Development of Artificial Intelligence Supported Tool for Anomaly Detection in Cloud Computing Systems

In the rapidly evolving landscape of Internet of Things (IoT) and Cloud systems, the complexity and scale of these technologies demand innovative solutions for efficient error detection and management. This paper proposes a robust framework for anomaly detection in cloud computing systems, emphasizing the critical role of automation and artificial intelligence (AI). The approach integrates Continuous Integration, Continuous Deployment (CICD) principles with AI-driven anomaly detection techniques to enhance system reliability and minimize risks. The study introduces a novel method employing one-dimensional convolutional layers and Autoencoder architecture to preprocess and analyze system metrics, enabling the development of an adaptive and customized anomaly detection system. The proposed framework is implemented through a Python-based agent, utilizing specific libraries and Continuous Integration/Continuous Deployment (CI/CD) processes executed via Jenkins. The agent is tested and deployed on Kubernetes, demonstrating its effectiveness in dynamic and complex environments. The results highlight the significance of iterative optimization, emphasizing continuous monitoring and adaptation for ensuring the robustness and reliability of cloud computing systems. This research contributes valuable insights into enhancing the efficiency of anomaly detection in large-scale cloud systems, paving the way for further advancements in the field of IoT and cloud technologies.