DeeP-TE: Data-enabled Predictive Traffic Engineering

TL;DR

DeeP-TE is a novel data-driven algorithm that adaptively optimizes network routing to minimize congestion, using historical data without explicit traffic matrix estimation, thus improving performance and stability.

Contribution

The paper introduces DeeP-TE, a predictive traffic engineering method that directly uses historical routing and link data for adaptive routing without traffic matrix estimation.

Findings

Achieves near-optimal network congestion control.

Reduces routing variations significantly compared to baseline methods.

Demonstrates effectiveness on real network topologies.

Abstract

Routing configurations of a network should constantly adapt to traffic variations to achieve good network performance. Adaptive routing faces two main challenges: 1) how to accurately measure/estimate time-varying traffic matrices? 2) how to control the network and application performance degradation caused by frequent route changes? In this paper, we develop a novel data-enabled predictive traffic engineering (DeeP-TE) algorithm that minimizes the network congestion by gracefully adapting routing configurations over time. Our control algorithm can generate routing updates directly from the historical routing data and the corresponding link rate data, without direct traffic matrix measurement or estimation. Numerical experiments on real network topologies with real traffic matrices demonstrate that the proposed DeeP-TE routing adaptation algorithm can achieve close-to-optimal control effectiveness with significantly lower routing variations than the baseline methods.