Can Neural Networks Provide Latent Embeddings for Telemetry-Aware Greedy Routing?

TL;DR

This paper introduces Placer, a neural network-based algorithm that creates latent embeddings for network nodes, enabling efficient and explainable telemetry-aware greedy routing without complex path computations.

Contribution

The paper presents Placer, a novel message passing neural network approach that generates latent node embeddings to improve routing efficiency and interpretability in telemetry-aware networks.

Findings

Enables quick greedy routing using latent embeddings

Provides visualization of network event impacts on routing

Reduces computational complexity compared to traditional methods

Abstract

Telemetry-Aware routing promises to increase efficacy and responsiveness to traffic surges in computer networks. Recent research leverages Machine Learning to deal with the complex dependency between network state and routing, but sacrifices explainability of routing decisions due to the black-box nature of the proposed neural routing modules. We propose \emph{Placer}, a novel algorithm using Message Passing Networks to transform network states into latent node embeddings. These embeddings facilitate quick greedy next-hop routing without directly solving the all-pairs shortest paths problem, and let us visualize how certain network events shape routing decisions.