Moreover, the number of nodes that the video server is responsible for sending the video segment toare not clearly defined. To solve these problems, in this paper, a more efficient approach of using thecoordinator nodes is proposed. Furthermore, the threshold is carefully calculated in order to use theserver capacity efficiently. The server is only responsible for sending the video segment to the thresholdnodes. After that, the video segment will be shared among nodes in the network until every node receivesthe video segment. The performance evaluation results show that our work is more efficient than DCO,and the server capacity can be used efficiently if the threshold is careful defined.
참고문헌 (20)
- Rescorla E., "Introduction to distributed hash tables", IETF-65 Technical Plenary, ~ (2006)
- Zhang H., "Distributed Hash Table: Theory, Platforms and Application" (2013), Springer
- Shen H., "A DHT-aided chunkdriven overlay for scalable and efficient peer-to-peer live streaming", IEEE Trans. on Parallel and Distributed Systems, 24(11), pp.2125~2137 (2013)
- Asaduzzaman S., "CliqueStream : an efficient and fault-resilient live streaming network on a clustered peer-to-peer overlay", Proc. Eighth Intl Conf. Peer-to-Peer Computing (P2P), pp.269~278 (2008)
- Yin H., "Design and deployment of a hybrid CDNP2P system for live video streaming : experiences with LiveSky", Proc. 17th ACM Intl Conf. Multimedia (MM), pp.25~34 (2009)
- Kouchi T., "Maintaining Tree-Structured P2P Overlay Being Resilient to Simultaneous Leave of Several Peers", IEICE Trans. Information and Systems, E98-D(9), pp.1667~1674 (2015)
- Picconi F., "Is there a future for meshbased live video streaming?", Proc. Eighth Intl Conf. Peer-to-Peer Computing(P2P), pp.289~298 (2008)
- Guo Y., "AQCS : adaptive queuebased chunk scheduling for P2P live streaming", Proc. Seventh Intl IFIP-TC6 Networking Conf. AdHoc and Sensor Networks, Wireless Networks, Next Generation Internet (2008)
- Bideh M. K., "Adaptive content-and-deadline aware chunk scheduling in mesh-based P2P video streaming", Peer-to-Peer Netw. Appl., 436~448 (2016)
- Venkataraman V., "Chunkyspread: multi-tree unstructured peer-to-peer multicast", Proc. Fifth Intl Workshop Peer-to-Peer Systems (IPTPS) (2006)
- Wang F., "mTreebone : a hybrid tree", mesh overlay for application-layer live video multicast, 2007, 49~
- Pal K., "A new hybrid approach for overlay construction in P2P live streaming", IEEE Intl Conf. ICACCI (2015)
- Lim P. U., "Cloud assist live video streaming over DHT overlay network", The Trans. of The Korean Institute of Electrical Engineers
- Zhou Y., "Simple model for analyzing P2P streaming protocols", IEEE Intl Conf. Network Protocols(ICNP), pp.226~235 (2007)
- Lan S., "Dynamic asynchronous buffer management to improve data continuity in P2P live streaming", IEEE 3rd Intl Conf. Computer Research and Development (ICCRD), pp.65~69 (2011)
- Chen Z., "Improving the Video Data Lookup Efficiency Peer-to-Peer Live Streaming Network", Journal of Advances in Computer Network, 1(3) (2013)
- Yiu W., "VMesh: Distributed Segment Storage for Peer-to-Peer Interactive Video Streaming", Selected Areas in Communications, IEEE Journal, 25(9), pp.1717~1731 (2007)
- 림펑언, "가상 메시 오버레이 네트워크상에서의 P2P VOD 서비스를 위한 클러스터 세그먼트 인덱스 기법", 전기학회논문지, 65(6), pp.1052~1059 (2016), 대한전기학회[DOI]
- 김기성, "소셜 네트워크의 연결관계를 고려한 주제별 잠재고객 탐색기법", 정보기술아키텍처 연구, 12(4), pp.603~611 (2015), 한국엔터프라이즈아키텍처학회
- 권응주, "소셜 네트워크에서 감정단어의 단계별 코사인 유사도 기법을 이용한 추천시스템", 정보기술아키텍처 연구, 9(3), pp.333~344 (2012), 한국엔터프라이즈아키텍처학회