Understanding the Modern Internet's Heterogeneous Congestion Control Landscape

Research in Internet congestion control has experienced a renaissance in recent years, driven by two key developments. In 2016, Google introduced and deployed BBR, a congestion control algorithm that marks a significant departure from traditional loss-based algorithms like CUBIC and New Reno. Additionally, Internet transport is transitioning to the userspace with the adoption of QUIC, a new transport stack that is already widely deployed and poised to become the default with HTTP3. While both these advancements present unique challenges, they also introduce considerable heterogeneity into the Internet's congestion control landscape. Historically, Internet congestion control has remained stable due to a homogeneous ecosystem of well-understood AIMD congestion control algorithms. However, the recent innovations in BBR and QUIC are set to challenge this paradigm. This extended abstract summarises the contributions of my thesis, which examines the impact of these two developments on the Internet's congestion control landscape and explores how they are likely to influence its evolution in the near future.