| （英） |
In IoT, content is created by combining data from various different fields generated in the region, expanding local production for local consumption of data to be utilized in the region. However, the amount of computation (computational resources) and the amount of transferred data (communication resources) required for content generation vary widely, and the efficiency of resource use decreases depending on the data transfer path and the server that processes the data. In a previous study, we proposed a resource utilization method that divides the content generation process into multiple processes and distributes data preprocessing among multiple servers, but the specific method to realize preprocessing and its effectiveness have not yet been evaluated. In this study, we propose a method for dividing preprocessing and a method for realizing the divided preprocessing in an edge network, and evaluate the effectiveness of the proposed method. Experimental results show that dividing preprocessing can reduce data and the processing load itself, and is effective in efficient use of computational resources. Furthermore, by distributing the computational load by executing the divided preprocessing on servers in the edge network, we can effectively utilize all resources in the edge network and expect to reduce the processing time by 4.7%. |