The paper uses the technology–organization–environment (TOE) framework to investigate the determinants of cloud adoption intention among firms operating in the upstream oil and gas industry. The issues of security, regulatory uncertainty, and trust deficit continue to delay cloud implementation despite the clear benefits offered by cloud computing technology. Survey data were collected from 190 credentialed information technology professionals and decision makers drawn from the Society of Petroleum Engineers and the LinkedIn Upstream Oil & Gas professional community. Multiple linear regression and factor analysis were employed to test three composite technology–organization–environment predictors against behavioral adoption intention. The final model accounted for 49.8% of variance in adoption intention (F(3, 186) = 63.535, p < .001, Adj. R2 = .498). Notably, environmental trust, fostered by confidence in cloud providers and regulatory clarity, emerged as the most influential factor, encouraging a sense of security and optimism about digital transformation.
Serverless computing has emerged as an efficient paradigm for scalable and cost-effective cloud application deployment through dynamic resource allocation. However, widely used autoscaling mechanisms, particularly CPU-threshold-based approaches such as the Knative Pod Autoscaler (KPA), often exhibit limited adaptability under highly dynamic and bursty workloads, leading to increased latency, SLA violations, scaling instability, and inefficient resource utilization. To address these limitations, this paper proposes a hybrid congestion-aware autoscaling framework for serverless platforms inspired by TCP slow-start and congestion control principles. The proposed method integrates multiple runtime metrics, including workload intensity, CPU utilization, memory usage, request latency, queue congestion, and TCP-inspired congestion window dynamics, into a unified weighted decision model to achieve both responsiveness and stability. The framework is implemented in a Knative-based Kubernetes environment and evaluated using 100,000 requests under 100 concurrent users, with comparisons against Knative KPA and an LSTM-based predictive autoscaling baseline. Experimental results demonstrate that the proposed approach consistently outperforms both baselines across key performance metrics. Specifically, it reduces P50, P90, and P99 response times to 0.29 s, 0.56 s, and 1.12 s, respectively, compared to 0.41 s, 0.88 s, and 2.31 s in KPA. Average response time is reduced to 0.32 s, while throughput increases to 246.80 req/s. In addition, the proposed method improves resource efficiency to 0.87 and decreases total execution time to 405.19 s. Stability is also significantly enhanced, with scaling oscillation rate reduced to 1.5, SLA violation rate reduced to 1.9%, and cold start delay reduced to 0.49 s. Overall, the results confirm that the proposed TCP-inspired hybrid autoscaling approach significantly improves latency performance, scalability, and resource efficiency in serverless environments.
Cloud computing has become a critical technology for modern businesses, offering scalable services and capabilities. Ensuring the security of cloud infrastructure is vital, as organizations rely heavily on these cloud resources. Although traditional Intrusion Detection Systems (IDS) often struggle to effectively detect and mitigate sophisticated security threats in dynamic cloud environments. This comprehensive survey explores the integration of Machine Learning (ML) and IDS techniques to enhance cloud computing security. The paper examines the key challenges in securing cloud environments, including the evolving nature of security threats and the limitations of conventional IDS. The study delves into the various ML-IDS models proposed to address these challenges. It reviews the fundamental concepts, architectures, and techniques employed in these integrated systems, covering a range of ML and IDS approaches. Furthermore, the survey discusses the emerging trends and future research directions in this field, highlighting the need for improved cloud security.
Traditional payroll systems maintain dedicated infrastructure on rigid batch schedules, yielding idle capacity rates of 97–99% between execution cycles. This paper introduces an artificial intelligence (AI)-enhanced, event-driven framework integrating managed container services with machine learning agents for payroll processing. Empirical analysis of 94 enterprise implementations spanning 500 to 50,000 employees demonstrates 71–74% monthly cost reduction, improvement in payroll accuracy from 94.2% to 99.7%, and an 87% reduction in errors relative to traditional architectures. A quantitative decision framework incorporating payroll complexity, regulatory requirements, and AI model selection is presented, alongside comparative analysis of Amazon Web Services Elastic Container Service Fargate, Azure Container Instances, and Google Cloud Run implementations. Four pseudocode algorithms formalize the event dispatch, anomaly detection, tax form extraction, and pipeline orchestration logic. Real-world validation with an 8,500-employee healthcare organization confirms $906 monthly infrastructure cost with zero compliance violations over 120 days.
Multi-cloud adoption is increasingly driven by resilience, regulatory compliance, and vendor independence. Serverless-container platforms such as AWS Fargate, Google Cloud Run, and Azure Container Instances promise portability through similar abstraction models, yet they exhibit differences in configuration semantics and runtime behavior. This study presents a metric-driven evaluation of multi-cloud compatibility using the Cloud Compatibility System, distinguishing between feature-level configuration coverage and empirical execution reliability. Unified container configurations are deployed across providers under controlled conditions to measure execution success, startup latency, and runtime variability. Results show high execution reliability despite limited feature compatibility, revealing a weak correlation between documented configuration support and runtime behavior. These findings highlight the limitations of feature-centric portability assessments and establish an empirically grounded framework for evaluating interoperability in serverless-container environments.
The expanding role of machine learning (ML) and artificial intelligence has become a primary reason behind the demand for high-performance GPUs. Cloud platforms such as Google Cloud Platform, Amazon Web Services, and Microsoft Azure provide scalable access to NVIDIA accelerators. However, variations in GPUs, pricing, usability, and deployment pipelines can be challenging for ML engineers. This paper presents a comprehensive study of GPU platforms across the major cloud providers, covering hardware families (T4, L4, A100, H100, H200, and emerging B200), virtual machine configurations, interconnect technologies, and pricing, including on-demand, spot, and reserved options. Various ecosystem factors were evaluated, including documentation quality, community support, ease of provisioning, and managed ML operations services, such as Vertex AI, SageMaker, and Azure ML. A literature review is provided of MLPerf results and benchmarks to analyze cost, performance, and scalability. The findings provide practical recommendations to guide ML engineers in selecting a suitable cloud for training, inference, and production.
The demand for smart safety solutions has increased, especially for vulnerable groups such as women, children, and the elderly. This study presents a wearable-based intelligent safety and emergency alert system that enables real-time monitoring and rapid response in critical situations. The system combines wearable sensors, a mobile application, and cloud communication to detect distress using signals such as panic button activation, abnormal motion, and irregular vital signs. When a potential emergency is detected, instant alerts are sent to pre-defined contacts and emergency services along with the user’s real-time location. The proposed solution integrates embedded systems, wireless communication, and internet of things technologies to ensure reliability and ease of use. The wearable device is lightweight and discreet, while the mobile application supports both manual and automated alerts. Ethical aspects are also addressed, with a focus on data privacy, secure communication, and user consent. This system aims to improve personal safety and reduce emergency response time.
Failure diagnosis in microservice systems is difficult due to the complex, multimodal nature of telemetry data. Existing methods use metrics, logs, and traces, but rely on message-passing graph neural networks with limited ability to model global context. This study introduces TransTVDiag, which replaces TVDiag’s GraphSAGE encoder with a Graph Transformer enhanced with structural encodings for microservice correlation graphs. The study provides four main contributions: (1) adapting Graphormer to multimodal alert graphs with degree centrality and shortest-path encodings, (2) analyzing these encodings in microservice diagnostics, (3) quantifying the individual and joint impact of metrics, logs, and traces, and (4) demonstrating robustness to missing or noisy alerts. TransTVDiag improves root cause localization in hit ratio by 5.3%, in ranking quality by 3.5%, while reducing inference time by 83.8% over TVDiag. The study also outlines how model outputs can be made more actionable for operators, showing that Graph Transformers offer an accurate and efficient alternative for multimodal failure diagnosis.
Mobile ad hoc networks (MANETs) are dynamic, infrastructure-less systems where nodes self-organize to communicate. They are vital for mission-critical domains—military operations, disaster relief, and rescue—and increasingly support Internet of Things in smart homes, smart environments, and Industry 4.0. Despite flexibility, MANETs face frequent topology changes, packet loss, and quality of service degradation. Software-defined networking offers centralized visibility and control, but current software-defined MANETs (SD-MANETs) suffer scalability limits and single points of failure. The authors implemented a decentralized SD-MANET that distributes control across clustered Ryu controllers, coordinated by Apache ZooKeeper for leader election and Redis for synchronized state. In Mininet-Wi-Fi, the framework delivers higher throughput, better scalability, and improved fault tolerance with lower control overhead than centralized SD-MANETs. Replacing a single controller with a cluster strengthens deployments and provides a resilient backbone for Internet of Things-driven smart environments where high reliability and mobility are essential.
Modern dependence on electrical equipment increases risks of overloading and short circuits, which can cause severe damage and loss of life. Conventional protection devices, such as fuses and miniature circuit breakers (MCBs), have drawbacks. Fuses require replacement, and MCBs rely on slow mechanical tripping, allowing damaging current surges and suffering from wear over time. The authors propose a solid-state MCB with ultra-fast tripping, enhanced reliability, and a longer lifespan. This MCB uses an air core current transformer, adjustable RC (Resistor Capacitor)-based delays, and analog signal processing, thereby ensuring precise fault detection and minimal response time. Comparative tests with digital controls and noise analysis confirmed their efficiency, making it a superior alternative to traditional devices.
The deployment of software-defined networking (SDN) in distributed Kubernetes environments across fog and cloud systems introduces complex security challenges. Traditional approaches often fail to ensure secure, resource-efficient control-plane operations and verifiable node coordination at scale. This study proposes a three-layered security framework: (a) flexible control plan (FCP) integrates lightweight SDN controllers with runtime attestation for trusted execution in resource-constrained fog nodes; (b) secure software-defined offloading (SSDO) enforces encrypted, policy-driven inter-node communication and signature verification to prevent unauthorized coordination; and (c) sentinel-adaptive intrusion detection (SAID) uses an unsupervised deep learning autoencoder to detect anomalies and identify zero-day threats. Combined, these layers offer scalable, adaptive, and real-time security for distributed SDN-Kubernetes environments.
Cloud computing is a transformative technology that has radically changed how companies manage and access their information technology resources. The widespread availability of high-quality cloud computing platforms, operated by companies like Microsoft, Amazon, and Google, has revolutionized the information technology landscape. This paper presents research on traditional virtualization and container technologies, which are fundamental in the shift towards cloud computing. Based on an analysis of the strengths and limitations of both approaches, this study aims to identify the contexts in which each method is most appropriate. It also applies a systematic literature review to analyze existing research published between 2015 and 2024 and discusses current trends and potential future directions. The results are expected to have a significant impact on both professionals and researchers, aiding in decision-making in terms of which technological approaches should be adopted for cloud computing environments, as the findings are based on strong empirical evidence.
Currently, smart mobile devices are used for more than just calling and texting. They can run complex applications such as GPS, antivirus, and photo editor applications. Smart devices today offer mobility, flexibility, and portability, but they have limited resources and a relatively weak battery. As companies began creating mobile resource-hungry and power-hungry applications, they have realized that cloud computing was one of the solutions that they could utilize to overcome smart device constraints. Cloud computing helps decrease memory usage and improve battery life. Mobile cloud computing is the current and expanding research area focusing on methods that allow smart mobile devices to take full advantage of cloud computing. Code offloading is one of the techniques that is employed in cloud computing with mobile devices. This research compares two dynamic offloading frameworks to determine which one is better in terms of execution time and battery life improvement. While executing light tasks Cuckoo does better with local execution while Aiolos outperforms Cuckoo when offloading a light computation task to the cloud. Similarly, Aiolos performs better than Cuckoo when offloading a heavy computation task to an EC2 instance. Regarding battery consumption, offloading using either framework saves 23% more power than the local environment. Aiolos consumes less battery power than Cuckoo when offloading a heavy computation task.
Remote cybersecurity laboratories (remote labs) have become increasingly central to academic and professional training in cybersecurity, offering hands-on experiences and reducing the infrastructure requirements for institutions. This paper provides a comprehensive analysis of the software requirements and challenges associated with implementing remote labs in cybersecurity education. The authors review existing literature; highlight gaps; propose best practices for designing robust, scalable, and secure remote lab environments; and emphasize the significance of addressing pedagogical needs alongside technical constraints. The insights presented aim to guide educators, system designers, and policymakers toward effective adoption and improvement of remote cybersecurity labs.
IoT environments face growing security threats due to their heterogeneity, resource limits, and scale. This paper evaluates an IoT-optimized Xtremely Boosted Network (XBNet) for multi-class attack detection, incorporating protocol-aware normalization, advanced neural architectures, and ensemble strategies. Using the UNB CIC IoT 2023 dataset (33 attacks, 105 devices), the authors conducted hyperparameter, complexity, and error analyses. XBNet achieved 99.5% binary, 94.5% 8-class, and 96.7% 34-class accuracy—outperforming traditional methods with efficient computation. SHAP analysis highlighted protocol-specific features: flow duration (DoS) and packet variance (DDoS). Error analysis showed 68% of DoS/DDoS misclassifications were due to temporal pattern issues. Runtime tests showed feasible deployment from edge to servers, with 42% memory savings via quantization at 98.8% accuracy. The results offer practical insights for real-world IoT security and guide future intrusion detection advances.
Cyberbullying is a fast-growing public-health hazard, demanding reliable, real-time detection of abusive language online. This study presents a unified transformer framework that compares bidirectional encoder representations from transformers, generative pre-trained transformer-2 and text-to-text transfer transformer (T5) on the 90 356-message Mendeley Cyber-Bullying corpus. A shared pipeline normalises text, removes stop-words, and using T5, augments minority classes to curb imbalance. Models are fine-tuned under identical splits (70% train/15% val/15% test, 15 epochs) and scored with accuracy, precision, recall, and F1. Augmented T5 leads with 92.7% accuracy, surpassing generative pre-trained transformer-2 (90.1%) and bidirectional encoder representations from transformers (89.4%). Confusion-matrix analysis shows T5 best balances true- and false-positive rates. Results validate (a) casting cyberbullying detection as sequence-to-sequence; (b) transformer-driven augmentation as an efficient remedy for skewed data; and (c) the feasibility of lightweight, fine-tuned transformers for scalable safety tool.
The adoption of cloud technology is reshaping auditing practices, offering significant advancements for non-Big4 audit firms. This study investigates the integration of cloud computing in audits, focusing on its transformative potential, challenges, and implications for efficiency and quality. The research employs a qualitative approach, using semi-structured interviews with external auditors from non-Big4 firms in Jakarta, registered with the OJK and PPPK of the Ministry of Finance of the Republic of Indonesia. Key findings reveal that cloud technology enhances audit efficiency through real-time data access, streamlined workflows, and improved collaboration, eliminating traditional barriers of time and geography. The integration of AI into cloud systems further optimizes processes, enabling faster data analysis, anomaly detection, and informed decision-making. These advancements significantly reduce operational costs and accelerate audit completion. However, the study also uncovers notable challenges, including data security vulnerabilities, the dependence on client-controlled systems, and difficulties in verifying document authenticity. Specific audit procedures, such as hierarchical reviews and physical site inspections, remain reliant on human judgment, underscoring the indispensable role of auditor expertise. The research highlights the critical role of regulatory compliance, advocating for tailored audit protocols that align with standards such as SA 500 and SA 505. Strong encryption, role-based access controls, and robust collaboration frameworks are identified as essential for mitigating risks and safeguarding audit data. This research contributes to bridging the gap in understanding cloud technology's application in auditing, providing actionable insights for audit firms transitioning into digital environments. It advocates a balanced approach where technological innovations complement human expertise, ensuring audit quality, efficiency, and sustainability. By addressing the challenges of integration and fostering collaboration between auditors and clients, this study addresses the issues of future advancements in audit methodologies, helping non-Big4 firms enhance their competitiveness in an evolving digital landscape.
Currently, smart mobile devices are used for more than just calling and texting. They can run complex applications such as GPS, antivirus, and photo editor applications. Smart devices today offer mobility, flexibility, and portability, but they have limited resources and a relatively weak battery. As companies began creating mobile resource-hungry and power-hungry applications, they have realized that cloud computing was one of the solutions that they could utilize to overcome smart device constraints. Cloud computing helps decrease memory usage and improve battery life. Mobile cloud computing is the current and expanding research area focusing on methods that allow smart mobile devices to take full advantage of cloud computing. Code offloading is one of the techniques that is employed in cloud computing with mobile devices. This research compares two dynamic offloading frameworks to determine which one is better in terms of execution time and battery life improvement. While executing light tasks Cuckoo does better with local execution while Aiolos outperforms Cuckoo when offloading a light computation task to the cloud. Similarly, Aiolos performs better than Cuckoo when offloading a heavy computation task to an EC2 instance. Regarding battery consumption, offloading using either framework saves 23% more power than the local environment. Aiolos consumes less battery power than Cuckoo when offloading a heavy computation task.
