Low Complexity Approaches for End-to-End Latency Prediction

Abstract

Software Defined Networks have opened the door to statistical and AI-based techniques to improve efficiency of networking. Especially to ensure a certain Quality of Service (QoS) for specific applications by routing packets with awareness on content nature (VoIP, video, files, etc.) and its needs (latency, bandwidth, etc.) to use efficiently resources of a network. Predicting various Key Performance Indicators (KPIs) at any level may handle such problems while preserving network bandwidth. The question addressed in this work is the design of efficient and low-cost algorithms for KPI prediction, implementable at the local level. We focus on end-to-end latency prediction, for which we illustrate our approaches and results on a public dataset from the recent international challenge on GNN [1]. We propose several low complexity, locally implementable approaches, achieving significantly lower wall time both for training and inference, with marginally worse prediction accuracy compared to state-of-the-art global GNN solutions.

Low Complexity Adaptive Machine Learning Approaches for End-to-End Latency Prediction

“…We will focus here on two very simple models, although other machine learning models have also been considered in [6]. Indeed, these two models lend themselves very easily to an adaptive formulation.…”

“…Several solutions were considered in [6] to define or choose the functions f n . Since we know that the Bernstein polynomials form a basis in the set of polynomial in the interval [0; 1]; and that the approximation of any continuous function on [0; 1[ by a Bernstein polynomial converges uniformly, we were led to these polynomials:…”

“…First, we establish the benchmark performances based on the global methods presented in the paper [6] and Sections 3.1 and 3.2.…”

“…The question addressed in this work is the design of efficient, low-cost adaptive algorithms for KPI estimation, monitoring and prediction. In the present paper, we improve our previously proposed lowcost estimators [6] by adding the adaptive dimension and show that the performances are minimally modified while gaining the ability to track varying networks. We focus on end-to-end latency prediction, for which we illustrate our approaches and results on data we generated using a public generator made available after the recent international challenge [12].…”

“…We present related works in section 2. Then we present in section 3 our main results from [6]. Instead of using high performances global and high-costs methods based on Graph Neural Networks (GNNs) [10,3,12], we proposed to use standard machine learning regression methods.…”