Online video is the killer application of the Internet. It is predicted that more than 85% of the consumer traffic on the Internet will be video-related by 2016. Yet, the future economic viability of online video rests squarely on our ability to understand how viewers interact with video content. For instance:
· If a video fails to start up quickly, would the viewer abandon?
· If a video freezes in the middle, would the viewer watch fewer minutes?
· If videos fail to load, is the viewer less likely to return to the same site?
In this talk, we outline scientific answers to these and other such questions, establishing for the first time a causal link between video performance and viewer behavior. One of the largest such studies, our work analyzes the video viewing habits of over 6.7 million viewers who in aggregate watched almost 26 million videos. To go beyond mere correlation and to establish causality, we develop a novel technique based on Quasi-Experimental Designs (QEDs). While QEDs are well known in the medical and social sciences, our work represents its first use in network performance research and is of independent interest.
Ramesh K. Sitaraman received his B. Tech. in electrical engineering from the Indian Institute of Technology, Madras. He obtained his Ph.D. in computer science from Princeton University. He is currently a faculty member in the School of Computer Science at the University of Massachusetts at Amherst and was named an Akamai Fellow. As a principal architect, he helped build the Akamai network that currently serves a significant fraction of the world’s web content, streaming videos, and online applications. His research focuses on all aspects of Internet-scale distributed systems, including algorithms, architectures, performance, energy efficiency, and economics.
Host: Ravi Sundaram