Causal-temporal analysis-based feature selection for predicting application performance degradation in edge clouds

Distributed, heterogeneous edge clouds are key enablers for next-generation network applications that need to be highly reliable, always available, with guaranteed Quality-of-Service (QoS). Preventing performance degradations by predicting infrastructure-related faults using Machine Learning-based analytics and handling them proactively is critical for maintaining application QoS.

Research paper

#FeatureSelection #CausalTemporalAnalysis #EricssonResearch

Predicting performance degradations can be challenging due to the diversity of the underlying causes. In this paper, we propose an automated feature selection system that uses causal-temporal analysis to find the infrastructure metrics that have causal relationships with application metrics. The selected features are further used to train ML models for predicting application performance degradation. We have validated a proof-of-concept of our system on a Kubernetes testbed.

Full abstract in IEEEXplore DOI: 10.1109/ICC45041.2023.10279527

Authors

Behshid Shayesteh, CIISE, Concordia University, Montréal, Canada

Chunyan Fu, Ericsson Research

Amin Ebrahimzadeh, CIISE, Concordia University, Montréal, Canada

Roch H. Glitho, CIISE, Concordia University, Montréal, Canada and University of Western Cape, Cape Town, South Africa

Presented at IEEE International Conference on Communications ICC 2023

©2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse.

Download the paper

Causal-temporal analysis-based feature selection for predicting application performance degradation in edge clouds

pdf