Publication

Abstract : The demand for effective, scalable, and adaptable resource provisioning in IoT-Cloud systems has increased due to the Internet of Things' (IoT) explosive growth. Traditional static or rule-based provisioning techniques are unable to adequately handle the varied resource requirements, stringent latency limits, and extremely dynamic workloads of contemporary IoT applications. The Autonomic Resource Provisioning Framework that this article suggests as a solution to the problems of dynamic workload variation, latency-sensitive decision-making, and effective resource allocation in IoT-Cloud environments is based on the MAPE-K (Monitor-Analyze-Plan-Execute over Knowledge) paradigm and the IoT-Cloud Resource Optimization Algorithm (ICROA). The system uses knowledge-driven reasoning to forecast resource needs, distributes cloud and edge resources independently, and continuously assesses IoT workloads to guarantee peak performance in order to address these issues. The suggested system integrates adaptive optimization, MCDM-based ranking, and fuzzy logic decision-making to boost responsiveness, lower operating costs, and guarantee service dependability. When compared to traditional methods, the MAPE-K-driven ICROA framework dramatically increases provisioning efficiency, as shown by experimental research and theoretical analysis. As a result, it is appropriate for extensive, real-time IoT-Cloud deployments.