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

TL;DR

TaNG introduces a neural network-based packet classification method that ensures complete rule coverage, reduces complexity, and significantly improves throughput and stability on large rulesets using GPU acceleration.

Contribution

It presents a novel semi-structured neural network approach with a rule update mechanism and a CPU-GPU framework for efficient large-scale packet classification.

Findings

Achieves 12.19x higher throughput than NuevoMatch.

Attains 98.84x higher performance stability.

Effectively handles 512k rulesets with improved efficiency.

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.