제 목 Trace-driven analysis of IPTV and online social networks
강 사 Dr. Meeyoung Cha
소 속 Max Planck Institute for Software Systems (MPI-SWS)
약 력 Meeyoung Cha is a post-doctoral researcher at Max Planck Institute for Software Systems (MPI-SWS). She received a Ph.D. degree in Computer Science from KAIST in 2008. Meeyoung’s research interests are in the design and analysis of large-scale networked systems. Her recent work has focused on multimedia streaming systems and online social networks. She won the best paper award at ACM IMC 2007 for her work characterizing the YouTube workload.
일 시 2008. 5. 26 (월) 오후 4시
장 소 301동 520호
Abstract
In this talk, I will present the first analysis of IPTV workloads based on traces collected from one of the largest IPTV networks. Our study characterizes several properties of what people watch and how people find content. I will also discuss the feasibility of using peer-to-peer distribution for scalable IPTV system and supporting advanced viewing controls such as DVD-like functionalities, content recommendations, and target advertisements.
In the second part of the talk, I will present a trace-driven analysis of how information cascades through friend links in the Flickr online social network. I refer to such information dissemination via social links as a social cascade. Our analysis of a real trace of 1,000 popular photos from Flickr.com reveals a strong correlation between the times when friends bookmark the same photo. Our findings show that bookmarking cascades constitute an important mechanism by which information propagates in the Flickr social network.