Automating In-Network Machine Learning

TL;DR

This paper introduces Planter, an open-source framework that enables the deployment of various machine learning models on programmable network devices, demonstrating feasibility and efficiency for real-time applications.

Contribution

It provides the first general, extensible solution for mapping trained machine learning models to programmable network hardware.

Findings

Planter supports diverse models and targets with easy extensibility.

In-network ML algorithms run at line rate with minimal latency impact.

The approach maintains high accuracy with negligible trade-offs.

Abstract

Using programmable network devices to aid in-network machine learning has been the focus of significant research. However, most of the research was of a limited scope, providing a proof of concept or describing a closed-source algorithm. To date, no general solution has been provided for mapping machine learning algorithms to programmable network devices. In this paper, we present Planter, an open-source, modular framework for mapping trained machine learning models to programmable devices. Planter supports a wide range of machine learning models, multiple targets and can be easily extended. The evaluation of Planter compares different mapping approaches, and demonstrates the feasibility, performance, and resource efficiency for applications such as anomaly detection, financial transactions, and quality of experience. The results show that Planter-based in-network machine learning algorithms can run at line rate, have a negligible effect on latency, coexist with standard switching functionality, and have no or minor accuracy trade-offs.