Security-Enhanced QoS-Aware Autoscaling of Kubernetes Pods Using Horizontal Pod Autoscaler (HPA)
Journal Title: Journal of Intelligent Management Decision - Year 2024, Vol 3, Issue 3
Abstract
Container-based virtualization has emerged as a leading alternative to traditional cloud-based architectures due to its lower overhead, enhanced scalability, and adaptability. Kubernetes, one of the most widely adopted open-source container orchestration platforms, facilitates dynamic resource allocation through the Horizontal Pod Autoscaler (HPA). This auto-scaling mechanism enables efficient deployment and management of microservices, allowing for rapid development of complex SaaS applications. However, recent studies have identified several vulnerabilities in auto-scaling systems, including brute force attacks, Denial-of-Service (DoS) attacks, and YOYO attacks, which have led to significant performance degradation and unexpected downtimes. In response to these challenges, a novel approach is proposed to ensure uninterrupted deployment and enhanced resilience against such attacks. By leveraging Helm for deployment automation, Prometheus for metrics collection, and Grafana for real-time monitoring and visualisation, this framework improves the Quality of Service (QoS) in Kubernetes clusters. A primary focus is placed on achieving optimal resource utilisation while meeting Service Level Objectives (SLOs). The proposed architecture dynamically scales workloads in response to fluctuating demands and strengthens security against autoscaling-specific attacks. An on-premises implementation using Kubernetes and Docker containers demonstrates the feasibility of this approach by mitigating performance bottlenecks and preventing downtime. The contribution of this research lies in the ability to enhance system robustness and maintain service reliability under malicious conditions without compromising resource efficiency. This methodology ensures seamless scalability and secure operations, making it suitable for enterprise-level microservices and cloud-native applications.
Authors and Affiliations
"Vani Rajasekar, Muzafer Saračević, Darjan Karabašević, Dragiša Stanujkić, Amor Hasić, Melisa Azizović, Srivarshan Thirumalai"
Keywords
Related Articles
- EP ID EP752390
- DOI https://doi.org/10.56578/jimd030304
- Views 35
- Downloads 0
Gamification in the Workplace: Enhancing Employee Engagement Through Gameful Experiences
In an era characterized by intense labor market competition for skilled and motivated personnel, the adoption of innovative strategies, such as gamification, has emerged as a critical factor for cultivating an engaging w...
Optimization of Production Scheduling Through a Multi-Objective Constrained Greedy Model
The traditional manufacturing sector in China is increasingly challenged by rising labour costs and the diminishing demographic advantage. These issues exacerbate existing inefficiencies, such as limited value addition,...
Evaluating the Logistics Performance of G8 Nations Using Multi-Criteria Decision-Making Models
Logistics performance plays a pivotal role in fostering economic growth and enhancing global competitiveness. This study aims to evaluate the logistics performance of G8 nations through multi-criteria decision-making (MC...
Integrated Multi-objective Optimization of Predictive Maintenance and Production Scheduling: Perspective from Lead Time Constraints
For the integrated optimization of job-shop production scheduling and predictive maintenance, this paper fully considers such constraints as product delivery time and changing machine failure rate, and establishes a mult...
Optimising the Efficiency of Municipal Utility Vehicle Fleets Using DEA-CRITIC-MARCOS: A Sustainable Waste Management Approach
The efficiency of utility vehicle fleets in municipal waste management plays a crucial role in enhancing the sustainability and effectiveness of non-hazardous waste disposal systems. This research investigates the operat...