Study On Improvement Of Ant Colony Algorithm Based On High Performance Computing Cluster

High-performance computing clusters are mainly used to handle complex computational problems.They are widely used in meteorology,oceans,and the environment(numerical forecasting,etc.),life sciences(gene sequencing,comparison,homology analysis,etc.),and computer-aided engineering(fluids).Mechanics,optimization analysis,characteristic analysis,simulation analysis,etc.).With the deepening of research and the requirement of calculation accuracy,the amount of data has increased exponentially,and higher requirements have been put on storage,calculation,node communication,job allocation,and resource scheduling of high-performance computing clusters.Blindly adding hardware devices to improve computing performance In addition to bringing huge power consumption,it also has a bottleneck.Researching resource scheduling strategies to maximize efficiency has always been the direction of high-performance development and the focus of scientific research.In the face of high concurrency,multi-computing models,and high-performance computing under big data storage,the timeliness of data and user response cannot really improve and improve.How to perform task allocation and resource invocation on high-performance clusters is to improve performance.The essential.This paper analyzes the principle,realization and advantages and disadvantages of PSO and ant colony algorithm,and combines the two algorithms to get PSO_ACO.Through simulation and verification,PSO_ACO is superior to other task scheduling algorithms in solving.In practical use,the PSO_ACO is further analyzed.The inadequacies of the algorithm are discovered and the improvement is continued.An adaptive algorithm based on PSO_ACO is proposed.Improvements are made on the basis of the original PSO_ACO,the search space is increased,the solution quality is improved,and the algorithm is avoided prematurely.Stagnation,trapped in a local optimal solution.Finally,experimental analysis shows that the performance of the algorithm has been further improved.The main work contents and innovations are as follows:(1)Analyze the architecture of high-performance computing clusters,study the job scheduling structure and key technologies;(2)Research task scheduling algorithm,put forward task scheduling model and load balance model;(3)The particle swarm optimization algorithm is combined with the ant colony algorithm.The parameters of the ant colony algorithm are calculated by the particle swarm optimization algorithm,and a PSO_ACO is proposed.(4)Through the further analysis of the PSO_ACO algorithm,an adaptive scheduling algorithm based on PSO_ACO is proposed to adapt to the large-scale calculation and improve the performance of the algorithm.(5)Through simulation experiments,the performance of PSO_ACO and other algorithms are compared and analyzed.It is concluded that the improved adaptive scheduling algorithm performs better than PSO_ACO.