TERA-Scaler for a Proactive Auto-Scaling of e-Business Microservices

Souheir Merkouche, Chafia Bouanaka

2023

Abstract

In this research work, we present a novel multicriteria auto-scaling strategy aiming at reducing the operational costs of microservice-based e-business systems in the cloud. Our proposed solution, TERA-Scaler for instance, is designed to be aware of dependencies and to minimize resource consumption while maximizing system performance. To achieve these objectives, we adopt a proactive formal approach that leverages predictive techniques to anticipate the future state of the system components, enabling earlier scaling of the system to handle future loads. We implement the proposed auto-scaling process for e-business microservices using Kubernetes, and conduct experiments to evaluate the performance of our approach. The results show that TERA-Scaler outperforms the Kubernetes horizontal pod autoscaler, achieving a 39.5% reduction in response time and demonstrating the effectiveness of our proposed strategy.

Paper Citation

in Harvard Style

Merkouche S. and Bouanaka C. (2023). TERA-Scaler for a Proactive Auto-Scaling of e-Business Microservices. In Proceedings of the 18th International Conference on Software Technologies - Volume 1: ICSOFT; ISBN 978-989-758-665-1, SciTePress, pages 448-455. DOI: 10.5220/0012093500003538

in Bibtex Style

@conference{icsoft23,
author={Souheir Merkouche and Chafia Bouanaka},
title={TERA-Scaler for a Proactive Auto-Scaling of e-Business Microservices},
booktitle={Proceedings of the 18th International Conference on Software Technologies - Volume 1: ICSOFT},
year={2023},
pages={448-455},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012093500003538},
isbn={978-989-758-665-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 18th International Conference on Software Technologies - Volume 1: ICSOFT
TI - TERA-Scaler for a Proactive Auto-Scaling of e-Business Microservices
SN - 978-989-758-665-1
AU - Merkouche S.
AU - Bouanaka C.
PY - 2023
SP - 448
EP - 455
DO - 10.5220/0012093500003538
PB - SciTePress