TUBO: A Tailored ML Framework for Reliable Network Traffic Forecasting

2025-07-21

International Conference on Distributed Computing Systems Workshops

AI Summary

Existing deep learning models struggle to reliably predict network traffic characterized by abrupt bursts and complex patterns. To address this challenge, this work proposes TUBO, a novel framework that integrates uncertainty quantification with deterministic forecasting and introduces a traffic-aware dynamic model selection mechanism. This mechanism comprises a multi-model pool, a burst detection module, and an adaptive scheduling component tailored to traffic dynamics. Experimental evaluations on three real-world network datasets demonstrate that TUBO improves overall prediction accuracy by fourfold, achieves a 94% accuracy rate for burst traffic prediction, and enhances throughput in proactive traffic engineering by factors of nine and three, respectively.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Network operation optimization through traffic forecasting holds great promise but faces key challenges due to the complex and bursty nature of traffic patterns. While deep learning models outperform traditional statistical methods in time series forecasting, they remain unreliable for network traffic due to limited adaptability to traffic fluctuations and diverse patterns. We present TUBO, a novel machine learning framework tailored for reliable traffic demand matrix (DM) forecasting. TUBO comprises two core components: a burst processor that isolates and predicts sudden traffic spikes, and a model selector that dynamically chooses the most suitable forecasting model from a diverse pool based on input characteristics and uncertainty estimation. This enables TUBO to deliver accurate, robust, and uncertainty-aware predictions. Evaluations on three real-world datasets (Abilene, GEANT, and CERNET) show that TUBO improves forecasting accuracy by up to $10 \times$ and achieves $\mathbf{9 4 \%}$ accuracy in burst occurrence prediction. As a downstream application, we apply TUBO to proactive traffic engineering (TE) and demonstrate throughput gains of $\mathbf{9} \times$ and $\mathbf{3} \times$ over reactive TE and the best prior proactive TE approach, respectively.

Problem

• network traffic forecasting
• burst traffic
• model reliability
• traffic patterns
• time series forecasting

Innovation

• traffic forecasting
• burst handling
• model selection
• uncertainty quantification
• proactive traffic engineering