Zhou, Pan (Beijing Jiaotong University), Wang, Xi (National Engineering Research Center of Rail Transportation Oper), Jin, Jun (China State Railway Group Co., Ltd.), Wang, Hongwei (Beijing Jiaotong University), Ying, ZhiPeng (China Academy of Railway Sciences), Fei, Zhenhao (CASCO Signal Ltd.), Wang, Lijun (China Railway Signal & Communication Research & Design Institute)

A Cloud Resource Allocation Method for Railway Safety Critical Computing Application

Scheduled for presentation during the Poster Session "Railway systems and applications" (ThBT13), Thursday, September 26, 2024, 14:30−16:30, Foyer

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on October 8, 2024

Abstract

As an important part of safety computer platform, railway safety critical computing application plays an important role in ensuring the safety of urban rail transit. With the construction of urban rail transit and the development of cloud computing technology, in order to reduce the number of physical equipment and maintenance costs, cloud computing is gradually introduced into rail transit signal system. However, the safety computer architecture must be considered when the railway safety critical computing applications are transplanted to the cloud server, and it is difficult to ensure the performance and power consumption after migration under the premise of meeting the safety requirements. This paper focuses on the migration of railway safety critical computing application from traditional hardware platforms to cloud servers. The reliability and performance of these systems are key to ensuring overall railway safety and efficiency. This paper proposes a virtual machine placement strategy and resource allocation strategy to balance application performance and reduce energy waste. A mixed integer linear programming model is used to consider dynamic changes in virtual machine resource requirements, taking into account the safety constraints of safety computer platforms and load balancing between hosts. According to the simulation results, the virtual machine placement strategy obtained by the model can meet the constraints of the safety constraints and reduce power consumption at the same time.