Placement of IoT services in fog environment based on complex network features: a genetic-based approach

The growth of the Internet of Things (IoT) has caused an ever-increasing number of devices to be added to the network. Fog computing is an emerging technology that aims to overcome the common challenges, such as delay, bandwidth usage, and security, by bringing the process and the storage closer to the user. Services that should serve IoT nodes usually have complex multi-component structures. Thus, mapping such structures onto the dynamic and complex fog environment is challenging. In this paper, we propose a novel Fog Service Placement algorithm based on Complex Networks feature (FSPCN) by considering the community concept to overcome this issue. Previous research commonly formed communities solely based on the network structure. We argue that grouping fog nodes into balanced communities before service placement, based on the network structure and nodes and links attributes, can lead to more effective placement of IoT services concerning resource use and application delay. In addition, we have defined a neighborhood distance metric, calculated based on the number of common neighbors among communities, to prioritize communities. This improves the average number of hops from requesting nodes to the requested services and reduces delay and traffic within the network. The experimental results show that the proposed algorithm significantly outperforms state-of-the-art algorithms in terms of resource use and response time. Thus, the FSPCN method deploys more applications in the fog environment and decreases up to 17% the number of applications placed in the cloud. This method also reduces the average delay about 30%.