AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model

Effective resource management in edge-cloud networks is crucial for meeting Quality of Service (QoS) requirements while minimizing operational costs. However, dynamic and fluctuating workloads pose significant challenges for accurate workload prediction and efficient resource allocation, particularl...

Full description

Saved in:
Bibliographic Details
Main Authors: Berend J. D. Gort, Godfrey M. Kibalya, Angelos Antonopoulos
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Transactions on Machine Learning in Communications and Networking
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10935743/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849337848479612928
author Berend J. D. Gort
Godfrey M. Kibalya
Angelos Antonopoulos
author_facet Berend J. D. Gort
Godfrey M. Kibalya
Angelos Antonopoulos
author_sort Berend J. D. Gort
collection DOAJ
description Effective resource management in edge-cloud networks is crucial for meeting Quality of Service (QoS) requirements while minimizing operational costs. However, dynamic and fluctuating workloads pose significant challenges for accurate workload prediction and efficient resource allocation, particularly in resource-constrained edge environments. In this paper, we introduce AERO (Adaptive Edge-cloud Resource Orchestration), a novel lightweight forecasting model designed to address these challenges. AERO features an adaptive period detection mechanism that dynamically identifies dominant periodicities in multivariate workload data, allowing it to adjust to varying patterns and abrupt changes. With fewer than 1,000 parameters, AERO is highly suitable for deployment on edge devices with limited computational capacity. We formalize our approach through a comprehensive system model and extend an existing simulation framework with predictor modules to evaluate AERO’s performance in realistic cloud-edge environments. Our extensive evaluations on real-world cloud workload datasets demonstrate that AERO achieves comparable prediction accuracy to complex state-of-the-art models with millions of parameters, while significantly reducing model size and computational overhead. In addition, simulations show that AERO improves orchestration performance, reducing energy consumption and response times compared to existing proactive and reactive approaches. Our live deployment experiments further validate these findings, demonstrating that AERO consistently delivers superior performance. These results highlight AERO as an effective solution for improving resource management and reducing operational costs in dynamic cloud-edge environments.
format Article
id doaj-art-cd33a92d5a4e481899c1c4ea57a917a9
institution Kabale University
issn 2831-316X
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Transactions on Machine Learning in Communications and Networking
spelling doaj-art-cd33a92d5a4e481899c1c4ea57a917a92025-08-20T03:44:33ZengIEEEIEEE Transactions on Machine Learning in Communications and Networking2831-316X2025-01-01346347810.1109/TMLCN.2025.355310010935743AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting ModelBerend J. D. Gort0https://orcid.org/0009-0007-4649-3890Godfrey M. Kibalya1https://orcid.org/0000-0002-7053-3756Angelos Antonopoulos2https://orcid.org/0000-0002-3546-1080Nearby Computing, Barcelona, SpainNearby Computing, Barcelona, SpainNearby Computing, Barcelona, SpainEffective resource management in edge-cloud networks is crucial for meeting Quality of Service (QoS) requirements while minimizing operational costs. However, dynamic and fluctuating workloads pose significant challenges for accurate workload prediction and efficient resource allocation, particularly in resource-constrained edge environments. In this paper, we introduce AERO (Adaptive Edge-cloud Resource Orchestration), a novel lightweight forecasting model designed to address these challenges. AERO features an adaptive period detection mechanism that dynamically identifies dominant periodicities in multivariate workload data, allowing it to adjust to varying patterns and abrupt changes. With fewer than 1,000 parameters, AERO is highly suitable for deployment on edge devices with limited computational capacity. We formalize our approach through a comprehensive system model and extend an existing simulation framework with predictor modules to evaluate AERO’s performance in realistic cloud-edge environments. Our extensive evaluations on real-world cloud workload datasets demonstrate that AERO achieves comparable prediction accuracy to complex state-of-the-art models with millions of parameters, while significantly reducing model size and computational overhead. In addition, simulations show that AERO improves orchestration performance, reducing energy consumption and response times compared to existing proactive and reactive approaches. Our live deployment experiments further validate these findings, demonstrating that AERO consistently delivers superior performance. These results highlight AERO as an effective solution for improving resource management and reducing operational costs in dynamic cloud-edge environments.https://ieeexplore.ieee.org/document/10935743/Edge computingcloud-edge orchestrationadaptive resource managementlightweight forecasting modelsworkload predictiondynamic periodicity detection
spellingShingle Berend J. D. Gort
Godfrey M. Kibalya
Angelos Antonopoulos
AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model
IEEE Transactions on Machine Learning in Communications and Networking
Edge computing
cloud-edge orchestration
adaptive resource management
lightweight forecasting models
workload prediction
dynamic periodicity detection
title AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model
title_full AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model
title_fullStr AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model
title_full_unstemmed AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model
title_short AERO: Adaptive Edge-Cloud Orchestration With a Sub-1K-Parameter Forecasting Model
title_sort aero adaptive edge cloud orchestration with a sub 1k parameter forecasting model
topic Edge computing
cloud-edge orchestration
adaptive resource management
lightweight forecasting models
workload prediction
dynamic periodicity detection
url https://ieeexplore.ieee.org/document/10935743/
work_keys_str_mv AT berendjdgort aeroadaptiveedgecloudorchestrationwithasub1kparameterforecastingmodel
AT godfreymkibalya aeroadaptiveedgecloudorchestrationwithasub1kparameterforecastingmodel
AT angelosantonopoulos aeroadaptiveedgecloudorchestrationwithasub1kparameterforecastingmodel