Design and implementation of large-scale network propagation simulation method inspired by Pregel mechanism

With the rapid development of the Internet and online social media, the law of information dissemination in social networks needs much experimentation on network propagation calculation. The current network propagation experiments based on the SIR model are widely used in disease research and information dissemination. However, because of the limitations of hardware and software, it is still difficult to conduct ultra-large-scale network propagation calculation. However, the current Internet information dissemination shows the characteristics of large-scale users, large amounts of information, and fast propagation. The shortcomings of small-scale network dissemination experiments based on abstract and simplified methods have been revealed. In this study, the Spark platform is used to implement an experimental algorithm for large-scale network propagation calculation. The performances of the algorithm and Nepidemix stand-alone computing components are compared, and the algorithms advantages and disadvantages are demonstrated. An orthogonal experimental design method was used to design the performance test experiment to find out the factors influencing the algorithm. When there are enough cluster computing resources, the algorithm can break the limitation of network node size and is difficult to develop, which lays the foundation for calculation experiments about very large-scale network propagation.