Power Trading Framework of Cloud-Edge Computing in the Artificial Intelligence Market
Gopal Krishna, Vineeta Singh, Deepak Kumar Pandey, Konda Hari Krishna, Kapil Kumar Joshi, Tushar Gupta · 2024
Cloud-edge computing and artificial intelligence proliferation make it necessary to learn how to properly manage energy in the environment. This paper will aim to design and assess a power trading framework on the possibility of adequately optimizing power allocation for cloud-edge AI systems. The problem that the paper seeks to address is the critical problem of energy acumen in cloud-edge AI. With the growing number of edge devices and the unpredictability of AI workload, power allocation must be effectively done. Therefore, the paper’s scope will include designing a power trading system that will effectively and dynamically allocate power resources in the AI to optimize and secure efficiency, effectiveness, reduce costs, and completely eliminate or reduce latencies. The research approach is wide and thorough. I conducted a collected theoretical framework, developed algorithms, acquired actual and credible real-world data, and analyzed big data. During this process, I conducted real-world simulations and deployed the framework in different case studies involving various cloud-edge AI environments. The results showed that power trading framework sufficiently optimizes power allocation, which has increased pyre efficacy, relatively reduced costs, and low latency. Implementing the framework will also secure sustainability ambitions, especially in reducing carbon emissions and depending less on non-renewable energy sources. The output of the study is more recommendable compared to traditional power management systems. This study has a promising starting point for power trading. However, some limitations proved to be challenging, such as some technical difficulties and problems in collecting and submitting data. Some future possibilities include creating new devices through collaboration or new novel applications. The framework can be applied to cloud-edge AI, and business and industries can apply it to their operations on how to manage energy in a rapidly dynamic and energy-intensive environment.