Prose-to-P4: Leveraging High Level Languages
This addresses the problem of reducing time, effort, and domain-knowledge for network programmers, though it is incremental as it builds on existing high-level language abstractions.
The paper tackles the difficulty of developing networking applications in low-level languages like P4 by proposing to use Large Language Models (LLMs) to translate natural language prose into high-level networking code, with preliminary results showing promise in generating Lucid code.
Languages such as P4 and NPL have enabled a wide and diverse range of networking applications that take advantage of programmable dataplanes. However, software development in these languages is difficult. To address this issue, high-level languages have been designed to offer programmers powerful abstractions that reduce the time, effort and domain-knowledge required for developing networking applications. These languages are then translated by a compiler into P4/NPL code. Inspired by the recent success of Large Language Models (LLMs) in the task of code generation, we propose to raise the level of abstraction even higher, employing LLMs to translate prose into high-level networking code. We analyze the problem, focusing on the motivation and opportunities, as well as the challenges involved and sketch out a roadmap for the development of a system that can generate high-level dataplane code from natural language instructions. We present some promising preliminary results on generating Lucid code from natural language.