Spiteri, Sitaraman Receive 2026 IEEE INFOCOM Test of Time Paper Award for Video Streaming Algorithm
Research by CICS PhD alumnus Kevin Spiteri ’21 and Distinguished Professor Ramesh Sitaraman introduced BOLA, an adaptive bitrate algorithm now widely used in online video streaming.
Content
Anyone who has watched a video on their phone while walking across campus or commuting knows the frustration of buffering or frozen playback when the network connection drops.
Research by Manning College of Information and Computer Sciences (CICS) alumnus Kevin Spiteri ’21PhD and Distinguished Professor Ramesh K. Sitaraman has received a 2026 IEEE International Conference on Computer Communications (INFOCOM) Test of Time Paper Award for its lasting impact on how video is streamed over the internet. The work helps ensure that online videos play smoothly at high quality—even when internet connections fluctuate—a challenge that affects billions of viewers worldwide.
The award recognizes the 2016 paper, “BOLA: Near-Optimal Bitrate Adaptation for Online Videos,” authored by Spiteri, Rahul Urgaonkar of Amazon Inc., and Sitaraman, who also served as Spiteri’s doctoral advisor.
Presented annually, the Test of Time Paper Award honors research published 10–12 years earlier that has demonstrated enduring scientific and real-world influence.
The widely-cited paper proposed a new adaptive bitrate (ABR) algorithm called BOLA (Buffer Occupancy-based Lyapunov Algorithm), which is designed to enable smooth, high-quality video playback without interruptions.
At the heart of the challenge of continuous video playback is how streaming platforms decide what video quality to deliver at any given moment.
Online video streaming relies on ABR algorithms that continuously adjust video quality based on changing network conditions. Streaming platforms divide videos into small segments, each available at multiple quality levels. As a video plays, the algorithm determines which segment to download next so that playback remains smooth while maintaining the highest possible quality.
BOLA improves earlier approaches by using advanced optimization techniques to dynamically adapt video quality. The algorithm is particularly notable for its strong theoretical foundation—its decisions on playback quality are mathematically proven to be near the best possible choices.
“As you’re walking and watching video on your phone, this algorithm is plotting its next steps,” Sitaraman said. “It’s figuring out what it should download so that even if you hit a dead spot with poor connectivity, your video will continue to play.”
Since publication, BOLA has become widely used in online video streaming and serves as a benchmark for new ABR algorithms. The approach has helped major streaming platforms, including Amazon Prime, deliver smooth, high-quality video to millions of viewers each day.
Further, BOLA is part of the MPEG-DASH open-standard for video streaming on the internet and is implemented in the reference video player maintained by the DASH Industry Forum (DASH-IF).
“Video streaming on the internet is truly essential for the functioning of society,” Sitaraman said. “The ability to meaningfully improve the experience of millions of users daily is what motivates our research.”
The work has received several additional honors, including the 2023 IEEE Communications Society William R. Bennett Prize. Spiteri’s doctoral dissertation—based on this research—received both the 2021 ACM SIGMM Award for Outstanding PhD Thesis in Multimedia Computing, Communications and Applications, and the 2021 DASH-IF Best PhD Dissertation Award.
