NUS CS COLLOQUIUM SERIES

Recent Developments in DASH - Low Latency and CMCD

Speaker
Dr Roger Zimmermann, Professor, School of Computing
Contact Person
Dr Kuldeep S. MEEL, NUS Presidential Young Professor, School of Computing
  meel@comp.nus.edu.sg

  19 Jan 2022 Wednesday, 04:00 PM to 05:30 PM

 via Zoom

Abstract:
In this talk we focus on two recent areas of interest in Dynamic Adaptive Streaming over HTTP (DASH). The first is the rise in live streaming, which has led to significant pursuits to lower and optimize the end-to-end latency. The second area explores the recently published Common Media Client Data (CMCD) standard which provides better, and more standardized, insights into large-scale streaming systems.

Live streaming remains a challenge in the adaptive streaming space due to the stringent requirements for not just quality and rebuffering, but also latency. Many solutions have been proposed to tackle streaming in general, but only few have looked into better catering to the more challenging low-latency live streaming scenarios. We describe our work on low latency, called Low-on-Latency (LoL and LoL+), in adaptive streaming systems to enhance the low-latency performance.

In September 2020, the Consumer Technology Association (CTA) published the CTA-5004: Common Media Client Data (CMCD) specification. Using this specification, a media client can convey certain information to the content delivery network servers with object requests. This information is useful in log association/analysis, quality of service/experience monitoring and delivery enhancements. We will provide an overview of CMCD, followed by a short investigation of the feasibility of CMCD in addressing one of the most common problems in the streaming domain: efficient use of shared bandwidth by multiple clients. To that effect, we implemented CMCD functions on an HTTP server and built a proof-of-concept system with CMCD-aware dash.js clients. We show that even a basic bandwidth allocation scheme enabled by CMCD reduces rebuffering rate and duration without noticeably sacrificing the video quality.


Biodata:
Roger Zimmermann is a professor of Computer Science in the School of Computing at the National University of Singapore (NUS). He is a key investigator in the Grab-NUS AI Lab, a joint research lab between the NUS Institute of Data Science and GrabTaxi Holdings. Earlier he was a deputy director with the Smart Systems Institute (SSI) and a co-director of the Centre of Social Media Innovations for Communities (COSMIC). Prior to joining NUS he held the position of Research Area Director with the Integrated Media Systems Center (IMSC) at the University of Southern California (USC). He received his M.S. and Ph.D. degrees from the University of Southern California in 1994 and 1998, respectively. Among his research interests are streaming media architectures, multimedia networking, distributed systems, applications of machine/deep learning, mobile and spatial data management. He has co-authored a book, seven patents and more than three hundred conference publications, journal articles and book chapters in the areas of multimedia, GIS and information management. He is an associate editor for IEEE MultiMedia, ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), Springer Multimedia Tools and Applications (MTAP), and IEEE Open Journal of the Communications Society (OJ-COMS). He is a distinguished member of the ACM and a senior member of the IEEE. Further information can be found at http://www.comp.nus.edu.sg/~rogerz/.