Efficient Online Algorithm for Assigning Resources of Fog Computers and Communication Networks
Masayuki Tsujino · 2020
There is interest in increasing the efficiency of assigning resources to fog demands in the IoT infrastructure while satisfying delay requirements. An online algorithm for assigning the resources of both communication networks and fog computers from the viewpoint of communication carriers is proposed. The proposed algorithm uses the method of the multi-concurrent flow problem studied in network theory. The parameters in the algorithm are also adjusted based on a machine learning approach for various conditions in actual IoT services. Numerical experiments were conducted to show that assignment efficiency in terms of blocking rate can be greatly improved by about 50% with the proposed algorithm, which can be further improved by adjusting the algorithm's parameters.