Research On Cloud Computing Multi-Objective Task Scheduling Problem Based On Improved Particle Swarm Algorithm

Task scheduling problem is a classic problem in the theoretical research of cloud computing,and it is also a hot issue in the research.The result of task scheduling is related to the service quality of cloud computing and affects the user experience.At present,it is a research hotspot to use heuristic optimization algorithms to solve task scheduling problems.Among them,particle swarm optimization is excellent in solving task scheduling problems.The traditional particle swarm optimization algorithm has the defects of premature convergence and low convergence accuracy when solving cloud computing multiobjective task scheduling.In this thesis,by optimizing the traditional particle swarm optimization algorithm,an improved particle swarm optimization algorithm(BIPSO)is proposed.The test verifies that the algorithm has superior performance.The improved particle swarm optimization algorithm is introduced into cloud computing multi-objective tasks scheduling.A multi-objective task scheduling strategy(MOTS-PSO)optimized by particle swarm optimization.This thesis improves the particle swarm optimization algorithm and applies the improved algorithm to cloud computing task scheduling,and the overall work content to achieve multi-objective task scheduling optimization is:1.Introducing dynamic adaptive inertia weight strategy and dynamic learning factor strategy to dynamically adjust the flying step size of particles,it improve the particle's optimization ability,and avoids falling into the local optimal solution in the late stage of algorithm optimization.2.Introducing the probability update mechanism of the flower pollination algorithm,balance the global search and local search of the particle swarm optimization algorithm,and optimize and improve the particle global search position update formula to improve the convergence accuracy of the algorithm.3.Introducing the firefly algorithm to generate "elite solution",combined with "elite solution" to improve the local search position update formula.It helps the algorithm jump out of the trap of the local optimal solution,improves the convergence accuracy of the algorithm;at the same time,monitors the continuous optimization of the particle swarm algorithm The number of iterations that does not change,introducing an "elite solution" to disturb the position of the particles,helps the particles jump out of the local optimum.4.Introducing a new boundary processing mechanism to optimize the position of out-of-bounds particles,it solves the problem that out-of-bounds particles cannot approach the optimal solution,and improves the optimization efficiency of the algorithm.5.Applying the improved particle swarm optimization algorithm to cloud computing multi-objective task scheduling and experimenting with the Cloud Sim platform,the results show that the improved particle swarm optimization algorithm has excellent performance when solving task scheduling problems of different sizes.