load-tolerant many-objective optimization for resource allocation in sdn-based iot

The increasing workload in internet of things (iot) environments leads to network congestion and resource imbalance, which significantly degrades quality of service (qos). software-defined networking (sdn) provides a flexible control paradigm for improving qos and load balancing; however, most existing approaches are limited to multi-objective formulations and do not explicitly address overload conditions. therefore, there is a need for many-objective qos optimization frameworks that can jointly consider multiple conflicting objectives while maintaining load tolerance in sdn-based iot systems.in this paper, a load-tolerant many-objective resource allocation framework is proposed to simultaneously optimize user and service provider qos requirements. specifically, cost and response time are minimized for users, while energy consumption is minimized and resource utilization is maximized for service providers. a pareto-based many-objective evolutionary optimization process is employed to generate diverse non-dominated solutions, which are evaluated under overload conditions.to address infeasible allocations under overload, a load-tolerance mechanism is introduced to identify admissible solutions and improve system robustness. this mechanism enables sustained task allocation even when pareto-optimal solutions violate load threshold, thereby increasing the task acceptance rate under overload conditions.simulation results demonstrate that the proposed many-objective grey wolf optimization (magwo)-based framework improves resource allocation cost by 13.85%, response time by 17.2%, energy consumption by 15.8%, and resource utilization by 10.25%. furthermore, the proposed load-tolerant strategy increases the task acceptance rate by 8.4% compared to nsga-iii and many-objective particle swarm optimization (mapso).