Cloud Computing Task Scheduling And Scheduling Optimization Decision Problems In The Research

Cloud computing carries the idea of parallel computing, distributed computing, grid computing and utility computing, which has become a vibrant dynamic mode. Cloud computing leads to future development of information technology trends and directions at the same time, the cloud computing platform is to provide services and IT applications in all kinds of fields. Cloud computing platform is built above the large data centers, managing and controlling system of the entire cloud computing services, forming a cloud computing environment. In cloud computing, task scheduling and rescheduling affect the performance, income and other important aspects of cloud computing system. Therefore, the improvement of task scheduling and rescheduling strategies has always been one of the core research directions in cloud computing applications.Traditional task scheduling model and the classic scheduling algorithms are widely used in the field of grid computing, which have the advantage of more fast in task scheduling processing at the same time but also have the inadequacies of single scheduling decision optimization goal, which need to be improved and enhanced in order to meet the requirements of modern computing model. Therefore, the traditional task scheduling model has not been gradually suited to the existing modeling of the actual needs of the system performance, income, and many other targets, task scheduling strategies and algorithms also need to be further improved. Therefore, that is why the improvement in task scheduling and rescheduling in the cloud modeling and scheduling algorithm becomes an important research focus in the cloud computing system.In the background of cloud computing application, this thesis makes the following work:(1) Engage the academic research about task scheduling and task rescheduling in cloud computing, and establish the task scheduling and task rescheduling model which are more in line with the actual requirements.(2) Analyze the optimization decision-making and evaluation objectives by using qualitative and quantitative methods. Learning from the advantages of traditional classical scheduling algorithm and the traditional single-objective optimization task scheduling strategy, the context proposes the task scheduling algorithm based on classical scheduling algorithm for multi-objective optimization decision-making. In the same time, this thesis analyzes the problem of system delay in task scheduling and puts forward the task of rescheduling policy to better solve this problem.(3) Build the cloud computing simulation environment, test and verify the improved task scheduling model and the task scheduling algorithm in order to solve such scheduling problem by optimization. The results showed that:based on the classical algorithm, the proposed improvement of FMOOSA scheduling algorithm can get better results than the traditional classical algorithm. In the same time, based the priority PBRSA rescheduling algorithm meets the system requirements better and thus the task scheduling system has better scheduling performance.Finally, the main work of this thesis is summarized, having the outlook of the other related research direction and research points.