TaNG: Modeling Packet Classification with TSS-assisted Neural Networks on GPUs

2026-01-06

arXiv.org

AI Summary

This work addresses the limitations of existing learning-based packet classification methods, which often suffer from incomplete coverage or redundancy when handling overlapping rules and lack efficient GPU support for scaling to large rule sets. The authors propose TaNG, the first approach to integrate a semi-structured neural network with the tuple space model, enabling a single network trained on multidimensional features to achieve complete rule coverage without rule replication while supporting efficient updates. Furthermore, TaNG introduces a CPU-GPU hybrid streaming architecture to enhance throughput. Experimental results demonstrate that, under a 512K-rule setting, TaNG achieves 12.19× and 9.37× higher throughput than NuevoMatch and NeuTree, respectively, and improves performance stability by up to 98.84× and 156.98×.

Technology Category

• Machine Learning: Structured Learning
• Natural Language Processing: Learning & Optimization for NLP
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Packet classification is a core function in software-defined networks, and learning-based methods have recently shown significant throughput gains on large-scale rulesets. However, existing learning-based approaches struggle with overlapping rules, leading to incomplete model coverage or excessive rule replication. Their limited GPU integration also hampers performance with large-scale rulesets. To address these issues, we propose TaNG, which utilizes a single neural network trained on multi-dimensional features to ensure complete coverage without duplicating rules. TaNG employs a semi-structured design that combines the neural network model with a tuple space, reducing model complexity. Furthermore, we develop a mechanism based on the semi-structure for rule updates. Finally, we implement a CPU-GPU hybrid streaming framework tailored for learning-based methods, further enhancing throughput. On our GPU-based classification framework with 512k rulesets, TaNG achieves 12.19x and 9.37x higher throughput and 98.84x and 156.98x higher performance stability compared to two state-of-the-art learning methods NuevoMatch and NeuTree, respectively.

Problem

• packet classification
• overlapping rules
• GPU integration
• large-scale rulesets
• model coverage

Innovation

• neural network
• packet classification
• tuple space
• GPU acceleration
• rule overlap