Cloud resource scheduling strategy based on improved particle swarm optimization algorithm for smart city system

Smart city facilitates people's lives from all aspects, but the complex equipment data analysis and processing, huge data calculation limit the development of smart city system, so reasonable scheduling of cloud computing resources make the smart city system hard to operate smoothly and orderly. To cope with problems above concerned, an novel method combined particle swarm algorithm with improved strategy is applied: First, the weight parameter of the social and cognitive coefficients in the particle swarm algorithm were adjusted to better conform to the optimal solution and avoid getting stuck in local optima; Second, an inertia weight function and an adjustment operator was defined during the search process to optimize position updates, thereby the convergence efficiency was improved and the global optimal solution was conformed; third, the load balance in cloud environments is considered by limiting the load rate of physical machines to place physical resources, then the service quality and user needs are assured. We can see in experiments that the proposed method can effectively integrate into the system, by searching the global optimal solution efficiently and improving the convergence efficiency of traditional particle swarm, the resource utilization and system performance was improved.