SocialVoD: a Social Feature-based P2P System Wei Chang, and Jie Wu Presenter: En Wang Temple University, PA, USA IEEE ICPP, September, Beijing, China1.

Slides:



Advertisements
Similar presentations
Cristian Lumezanu Neil Spring Bobby Bhattacharjee Decentralized Message Ordering for Publish/Subscribe Systems.
Advertisements

謝文婷 SocialTube: P2P-assisted Video Sharing in Online Social Networks Authors: Ze Li ; Haiying Shen ; Hailang Wang ; Guoxin Liu ; Jin Li.
Efficient Information Retrieval for Ranked Queries in Cost-Effective Cloud Environments Presenter: Qin Liu a,b Joint work with Chiu C. Tan b, Jie Wu b,
Massively Distributed Database Systems Distributed Hash Spring 2014 Ki-Joune Li Pusan National University.
Search and Replication in Unstructured Peer-to-Peer Networks Pei Cao, Christine Lv., Edith Cohen, Kai Li and Scott Shenker ICS 2002.
Search in Power-Law Networks Presented by Hakim Weatherspoon CS294-4: Peer-to-Peer Systems Slides also borrowed from the following paper Path Finding Strategies.
LightFlood: An Optimal Flooding Scheme for File Search in Unstructured P2P Systems Song Jiang, Lei Guo, and Xiaodong Zhang College of William and Mary.
Small-world Overlay P2P Network
Subscription Subsumption Evaluation for Content-Based Publish/Subscribe Systems Hojjat Jafarpour, Bijit Hore, Sharad Mehrotra, and Nalini Venkatasubramanian.
Peer-to-peer Multimedia Streaming and Caching Service Jie WEI, Zhen MA May. 29.
Peer-to-Peer Based Multimedia Distribution Service Zhe Xiang, Qian Zhang, Wenwu Zhu, Zhensheng Zhang IEEE Transactions on Multimedia, Vol. 6, No. 2, April.
HyperCuP – P2P Network Boyko Syarov. 2 Outline  HyperCup: What is it?  Basic Concepts  Broadcasting Algorithm  Topology Construction  Ontology Based.
Responder Anonymity and Anonymous Peer-to-Peer File Sharing. by Vincent Scarlata, Brian Levine and Clay Shields Presentation by Saravanan.
Efficient Content Location Using Interest-based Locality in Peer-to-Peer Systems Presented by: Lin Wing Kai.
A Scalable Content-Addressable Network Authors: S. Ratnasamy, P. Francis, M. Handley, R. Karp, S. Shenker University of California, Berkeley Presenter:
Exploiting Content Localities for Efficient Search in P2P Systems Lei Guo 1 Song Jiang 2 Li Xiao 3 and Xiaodong Zhang 1 1 College of William and Mary,
Object Naming & Content based Object Search 2/3/2003.
Peer-to-peer file-sharing over mobile ad hoc networks Gang Ding and Bharat Bhargava Department of Computer Sciences Purdue University Pervasive Computing.
1CS 6401 Peer-to-Peer Networks Outline Overview Gnutella Structured Overlays BitTorrent.
P2P File Sharing Systems
Mobile Ad-hoc Pastry (MADPastry) Niloy Ganguly. Problem of normal DHT in MANET No co-relation between overlay logical hop and physical hop – Low bandwidth,
INTRODUCTION TO PEER TO PEER NETWORKS Z.M. Joseph CSE 6392 – DB Exploration Spring 2006 CSE, UT Arlington.
Roger ZimmermannCOMPSAC 2004, September 30 Spatial Data Query Support in Peer-to-Peer Systems Roger Zimmermann, Wei-Shinn Ku, and Haojun Wang Computer.
© Y. Zhu and Y. University of North Carolina at Charlotte, USA 1 Chapter 1: Social-based Routing Protocols in Opportunistic Networks Ying Zhu and.
1 Speaker : 童耀民 MA1G Authors: Ze Li Dept. of Electr. & Comput. Eng., Clemson Univ., Clemson, SC, USA Haiying Shen ; Hailang Wang ; Guoxin.
Fundamental Lower Bound for Node Buffer Size in Intermittently Connected Wireless Networks Yuanzhong Xu, Xinbing Wang Shanghai Jiao Tong University, China.
By Shobana Padmanabhan Sep 12, 2007 CSE 473 Class #4: P2P Section 2.6 of textbook (some pictures here are from the book)
Developing Analytical Framework to Measure Robustness of Peer-to-Peer Networks Niloy Ganguly.
09/07/2004Peer-to-Peer Systems in Mobile Ad-hoc Networks 1 Lookup Service for Peer-to-Peer Systems in Mobile Ad-hoc Networks M. Tech Project Presentation.
A Delaunay Triangulation Architecture Supporting Churn and User Mobility in MMVEs Mohsen Ghaffari, Behnoosh Hariri and Shervin Shirmohammadi Advanced Communications.
HERO: Online Real-time Vehicle Tracking in Shanghai Xuejia Lu 11/17/2008.
Lyon, June 26th 2006 ICPS'06: IEEE International Conference on Pervasive Services 2006 Routing and Localization Services in Self-Organizing Wireless Ad-Hoc.
Social Feature-based Multi-path Routing in Delay Tolerant Networks
Stochastic sleep scheduling (SSS) for large scale wireless sensor networks Yaxiong Zhao Jie Wu Computer and Information Sciences Temple University.
An efficient secure distributed anonymous routing protocol for mobile and wireless ad hoc networks Authors: A. Boukerche, K. El-Khatib, L. Xu, L. Korba.
Peer to Peer Research survey TingYang Chang. Intro. Of P2P Computers of the system was known as peers which sharing data files with each other. Build.
Tonghong Li, Yuanzhen Li, and Jianxin Liao Department of Computer Science Technical University of Madrid, Spain Beijing University of Posts & Telecommunications.
SOS: Security Overlay Service Angelos D. Keromytis, Vishal Misra, Daniel Rubenstein- Columbia University ACM SIGCOMM 2002 CONFERENCE, PITTSBURGH PA, AUG.
Dynamic Source Routing in ad hoc wireless networks Alexander Stojanovic IST Lisabon 1.
Content Sharing over Smartphone-Based Delay- Tolerant Networks.
Quantitative Evaluation of Unstructured Peer-to-Peer Architectures Fabrício Benevenuto José Ismael Jr. Jussara M. Almeida Department of Computer Science.
A Collaborative and Semantic Data Management Framework for Ubiquitous Computing Environment International Conference of Embedded and Ubiquitous Computing.
Load-Balancing Routing in Multichannel Hybrid Wireless Networks With Single Network Interface So, J.; Vaidya, N. H.; Vehicular Technology, IEEE Transactions.
A Novel Multicast Routing Protocol for Mobile Ad Hoc Networks Zeyad M. Alfawaer, GuiWei Hua, and Noraziah Ahmed American Journal of Applied Sciences 4:
An Improved Kademlia Protocol In a VoIP System Xiao Wu , Cuiyun Fu and Huiyou Chang Department of Computer Science, Zhongshan University, Guangzhou, China.
User-Centric Data Dissemination in Disruption Tolerant Networks Wei Gao and Guohong Cao Dept. of Computer Science and Engineering Pennsylvania State University.
Kaleidoscope – Adding Colors to Kademlia Gil Einziger, Roy Friedman, Eyal Kibbar Computer Science, Technion 1.
1 Peer-to-Peer Technologies Seminar by: Kunal Goswami (05IT6006) School of Information Technology Guided by: Prof. C.R.Mandal, School of Information Technology.
Scalable Content- Addressable Networks Prepared by Kuhan Paramsothy March 5, 2007.
Efficient P2P Search by Exploiting Localities in Peer Community and Individual Peers A DISC’04 paper Lei Guo 1 Song Jiang 2 Li Xiao 3 and Xiaodong Zhang.
SocialTube: P2P-assisted Video Sharing in Online Social Networks
A Multicast Mechanism in WiMax Mesh Network Jianfeng Chen, Wenhua Jiao, Pin Jiang, Qian Guo Asia-Pacific Conference on Communications, (APCC '06)
1 - CS7701 – Fall 2004 Review of: Detecting Network Intrusions via Sampling: A Game Theoretic Approach Paper by: – Murali Kodialam (Bell Labs) – T.V. Lakshman.
DHT-based unicast for mobile ad hoc networks Thomas Zahn, Jochen Schiller Institute of Computer Science Freie Universitat Berlin 報告 : 羅世豪.
Efficient AOI-Cast for Peer-to-Peer Networked Virtual Environments.
Dual Centric Data Center Network Architectures DAWEI LI, JIE WU (TEMPLE UNIVERSITY) ZHIYONG LIU, AND FA ZHANG (CHINESE ACADEMY OF SCIENCES) ICPP 2015.
Copyright © 2011, Scalable and Energy-Efficient Broadcasting in Multi-hop Cluster-Based Wireless Sensor Networks Long Cheng ∗ †, Sajal K. Das†,
KAIS T On the problem of placing Mobility Anchor Points in Wireless Mesh Networks Lei Wu & Bjorn Lanfeldt, Wireless Mesh Community Networks Workshop, 2006.
Cooperative Mobile Live Streaming Considering Neighbor Reception SPEAKER: BO-YU HUANG ADVISOR: DR. HO-TING WU 2015/10/15 1.
Peer to Peer Network Design Discovery and Routing algorithms
1 Scalable AOI-cast for Peer-to-Peer Networked Virtual Environments Jehn-Ruey Jiang, Yu-Li Huang, and Shun-Yun Hu CSIE, National Central University, Taiwan.
LOOKING UP DATA IN P2P SYSTEMS Hari Balakrishnan M. Frans Kaashoek David Karger Robert Morris Ion Stoica MIT LCS.
P2P Search COP P2P Search Techniques Centralized P2P systems  e.g. Napster, Decentralized & unstructured P2P systems  e.g. Gnutella.
Large Scale Sharing Marco F. Duarte COMP 520: Distributed Systems September 19, 2004.
Malugo – a scalable peer-to-peer storage system..
1 UNIT 13 The World Wide Web. Introduction 2 The World Wide Web: ▫ Commonly referred to as WWW or the Web. ▫ Is a service on the Internet. It consists.
Distributed Caching and Adaptive Search in Multilayer P2P Networks Chen Wang, Li Xiao, Yunhao Liu, Pei Zheng The 24th International Conference on Distributed.
CS Spring 2010 CS 414 – Multimedia Systems Design Lecture 24 – Introduction to Peer-to-Peer (P2P) Systems Klara Nahrstedt (presented by Long Vu)
A Scalable content-addressable network
A Semantic Peer-to-Peer Overlay for Web Services Discovery
Presentation transcript:

SocialVoD: a Social Feature-based P2P System Wei Chang, and Jie Wu Presenter: En Wang Temple University, PA, USA IEEE ICPP, September, Beijing, China1

Outline 1. Introduction 2. Background and Related Work 3. Vimeo Category and User Interests 3. Vimeo Category and User Interests 6. Conclusion 5. Evaluation 5. Evaluation 4. Overlay Design and File Searching IEEE ICPP, September, Beijing, China2

Introduction  Video-on-demand (VoD) service has been explosively growing since its first appearance. For maintaining an acceptable buffering delay, the bandwidth costs have become a huge burden for the service providers.  Complementing the conventional client-server architecture (CS) with a peer-to-peer system (P2P) can significantly reduce the central server’s bandwidth demands.  Previous works focus on establishing a P2P overlay for each video, which not only generates prohibitive costs for maintaining the overlay, but also create plenty of redundant links between a pair of nodes on different overlays. IEEE ICPP, September, Beijing, China3

Introduction (cont.)  Many VoD sites provide channel subscription functions to users.  Considering that the subscribers of a channel are likely to watch the same videos from the channel, SocialTube proposes an interest-based per-channel hierarchical overlay. IEEE ICPP, September, Beijing, China4

Introduction (cont.)  SocialTube has two problems. It is not efficient for subscribers to explore (unsubscribed channels’) videos belonging to other categories. It is not applicable for the VoD sites, such as Vimeo, who are rich in category information. 5

Introduction (cont.)  In this paper, we propose a new hierarchical P2P overlay structure, called SocialVoD, by exploring both the subscription relationships and channels’ similarities on Vimeo.  The key design of SocialVoD is the utilization of channel subscribers’ social closeness: the closer the watching preferences of users are, the better the overlay connectivity should be between them.  We not only build interest-based per-channel overlays, but also, in a high-level, we organize these overlays according to channels’ existing category information: by neighboring the channel overlays, whose category information differs in one dimension, the resulting graph contains enough routing “hints” for seeking other unsubscribed channels’ videos. 6

Outline 1. Introduction 2. Background and Related Work 3. Vimeo Category and User Interests 3. Vimeo Category and User Interests 6. Conclusion 5. Evaluation 5. Evaluation 4. Overlay Design and File Searching IEEE ICPP, September, Beijing, China7

Background and Related Work P2P overlay design: Without social information Hypercup and et al. With social information NetTube: same watching history SocialTube: same channel subscription SocialVoD: neighboring nodes with similar watching preferences IEEE ICPP, September, Beijing, China8

Outline 1. Introduction 2. Background and Related Work 3. Vimeo Category and User Interests 3. Vimeo Category and User Interests 6. Conclusion 5. Evaluation 5. Evaluation 4. Overlay Design and File Searching IEEE ICPP, September, Beijing, China9

Vimeo Category and User Interests  It has been shown that most videos on VoD systems are short, and there are strong clustering behaviors among users.  We further investigate the category features on Vimeo, and the change of users’ interests over years.  There are 58.06% popular channels (>300 subscribers) associated with more than one category feature.  Users’ watching preferences do change over the years. 10

Watching Pattern of Users  We summarize users’ watching pattern of as follows: Users frequently look for the videos from their subscribed channels; Users are likely to seek other unsubscribed channels’ videos, which satisfy their watching habits; The watching habits may change. Users may access certain videos to explore new preferences.  Based on these features, we design the new P2P overlay structure, SocialVoD. IEEE ICPP, September, Beijing, China11

Outline 1. Introduction 2. Background and Related Work 3. Vimeo Category and User Interests 3. Vimeo Category and User Interests 6. Conclusion 5. Evaluation 5. Evaluation 4. Overlay Design and File Searching IEEE ICPP, September, Beijing, China12

Overlay Design: basic model  Feature coordinate of a video:,  e.x., in “Language” category dimension, may indicate English, Chinese, Spanish, and et al.  Based on the feature coordinates, the category-feature space is partitioned into groups, and each channel belongs to one and only one group.  Mapping: real-world users  nodes on P2P  Establishing a logical connection in P2P: storing the IP address of another user. IEEE ICPP, September, Beijing, China13

Overlay Design: overlay construction  There are three types of connections in SocialVoD, and the number of connections for each type following:, Intra-channel: V i,V j subscribes to the same channel, i.e. V 1 V 2. Intra-group: different channel but same coordinate, i.e. V 1 V 5. Inter-group: feature coordinates differ in one dimension, i.e. V 1 V 6, where coordinates of V 1 and V 6 are (0,1) and (1,1). 14

Overlay Construction (cont.)  SocialVoD creates a two-tier community structure and three levels of overlays. Different types of queries are conducted by different overlays.  Link management New node joining Link updating SocialVoD gives more emphasis on how to reach a destination, instead of where the destination is. 15

Routing in SocialVoD  A query is conducted by two steps in SocialVoD: Feature-spaced inter-group routing: use the highest-level overlay to locate the group with the specific feature coordinate. Intra-group/channel searching: apply the conventional random walk-based approach to search the target file within the specific group or channel.  In order to control the network traffic, each query has a limited number of query copies, assuming h. During inter-group routing, one should avoid the situations where several query messages search the same space. IEEE ICPP, September, Beijing, China16

Coordinate Routing Sequence  Feature difference set : a set of different dimensions between source S and destination D.  Coordinate routing sequences: suppose S differs D in k dimensions, ; there are k coordinate sequences forming k node-disjoint paths.  For example: : blue path : red path 17

The Selection of Coordinate Routing Sequences  When h<k, the source S needs to determine which path (or coordinate routing sequence) has more of a chance to reach the destination group.  We approximate a coordinate routing sequence’s success rate by subcategories’ popularity:  Source node always selects the top-h sequences with the highest success rate. This method essentially explores the rare intergroup links first., where IEEE ICPP, September, Beijing, China18

Feature Space Routing  Source node: Are there redundant searching paths? If not, which coordinate sequences should be used?  Relay node: Am I in the destination feature group? For a query message, does it use shortest- path mode or non- shortest path mode? Who should I forward the message to? 19

Extension  Multi-channel subscription Initial node selection Feature matching shortcut  Pre-fetch for new videos Random walk-based propagation  Routing for non-subscribers Most frequently searched subcategory features Local overlay keeps updating while developing his own watching preference IEEE ICPP, September, Beijing, China20

Outline 1. Introduction 2. Background and Related Work 3. Vimeo Category and User Interests 3. Vimeo Category and User Interests 6. Conclusion 5. Evaluation 5. Evaluation 4. Overlay Design and File Searching IEEE ICPP, September, Beijing, China21

Evaluation 22

Outline 1. Introduction 2. Background and Related Work 3. Vimeo Category and User Interests 3. Vimeo Category and User Interests 6. Conclusion 5. Evaluation 5. Evaluation 4. Overlay Design and File Searching IEEE ICPP, September, Beijing, China23

Conclusion  We find that the existing P2P systems do not fully utilize the social properties behind the users’ watching patterns.  Based on the existing channel subscription knowledge and category structure on the Vimeo platform, we introduce a new hierarchical P2P overlay structure.  Our new system explores the existing category information as an overlay’s construction “guide” and routing “hints”, which significantly reduces the searching space of videos.  We provide extensive simulations to show that our new system can efficiently locate the files and is applicable in a large scale. IEEE ICPP, September, Beijing, China24

Thank you! IEEE ICPP, September, Beijing, China25