Toward Non-Expert Customized Congestion Control

2025-06-08

ICC 2025 - IEEE International Conference on Communications

AI Summary

This work proposes NECC, a novel framework that empowers non-expert users to design, generate, and deploy customized congestion control algorithms through natural language descriptions. Existing general-purpose congestion control schemes often fail to meet diverse application requirements, while developing tailored solutions typically demands deep expertise in network systems, limiting broader participation. NECC bridges this gap by integrating large language models with Berkeley Packet Filter (BPF) technology, enabling intuitive specification and automatic implementation of user-defined congestion control logic. Experimental evaluation in real-world network environments demonstrates that NECC is both efficient and practical, significantly lowering the barrier to entry for network protocol development and establishing a new paradigm for programmable congestion control.

Technology Category

• Multiagent Systems: Agent Communication
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

General-purpose congestion control algorithms (CCAs) are designed to achieve general congestion control goals, but they may not meet the specific requirements of certain users. Customized CCAs can meet certain users' specific requirements; however, non-expert users often lack the expertise to implement them. In this paper, we present an exploratory non-expert customized CCA framework, named NECC, which enables non-expert users to easily model, implement, and deploy their customized CCAs by leveraging Large Language Models and the Berkeley Packet Filter (BPF) interface. To the best of our knowledge, we are the first to address the customized CCA implementation problem. Our evaluations using real-world CCAs show that the performance of NECC is very promising, and we discuss the insights that we find and possible future research directions.

Problem

• congestion control
• customized CCA
• non-expert users
• networking
• implementation

Innovation

• customized congestion control
• large language models
• Berkeley Packet Filter
• non-expert framework
• networking