A Whale Optimization Algorithm for Resource Optimized Load Balancing in IoT Based Edge Network

With the rise of IoT applications, load balancing at edge has attracted more and more attention to achieve effective resource utilization cost-effectively with latency minimized as well as energy efficiency. Traditional load balancing heuristics and metaheuristics do not succeed in dynamic environments due to their lack of scalability, high energy consumption and the way they mismanage job allocations. In this paper we introduce a bio-inspired optimization framework for edge networks, referred as WOLOPA (Whale Optimization Algorithm for Resource Optimized Load Balancing), which is addressed to efficient resource allocation, reduction of network latency and overall system improvement. The task scheduler is augmented with SMP factors to locally allocate workload efficiently by leveraging the exploration exploitation benefits of WOA. For network lifetime simulation, better resource utilization and 40 % less latency is achieved by WOROLB than LBRO and 25 % more throughput than that of LBRO (EMOPSOC). The WOROLB also has good interactions with resource management strategies, and can achieve better performance at low cost.