Self-Organized Bio-inspired Service Placement Algorithm in Future Networks

الملخص

Through this research paper, we studied the service placement issue in future networks, and proposed a Bio-inspired algorithm for placing replicas of services in the network based on inspired information of the load generated by client nodes, and the network topology that is constantly changing due to the change in the locations of the nodes and their mobility. It is known that future networks demand a high degree of self-organization to enable coping with the high dynamics of the networks while keeping performance optimized. One of the main challenging tasks in this context is a self-organized service placement. Service placement issue refers to the problem of selecting which node in the network is most suitable for hosting a service. An optimal placement of service instances (replicas) in a network minimizes the cost of serving clients, improves the connectivity between clients and servers, enables maintaining the quality of service and improves the use of available resources. Experience and results have shown that the proposed algorithm achieves an improvement in performance in terms of meeting the required services within a shorter time and a smaller bandwidth, and thus lower cost compared to the general network model (client / server). As this algorithm achieves the ideal placement of service replicas via monitoring the load within the server node and its neighborhood, selecting the node from which the largest load is received and copying the service to it, so that the distance crossed by requests coming from the client nodes becomes as small as possible as a result of placing services in locations close to them.