OPTIMIZING CLOUD COMPUTING PERFORMANCE USING EDGE AI: A HYBRID APPROACH

Abstract

Background: Data: cloud, web, character string, encoded, decoupled, run, scale, storage, availability, 2030, 2050, 975, describe, encrypt, optimize, prep, in-context, concurrency, level, accessibility. Traditionally designed cloud structures are not efficient for the on-demand real-time processing most applications require, particularly for low latency and high efficiency. Local data processing enabled by edge AI became the solution, but edge only systems are limited due to computational constraints. This can be achieved with a hybrid cloud-edge AI based approach that dynamically distributes the tasks between cloud servers and edge devices using AI to help facilitate intelligent workload management. Objective: The effect of hybrid cloud-edge AI model in improving performance of cloud computing is presented in this study. A hybrid approach involves the integration of cloud and edge computing, which addresses the inefficiencies of cloud and edge computing working in isolation, by honing in on the most effective balance in workload between the two architectures, and maximizing resource usage. Method: This study analyzes efficiency, security and resource utilization improvements through a mix of cloud-based simulations and real-world edge computing tests. Federated learning enables decentralized training of AI models on edge devices, while load balancing and edge inference techniques are employed to reduce the load on a centralized cloud server. Latency reduction, use of limited bandwidth, computational cost, and effective encryption are all performance metrics. Result: The study shows how a hybrid model is able to help you obviate latency challenges, speed up processes and manage your bandwidth far better than simple cloud and edge-only models. Data integrity and privacy are ensured through AI driven security measures, and scalable task distribution allows seamless integration across multiple devices. Conclusion: These results highlight the potential of hybrid cloud-edge AI systems for real-time processing in a range of applications from healthcare and IoT, to smart cities and industrial automation.Simply speaking, hybrid cloud-edge AI is able to enhance the quality of end-edge cloud processing for the real-time use cases. Moreover, some future studies may focus on dealing with the blockchain-based security mechanisms as well as advanced federated learning methods to improve optimization and reliability further.