Software aging issues and rejuvenation strategies for a container orchestration system

Cloud-native computing is a software development approach that uses cloud computing to build and run scalable applications by leveraging best practices and technologies in the field such as DevOps, general unified architectures (e.g., based on serverless functions, containers, container orchestrators, sidecar, service meshes, microservices), agile software development and so on. It is also used for application development environments that enable an abstraction layer on top of non-cloud provider services. Cloud-native is expected to increase the efficiency of development, operation, and verification of 5G and beyond network functionalities along with automated service upgrades and deployments, and ultimately reduce vendor lock-in so that applications can be easily migrated to other environments seemlessly. In this paper, we provide a state-of-the-art overview of existing cloud-native approaches for telcos, their key features, the role of open-source and Linux Foundation projects, and all associated standardization efforts. After presenting key findings of the recent research studies and related implications under “Lessons Learned”, we highlight the potential challenges, current issues, and future directions of cloud-native computing for telecommunication networks.

ABSTRACT The dynamic nature of cloud‐native applications necessitates robust resource elasticity to meet performance objectives while optimizing costs. However, conventional container orchestration systems often rely on reactive, static‐threshold mechanisms and centralized control architectures, which lead to inefficient resource utilization and scalability bottlenecks. This paper presents EffiScale, a novel, self‐adaptive framework for orchestrating container elasticity in cloud environments. EffiScale introduces a decentralized microservice architecture, extending the MAPE‐K model, that integrates four core innovations: (1) a decentralized control plane governed by a Byzantine fault‐tolerant consensus protocol, eliminating single points of failure and ensuring resilient coordination; (2) a hybrid scaling orchestration engine that frames the choice between vertical and horizontal scaling as a multi‐objective optimization problem, enabling granular and cost‐effective resource allocation; (3) a predictive, risk‐aware thresholding mechanism that leverages an ensemble of machine learning models to forecast workloads and dynamically adjusts scaling triggers based on both the prediction and its associated uncertainty; and (4) a federated knowledge base for continuous, collaborative refinement of scaling policies across distributed controllers. Experimental evaluation on a multi‐node cluster with a stateless Nginx web application demonstrates that EffiScale can reduce response times, increase throughput, and lower resource usage compared to state‐of‐the‐art solutions, highlighting its effectiveness in managing dynamic containerized workloads.

Modern particle accelerator facilities require sophisticated control systems capable of managing thousands of process variables in real-time while ensuring high availability, scalability, and ease of maintenance. This paper presents EPIK8S (EPICS on Kubernetes), a framework that leverages Kubernetes container orchestration to deploy, manage, and scale EPICS (Experimental Physics and Industrial Control System) infrastructure for accelerator control systems. A central design principle is that the entire control system configuration for a beamline — comprising IOCs, services, and infrastructure — is captured in a single YAML file, which is processed through a Jinja2 templating layer to produce device-specific IBEK runtime configurations. This three-tier architecture radically simplifies IOC management, enables complete change tracking via Git, and eliminates an entire class of manual configuration errors. The framework introduces a GitOps-based approach using ArgoCD for continuous deployment, providing declarative configuration management and automated synchronisation. We describe the architecture and implementation, and report operational experience from three INFN facilities: the SPARC_LAB photoinjector, the DAFNE Beam Test Facility (BTF), and the ELI-NP gamma beam system in Romania. Performance metrics — including dedicated Channel Access round-trip latency measurements comparing bare-metal, pod-to-pod, and external-to-cluster scenarios — and lessons learned from over two years of production operation demonstrate significant improvements in deployment efficiency, maintainability, and reliability compared to traditional bare-metal EPICS deployments, with containerisation overhead well within acceptable bounds for accelerator control applications.

Aim: The primary aim of this research is to design a unified architectural framework that seamlessly integrates microservices with DevOps practices, continuous integration and continuous delivery pipelines, and runtime orchestration mechanisms. The proposed framework addresses the growing complexity of managing distributed microservice systems in dynamic cloud-native environments by improving deployment consistency, operational efficiency, and system scalability while significantly reducing manual intervention across the software delivery lifecycle. Emphasis is placed on reliability, automation, and continuous feedback to enable faster, safer, and more resilient software delivery. Method: This study adopts a system-oriented architectural design approach that combines conceptual modeling with workflow abstraction. Existing microservices architectures, DevOps workflows, and orchestration platforms are systematically analyzed to identify integration gaps and operational inefficiencies. Based on this analysis, a layered architectural framework is proposed to unify development, deployment, and runtime management. The framework incorporates standard DevOps tools, CI/CD automation strategies, and container orchestration principles, with logical workflows validated through representative architectural scenarios. The methodological focus remains on modularity, interoperability, and end-to-end automation. Results: The results demonstrate that the proposed framework improves deployment frequency and reduces system downtime through tight integration of CI/CD pipelines with orchestration engines, enabling automated scaling and rollback mechanisms. Centralized monitoring and logging enhance operational observability, while the unified architecture supports continuous delivery with minimal configuration overhead. Performance analysis further indicates faster system recovery and improved fault isolation, confirming the effectiveness of the unified architectural approach. Conclusion: In conclusion, this research establishes that integrating DevOps practices, CI/CD pipelines, and runtime orchestration within a unified microservices architecture significantly enhances system manageability and reliability. The framework reduces operational complexity while ensuring deployment consistency and rapid feedback across the software lifecycle. Its adaptability to diverse cloud-native platforms positions it as a strong foundation for future intelligent, autonomous, and self-healing software systems.

Architecturally, the AI Runtime Infrastructure, or AIRI, is a foundational layer of distributed architecture designed to enable the execution of large-scale AI workloads. Most modern distributed architectures, heavily influenced by cloud-native design principles, are designed for stateless, deterministic, synchronous, and microservices-based workloads. As such, they are not designed to manage efficiently the stateful, probabilistic, and adaptive workloads that AI execution entails. AIRI is proposed as a runtime layer and reference architecture providing application-agnostic support across compute, storage, and networking infrastructure. It supports core runtime responsibilities such as model lifecycle management, orchestration of heterogeneous accelerators, cross-model coordination, and inference-time policy enforcement. In addition, the architecture includes control-plane capabilities such as model-aware routing, which aid efficiency and governance, as well as data-plane capabilities including feature servers, embedding infrastructure, and vector search. Engineering challenges include multi-model coherence, runtime safety, model-aware scheduling, dynamic batching, and fairness scheduling in multi-tenant environments. As with virtualization and container orchestration in previous generations of computing, AIRI establishes AI workloads as first-class distributed system workloads that require a dedicated runtime and layered abstractions for optimal performance. It eases the scalable, reliable, and efficient deployment of generative models, multimodal systems, and agentic architectures in diverse cloud-native environments. This paper presents a layered architectural model for AIRI, identifies key engineering challenges, and discusses implications for future distributed systems infrastructure.

Container orchestration platforms such as Kubernetes are increasingly deployed in cloud-native and edge computing environments, where ensuring secure network isolation and trustworthy interactions between distributed components remains a critical challenge. Lightweight Kubernetes clusters, in particular, often lack robust mechanisms for decentralized trust management and tamper-resistant security auditing. This paper proposes a secure network-isolation and trust model for Kubernetes environments based on the Multichain blockchain. The study aims to enhance security assurance and audit transparency by introducing a decentralized trust layer that complements native Kubernetes networking mechanisms. The research objectives include analyzing network-level security threats in Kubernetes clusters, designing a blockchain-based trust-and-audit architecture, integrating Multichain with Kubernetes networking components, and evaluating the effectiveness of the proposed model using quantitative metrics. The proposed approach is validated through experimental deployment in a controlled Kubernetes environment. Effectiveness is assessed using normalized indicators for security, performance, reliability, and integration, combined into an integrated effectiveness index. Radar-based visualization is employed to compare the proposed solution with a baseline Kubernetes configuration without blockchain support. Experimental results demonstrate that the proposed model significantly improves security and reliability metrics while maintaining acceptable performance overhead. The integrated effectiveness index confirms a measurable overall improvement compared to traditional Kubernetes deployments. The scientific contribution of this work lies in integrating decentralized, blockchain-based trust and immutable audit logging with Kubernetes network isolation mechanisms. The proposed model provides a practical, scalable approach to enhancing the security of Kubernetes clusters across cloud-native and edge computing infrastructures.

Electric vehicle fleet operations face significant connectivity challenges across heterogeneous network environments, requiring innovative solutions for seamless communication maintenance. This article presents a comprehensive connectivity architecture comprising Multi-Network Adaptive Routing (MNAR), Context-Aware Data Prioritization (CADP), and Predictive Connectivity Health Modeling (PCHM) frameworks designed to address critical gaps in current fleet communication systems. The MNAR framework enables dynamic network discovery and intelligent switching across cellular, Wi-Fi, satellite, and mesh technologies through real-time performance optimization algorithms. CADP implements hierarchical data classification and priority assignment mechanisms that optimize bandwidth utilization based on operational context and mission criticality. PCHM employs machine learning algorithms for connectivity degradation prediction and proactive network management strategies. The integrated security architecture addresses multi-network vulnerabilities through end-to-end encryption, zero-trust authentication, and comprehensive intrusion detection systems. Performance evaluation demonstrates substantial improvements in connectivity uptime, latency reduction, bandwidth efficiency, and cost optimization compared to traditional single-network approaches. The proposed architecture exhibits superior scalability across diverse fleet deployments while maintaining consistent performance characteristics. Implementation frameworks include microservices-based deployment, container orchestration, and blockchain integration for secure fleet-to-fleet communication. Validation through simulation environments and real-world pilot implementations confirms system effectiveness across urban, rural, and challenging operational scenarios. Future development opportunities encompass 5G network slicing integration, quantum-resistant encryption protocols, and AI-driven autonomous optimization capabilities with cross-industry applicability to maritime and aerospace domains.

Enhancing fog IoT container deployment: A customizable Kubernetes scheduler

Abstract - Cloud-based applications increasingly rely on multiple database systems to handle diverse data models and workloads, yet managing these heterogeneous environments remains complex and resource-intensive. Traditional Database-as-a-Service platforms often introduce vendor lock-in, limited flexibility, and high costs, restricting their suitability for academic and research use. To address these challenges, this research proposes an open-source, AI-powered Cloud Database-as-a-Service platform that unifies the management of SQL, NoSQL, and in-memory databases using Kubernetes-based container orchestration. The system integrates AI-driven natural language assistance for schema generation and query formulation, along with real-time monitoring using Prometheus and Grafana. By combining automation, intelligent interaction, and cost-effective deployment, the platform aims to improve accessibility, efficiency, and scalability in cloud-native database management. Key Words: Database-as-a-Service (DBaaS); Cloud Computing; Kubernetes Orchestration; Container Management; Artificial Intelligence (AI); Natural Language Processing (NLP); Multi-Database Systems; Microservices Architecture; Performance Monitoring; Prometheus; Grafana; Real-Time Analytics; Open-Source Platform

Kubernetes has emerged as the leading platform for container orchestration, delivering unprecedented flexibility and scalability for cloud-native applications. However, its dynamic, distributed nature introduces significant complexity in maintaining operational visibility and reliability. Effective monitoring is no longer luxury, rather it is essential for detecting failures, optimizing performance, and securing workloads. This paper presents a comprehensive review of best practices for monitoring Kubernetes clusters, synthesizing recommendations from industry leaders, open-source communities, and recent academic research. By adopting these practices, organizations can build resilient, observable, and manageable Kubernetes environments.

The convergence of artificial intelligence, fraud detection, and multi-cloud infrastructure presents unique challenges at the intersection of software engineering, cybersecurity, and distributed systems. This review examines the current state of research on optimizing software engineering pipelines for deploying AI-based fraud detection systems across multi-cloud environments. We synthesize findings from contemporary literature, analyzing architectural patterns, security frameworks, deployment strategies, and performance optimization techniques. The review addresses three critical research questions concerning architectural patterns and pipeline optimization strategies for multi-cloud deployments, security requirements influencing pipeline design, and current limitations with future research directions. Key findings indicate that microservices-based architectures leveraging container orchestration, event-driven processing, and hierarchical feature stores enable effective multi-cloud deployment. However, significant complexities persist in model versioning, data governance, cross-cloud data transfer costs, and security orchestration. We identify critical gaps in standardized pipeline architectures and propose a research agenda focusing on AI-native infrastructure, confidential computing, automated optimization, and sustainable ML practices. Case studies from financial services and e-commerce sectors illustrate practical implementations, while identified challenges and future directions provide a roadmap for advancing this critical domain.

The shift from monolithic systems to cloud-native architectures has become a critical priority for modern enterprises seeking greater scalability, agility, and operational efficiency. Cloud-native principles emphasize re-architecting legacy applications to align them with contemporary infrastructure standards, incorporating container orchestration, microservice-based designs, and modern deployment practices. The paper provides a thorough comparative analysis of two prominent platforms, Microsoft Azure and Red Hat OpenShift, and how they are effective in enabling cloud-native refactoring programs. It interrogates their architectural designs, service platform, DevOps implementation, security, cost platform, and scalability. Azure has the advantages of managed services and extensive integration with the Microsoft ecosystem, whereas open-source OpenStack has more control, flexibility of deployment to be hybrid, and is well-aligned with open-source technologies. The results are summed up with strategic recommendations on which platform organizations should choose in order to meet the needs of cloud-native modernization and address their long-term objectives.

In this article, we present a timing analysis of orchestration times for containerized services, revealing the inability of current container orchestrators to fully prioritize services under concurrent requests. The analysis identifies the sources of orchestration delays that impact services to be prioritized potentially violating their Service Level Objectives (SLOs). Based on the findings of the timing analysis, we highlight three alternative SLO-aware orchestration system designs aimed at preventing and/or mitigating delays for high-priority services. We provide principles and guidelines that must drive the implementation of these designs. We then introduce Ulysses , a Kubernetes -based prototype embodying the simplest of the three designs. Ulysses modifies the core Kubernetes control plane components to manage events synchronously and with fixed priority. Through experiments conducted with both synthetic workloads and a containerized cloud-native 5G core network, we demonstrate that Ulysses ensures stable orchestration times for high-priority services, with a reduction of up to 78% under high orchestration load.

The challenges of performance, scalability, and reliability of modern enterprise integration platforms have never been higher due to the rapid uptake of cloud-native infrastructures and microservices-based applications by organizations. The optimization of infrastructure has become a highly important field of study with direct implications on system throughput, the use of resources, and the cost of operation. The paper looks at the basic techniques and the best practices that can be used to optimize infrastructure within enterprise integration environments, and this includes architectural patterns, deployment strategies, resource management approaches, and continuous improvement methodologies. The API-led connectivity is a paradigm shift from the traditional point-to-point integration methods, where assets of integration are structured to form three different layers that isolate concerns and allow their independent optimization. Container orchestration platforms transformed the management of the integration workloads through the automation of the deployment, scaling, and recovery processes. Directly dependent on efficient data transformation and processing methods are the computational intensity and resource consumption of integration workloads. Transformations that stream data as it arrives instead of loading an entire set of data can save a large amount of memory. Extensive monitoring and performance analysis are the basis of determining optimization opportunities and ensuring the efficiency of infrastructure changes. Companies that use overall optimization plans are always able to make substantial improvements in throughput with lower infrastructure costs, which provides substantial value to the business as a whole in terms of the capability of the system and the overall cost-effectiveness.

Event-based automation systems are a revolutionary paradigm in the healthcare information technology activities that respond to the critical issue of operational resiliency in the growing and increasingly complex digital health ecosystems. The SupportPlus model proves how intelligent automation may radically rethink the validation process, moving away from a manual process of reactivity towards validation to a system-driven mechanism of assurance. Using microservices architecture, serverless computing, and container orchestration on the Azure Kubernetes service, the framework integrates validation, monitoring, and compliance functions in a metadata-based ecosystem. Application to mission-critical healthcare services also demonstrates significant operational benefits such as radical decreases in manual validation cost, faster incident response settings, and system availability that is approaching five-nines reliability. The architecture uses event-driven orchestration to instigate automated validation gates built into continuous delivery pipelines, modifying compliance with periodic manual audits into continuous compliance processes. The self-healing functionality is defined as the capability of the system to automatically recover most of the validation failures, and this aspect is fundamentally changing the dynamic between the operations of IT and the system reliability. In addition to the short-term efficiency, the framework harmonizes the validation practices throughout organizational teams, eradicates tribal knowledge dependence whatsoever, and develops ongoing compliance evidence that can be examined during a regulatory audit. The measurable effect proves that intelligent healthcare IT automation is not only cost optimization but also meets the main demands of patient safety, care continuity, and trust of people in digital health infrastructure.

The financial services and insurance industries are increasingly confronted with challenges relating to adapting compliance within dynamic digital environments, wherein legacy brick-and-mortar architectures become inefficient for handling large volumes of transaction and claims data. Cloud-native architectures bring about paradigm shifts with microservices based on event processing and intelligent orchestration, aiming at enabling compliance monitoring at a real-time scale while ensuring integrity within audit trails. The inclusion of ethical artificial intelligence brings about adaptive learning and pattern recognition capabilities within autonomous decision-making, but it poses serious questions with regard to fairness and bias, particularly within vulnerable sections. Contemporary compliance systems seek to integrate efficiency and societal responsibilities within compliance and fairness constraints, explainability, and human review processes. Multi-level caching architectures and computational optimization improve system performance with accuracy. Container orchestration solutions bring about automation with sophisticated resource allocation strategy support within high-priority fraud and batch reporting processes. The road ahead focuses on federated learning methods enabling collaborative fraud analysis without compromising customer data privacy, making cloud-native architectures an infrastructural component within trustworthy and ethical compliance system developments.

Container orchestration platforms have fundamentally transformed enterprise infrastructure managementin regulated sectors through sophisticated security and compliance frameworks. This article examines comprehensive strategies for deploying secure container platforms

Zero-Trust Security Architecture for Containerized Microservices in Enterprise Telecommunications Networks

In an information technology landscape increasingly dominated by high-level abstraction layers, configuration management frameworks, and sophisticated Infrastructure as Code (IaC) platforms, the role of foundational shell scripting often faces scrutiny regarding its continued relevance. This paper presents an analytical examination of shell scripting's evolving but persistent significance in both traditional system administration and modern DevOps methodologies. The central thesis posits that shell scripting, far from being obsolete, functions as an indispensable, ubiquitous "glue logic" that connects disparate systems, facilitates rapid automation of granular tasks, and provides essential customization capabilities that high-level tools cannot efficiently address. We explore the functional advantages of shell scripting—including its universality across POSIX-compliant systems, low resource overhead, and unparalleled access to core operating system utilities—in automating routine system maintenance, diagnostics, and monitoring. Furthermore, the analysis extends to the DevOps lifecycle, demonstrating how shell scripts underpin crucial processes within Continuous Integration/Continuous Delivery (CI/CD) pipelines, container orchestration (Docker/Kubernetes), and environment bootstrapping. By comparing the capabilities of shell scripting against modern alternatives like Python and dedicated IaC tools (e.g., Ansible, Terraform), this paper identifies the specific niches where shell scripting excels and analyses the threshold at which task complexity necessitates migration to more robust programming languages. The research concludes that proficiency in shell scripting remains a critical skill for IT professionals, enhancing operational agility, reducing human error, and maximizing efficiency by bridging the gap between legacy infrastructure and contemporary cloud-native architectures.