Computing Model and Adaptation Service of State Grid Cloud Open Source Components for Cloud Big Data

When processing large-scale and diverse power data, State Grid Cloud faces challenges such as efficient use of computing resources and real-time data processing. In order to build a cloud platform with strong adaptability and good scalability, it is necessary to design a reasonable computing model and provide flexible adaptation services to meet the needs of different power business systems. This paper adopts a cloud platform tool chain and operation and maintenance method based on open source components, and builds an open source component computing model for State Grid Cloud by integrating open source technologies such as Kubernetes and Docker. The paper introduces open source technologies such as Kubernetes and Docker to build the underlying infrastructure of the State Grid Cloud. It integrates open source cloud management platforms such as OpenStack to achieve unified management and scheduling of computing, storage, network and other resources. Designing a computing model based on open source components to support the dynamic allocation and release of elastic computing resources. Using containerization technology to deploy applications of the power business system in containers to achieve rapid deployment and migration of applications. Designing the adaptation service interface and protocol according to the needs of the power business system. Using the cloud platform operation and maintenance tool chain to realize the automated deployment and iteration of the adaptation service. Through the configuration and management of cloud platform operation and maintenance tasks, customized services for power business systems can be achieved. The experimental group that adopted the State Grid Cloud open source component computing model and adaptation service outperformed the control group deployed with traditional physical servers in terms of memory utilization, response time and mean fault repair time. The memory utilization rate of the experimental group mostly exceeded 70%, the response time remained between 301 milliseconds and 632 milliseconds, and the average fault repair time was significantly lower than that of the control group, showing higher efficiency and faster response speed. Based on the above, the research results of this paper promote the digital transformation and intelligent upgrading of the power industry.