Research On Cloud Computing Task Scheduling Based On Improved Particle Swarm Optimization Algorithm

Cloud computing is an emerging technology based on distributed computing,network computing,parallel computing,and virtualization.Allocating tasks submitted by users over the Internet to a virtual resource pool composed of a large number of heterogeneous or homogeneous computer infrastructure devices,Users only need to obtain various types of virtual resources through the network on demand,and do not need to care about their specific distribution location and configuration.With the rapid development of cloud computing,the number of users and demand is also increasing,How to allocate large-scale tasks to reasonable resource nodes and improve the scheduling efficiency and user satisfaction of cloud task scheduling is the focus of cloud computing research.Therefore,it is important to choose an efficient and feasible task scheduling algorithm.Because of the advantages of simple process,few algorithm parameters,and fast convergence speed,particle swarm algorithm is widely used in cloud computing task scheduling.The thesis analyzes the development status of cloud computing,and conducts in-depth research on the concept of cloud computing task scheduling,workflow and other related content.At the same time,the algorithmic ideas and main characteristics of the particle swarm algorithm were analyzed in depth.Combining the characteristics of particle swarm algorithm and the characteristics of cloud computing task scheduling problem,this paper proposes a cloud computing task scheduling strategy based on improved particle swarm algorithm—IPSO.First,the problem of uneven distribution of particles caused by random initialization of the population is first addressed.Chaos theory is added to the initialization of the particle population,and a large number of particles are generated using the Logistic mapping sequence,and the particles of better quality are selected for initialization,which improves the quality of the particles and enables the particles to be uniformly distributed during initialization.Then,because the particle swarm algorithm’s late convergence speed makes the algorithm easily fall into the local optimal solution,a dynamic inertia weight update method based on fitness value is designed to improve the algorithm’s late convergence speed and improve the quality of the global optimal solution So that the particles can perform a global search and finally find the optimal solution for the population;Finally,a fitness function based on task completion time is designed,and a mapping method of the continuous solution space of particle swarm algorithm to the discrete solution space of cloud computing task scheduling problems is proposed.Design comparative experiments on the Cloud Sim simulation platform to verify that the IPSO algorithm is suitable for cloud computing task scheduling problems,and test the performance of the IPSO algorithm.The IPSO algorithm is compared with the standard particle swarm algorithm and simulated annealing algorithm,and the experimental results show that the IPSO algorithm Good convergence,high accuracy of solution,can find the optimal solution in a short time,and has a wide range of application scenarios in cloud computing task scheduling to handle massive tasks.