Contracts: A unified lens on congestion control robustness, fairness, congestion, and generality

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Agarwal, Anup, Arun, Venkat, Seshan, Srinivasan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908395808751616
author Agarwal, Anup
Arun, Venkat
Seshan, Srinivasan
author_facet Agarwal, Anup
Arun, Venkat
Seshan, Srinivasan
contents Congestion control algorithms (CCAs) operate in partially observable environments, lacking direct visibility into link capacities, or competing flows. To ensure fair sharing of network resources, CCAs communicate their fair share through observable signals. For instance, Reno's fair share is encoded as $\propto 1/\sqrt{\texttt{loss rate}}$. We call such communication mechanisms \emph{contracts}. We show that the design choice of contracts fixes key steady-state performance metrics, including robustness to errors in congestion signals, fairness, amount of congestion (e.g., delay, loss), and generality (e.g., range of supported link rates). This results in fundamental tradeoffs between these metrics. Using properties of contracts we also identify design pitfalls that lead to starvation (extreme unfairness). We argue that CCA design and analysis should start with contracts to conscientiously pick tradeoffs and avoid pitfalls. We empirically validate our findings and discuss their implications on CCA design and network measurement.
format Preprint
id arxiv_https___arxiv_org_abs_2504_18786
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Contracts: A unified lens on congestion control robustness, fairness, congestion, and generality
Agarwal, Anup
Arun, Venkat
Seshan, Srinivasan
Networking and Internet Architecture
Congestion control algorithms (CCAs) operate in partially observable environments, lacking direct visibility into link capacities, or competing flows. To ensure fair sharing of network resources, CCAs communicate their fair share through observable signals. For instance, Reno's fair share is encoded as $\propto 1/\sqrt{\texttt{loss rate}}$. We call such communication mechanisms \emph{contracts}. We show that the design choice of contracts fixes key steady-state performance metrics, including robustness to errors in congestion signals, fairness, amount of congestion (e.g., delay, loss), and generality (e.g., range of supported link rates). This results in fundamental tradeoffs between these metrics. Using properties of contracts we also identify design pitfalls that lead to starvation (extreme unfairness). We argue that CCA design and analysis should start with contracts to conscientiously pick tradeoffs and avoid pitfalls. We empirically validate our findings and discuss their implications on CCA design and network measurement.
title Contracts: A unified lens on congestion control robustness, fairness, congestion, and generality
topic Networking and Internet Architecture
url https://arxiv.org/abs/2504.18786