The existing steady state analysis of power system application architecture has poor scalability and cannot adapt to the large-scale application requirement across provincial and regional dispatching centers. To solve this problem, a cross regional parallel power system steady state analysis service elastic scheduling management system based on cloud computing technology is proposed. The key technologies of the system, such as overall architecture, container packaging, resource isolation, service elastic scheduling, operation monitoring and so on are introduced and analyzed in detail. Micro service technology is used to modularize analysis services. The container technology is used to realize the deployment, isolation and dynamic expansion of analysis services. An algorithm based on the stochastic high-level Petri net model is used to realize the elastic scheduling of analysis services. A variety of practical steady-state analysis service applications proves that the overall operation of the system is stable, the system architecture has good scalability, and the application requirement of the wide range steady-state analysis and calculation across provincial and regional dispatching centers can be met. This work is supported by the National Key Research and Development Program of China (No. 2017YFB0902600).