| Title | : | Optimizing Streaming Media QoE over Poor Bandwidth Networks |
| Speaker | : | Dr. Mukulika Maity (IIITD) |
| Details | : | Mon, 28 Oct, 2024 2:00 PM @ SSB 334 |
| Abstract: | : | With the push towards HTTP/3, most modern browsers have started
supporting it. HTTP/3 uses QUIC, which runs on top of UDP. However, a
few Internet middleboxes tend to block or rate-limit UDP traffic;
therefore, the browsers ensure compatibility by enabling connection
racing via simultaneously initiating a TCP connection with the QUIC
one. Each time the QUIC protocol suffers, connection racing is
activated, and whichever protocol wins the race is further used for the
application. In this talk, I will present how browsers implement this
connection racing mechanism and analyze its impact on applications that
require a long-lived Internet connection, such as video streaming. We
perform a large-scale measurement study across different browsers
(Chrome/Chromium and Firefox), which helps to analyze why and how the
repeated connection racing between protocols affects adaptive streaming
QoE over 6265 YouTube sessions covering 5647 hours of
streaming. Interestingly, we observe that YouTube QoE over an HTTP/3
supported browser suffers many times, and repeated connection racing is
one of the major reasons that hinder the performance. We modified the
Chromium browser source code to disable the connection racing altogether
and observed that it improves the QoE for YouTube streaming over this
modified browser. We then design solutions to (1) decide when to enable
connection racing, (2) minimize the impact of repeated connection racing
on application QoE. We implement (1) by modifying the Chromium browser
and (2) by modifying server behavior. We evaluate both solutions
separately. We observe that both solutions improve the QoE compared to
the original case. This talk will highlight the requirement of
revisiting how browsers and application senders handle and switch
between protocols through connection racing to ensure compatibility with
middleboxes.
Bio: Mukulika Maity is an assistant professor at the computer science department of Indraprastha Institute of Information Technology Delhi (IIITD), New Delhi, India. She received her M.Tech + Ph.D. dual degree from the computer science department of the Indian Institute of Technology (IIT) Bombay in 2017. She received B.E. in 2010 in the computer science department of Bengal Engineering and Science University, Shibpur. Her Ph.D. topic was health diagnosis and congestion mitigation of wireless networks. She has worked with Prof. Bhaskaran Raman and Prof. Mythili Vutukuru at IIT Bombay. Her research interests broadly lie in the areas of wireless networks, computer networks, mobile systems, and WiFi sensing. Her research has focused on addressing some of the networking challenges faced in developing countries like India, such as dense WiFi deployment, intermittent connectivity, poor network bandwidth. She is currently working on multiple projects funded by both Govt (DST, NSC, TiH) and private organizations (Arista Networks). She is awarded an early career research award by DST, best paper awards at multiple conferences, WiFi thinkfest award, and research & teaching excellence awards at IIITD. |
