Application of Chaotic Cat Swarm Optimization in Cloud Computing Multi Objective Task Scheduling

Abstract

In cloud computing multi-task scheduling, a large number of tasks require the system to have excellent task scheduling capabilities to ensure stable and efficient operation of the system. However, the integration and scheduling of multi-task resources in cloud computing directly affect the effectiveness of cloud computing services. Traditional cloud computing task scheduling has low efficiency and single objective optimization, which cannot meet the requirements of cloud computing tasks. In this regard, cat swarm optimization model is used to construct a multi-objective task scheduling model for cloud computing, achieving cloud computing tasks optimization. Cloud computing task objectives were analyzed, and a multi-objective task scheduling model was constructed with execution time and system load as scheduling objectives. Considering multi-objective task scheduling complexity, cat swarm optimization model was introduced for solution. Cat swarm optimization model can easily fall into local optimization. This model was improved by adjusting weight factor and fitness. In algorithm model performance analysis, Ackley function was used for optimization testing. The new model tends to converge after 445 iterations. At this moment, its optimal optimization value is 0.506, which is superior to other two optimization models. In cloud computing instances analysis, the cloud computing execution time was tested. This proposed model tends to converge after 102 iterations, with a task execution time of 6.23 seconds. The traditional optimization model has a task execution time of 6.51 seconds. The particle swarm optimization model has a task execution time of 6.96 seconds. In task cost analysis, this proposed model has a task execution cost of 3326 yuan when tasks number is 500, which is lower than other two models. From this, it can be seen that the proposed multi-objective task scheduling model has excellent application effects, which can effectively optimize and improve traditional cloud computing multi-objective scheduling, and ensure the stability of system operation. The research content provides important technical support for the management and scheduling optimization of cloud computing resources.