Improved Load Balancing for Large-Scale Smart Cities in SDN-IoT Networks

The SDN-IoT execution approach, which relies on a distributed set of many controllers, is inflexible in the face of exponential growth in data traffic. As a consequence, there is an imbalance in the workload between the managers, which in turn increases the probability of packet loss, slows down responses, and causes other issues with overall network performance. This study sets the basis for the multiple distributed controller load balancing method on a massive-scale SDN-IoT for smart cities. Then, depending on the variable CPU Usage in the centralized plane, the scientists proposed a technique for disadvantaging the load through numerous controllers. Experimental findings are analyzed using the mininet simulator, and the results are verified in the RYU controller using dynamic load balancing algorithms including Nash bargaining, efficient switch migration load balancing, efficiency aware load balancing, and the suggested method. CPU Time is used to run and evaluate these methods, and it is guaranteed that CPU usage with task scheduling will be 20% higher than without it.