Fair Queuing Aware Congestion Control
Maximilian Bachl

TL;DR
This paper introduces a novel algorithm that detects fair queuing on network paths with high accuracy, enabling adaptive congestion control that reduces latency and allows flows to customize delay- throughput tradeoffs.
Contribution
It presents the first continuous fair queuing detection algorithm and demonstrates how it can improve congestion control flexibility and network performance.
Findings
Detection accuracy over 95%
Reduced latency when fair queuing is detected
Flows can specify individual delay tolerances
Abstract
Fair queuing is becoming increasingly prevalent in the internet and has been shown to improve performance in many circumstances. Performance could be improved even more if endpoints could detect the presence of fair queuing on a certain path and adjust their congestion control accordingly. If fair queuing is detected, the congestion control would not have to take cross traffic into account, which allows for more flexibility. In this paper, we develop the first algorithm that continuously checks if fair queuing is present on a path, with an accuracy of over 95%. When fair queuing is detected, a different congestion control can be chosen, which can result in reduced latency. Also, each flow can then specify how much queuing delay it allows, meaning that it can choose its own tradeoff between throughput and latency.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsNetwork Traffic and Congestion Control · Advanced Wireless Network Optimization · Optimization and Search Problems
