Multi-objective resource allocation in edge computing using TF-IDF scheme

Recently, the rapid increase in the number of mobile phones and IoT devices connected to networks has led to a decrease in the throughput capacity of Internet network channels and delays in delivering processed data to users. To eliminate these delays in network channels, edge computing systems are widely used. Edge computing systems enable data to be processed on computing devices close to the users, without being sent to remote cloud servers. These systems reduce network latency, allow fast real-time data processing, and enable a reduction in the costs of services. This paper proposes a model of an edge computing system based on data processing centers. Two criteria, resource usage frequency and number of users, are taken as the basis for the allocation of cloud resources in edge computing networks. For balancing these criteria, the TF-IDF scheme is used. The proposed approach is formalized as a two-objective optimization model. An algorithm for selecting the optimal solution from the Pareto-optimal front is employed.