A profit-aware server deployment approach for edge computing

A profit-aware server deployment approach for edge computing

연구 분야: Networking

학회: Computing

In recent years, the demand for computing resources in daily life has risen due to the rapid development of smart terminal devices. Cloud computing facilities may not be able to handle the massive amounts of data that will be generated in the future. Edge computing, a new paradigm after cloud computing, places edge servers closer to the terminals, improving the efficiency of processing the massive data generated by the terminals. However, during actual use, the edge computing system is limited by factors such as population density and service demand, resulting in suboptimal data processing efficiency and operator profits. This paper studies the deployment of edge servers in a fixed area, focusing on operator profitability. We propose the Advanced K-Means method, an improvement of K-means for users with different needs in a fixed area, to maximize system profitability at various times. Experimental results based on subscriber and base station datasets from Melbourne’s Central Business District (CBD) demonstrate that this deployment strategy improves profitability by at least 22 , outperforming several other representative approaches.