A Fog-Cluster Based Load-Balancing Technique

The Internet of Things has recently been a popular topic of study for developing smart homes and smart cities. Most IoT applications are very sensitive to delays, and IoT sensors provide a constant stream of data. The cloud-based IoT services that were first employed suffer from increased latency and inefficient resource use. Fog computing is used to address these issues by moving cloud services closer to the edge in a small-scale, dispersed fashion. Fog computing is quickly gaining popularity as an effective paradigm for providing customers with real-time processing, platforms, and software services. Real-time applications may be supported at a reduced operating cost using an integrated fog-cloud environment that minimizes resources and reduces delays. Load balancing is a critical problem in fog computing because it ensures that the dynamic load is distributed evenly across all fog nodes, avoiding the situation where some nodes are overloaded while others are underloaded. Numerous algorithms have been proposed to accomplish this goal. In this paper, a framework was proposed that contains three subsystems named user subsystem, cloud subsystem, and fog subsystem. The goal of the proposed framework is to decrease bandwidth costs while providing load balancing at the same time. To optimize the use of all the resources in the fog sub-system, a Fog-Cluster-Based Load-Balancing approach along with a refresh period was proposed. The simulation results show that Fog-Cluster-Based Load Balancing decreases energy consumption, the number of Virtual Machines (VMs) migrations, and the number of shutdown hosts compared with existing algorithms for the proposed framework.

Automated summary

The Internet of Things is crucial for the development of smart homes and smart cities. Fog computing, which brings cloud services closer to the edge, addresses issues such as increased latency and inefficient resource use. Load balancing plays a critical role in fog computing to distribute dynamic load evenly and optimize resource utilization. A framework was proposed in this paper to decrease bandwidth costs and provide load balancing, achieving energy savings and improved performance compared to existing algorithms.