Research on Architecture and Real-Time Performance of a Safety Cloud Platform Based on Commodity Servers

As the contradiction between the demand for high speed, high efficiency, high safety of rail transit and the lack of computing power, limited equipment space and high equipment cost in the existing train control system of rail transit becomes increasingly fierce. At the same time, cloud computing technology stands out. It has the characteristics of high resource utilization, large scale, strong scalability, high availability and on-demand service. The migration of train control systems to cloud computing can effectively improve the resource utilization, reduce the number of equipment and system energy consumption. In this paper, based on the commodity server hardware and the advanced virtualization technology, an architecture of a safety cloud platform was proposed. This safety cloud platform can deploy SIL4 safety-critical applications and enhance the rail transit intelligence. However, the introduction of virtualization technology may also bring some problems about system real-time performance. To address this issue, a set of optimization strategies are proposed to enhance the real-time performance of the safety cloud platform. To validate the effectiveness of optimization measures, real-time performance testing was conducted. Experimental results demonstrate substantial enhancements in CPU throughput, memory efficiency, and network latency after optimization. This study aims to provide theoretical support and practical references for the development of safety cloud platforms based on commodity servers.