The Research Of Real Time Task Scheduling Based On Multi-Core Processors

With the rapid development of the communications industry,4G Technology has matured,while 5G research and development are also moving towards the 2020 commercial planning,the business volume of the communications industry is showing an explosion of growth.Therefore,how to use the limited processor resources to support the increasing number of communication information's calculation and transmission tasks is the problem that cannot be ignored.In order to solve this problem,we can use the multicore array platform,which can convert the traditional Serial task processing mode to the parallel task computing mode.The key point of this part is also the focus of this paper,which is to study the task scheduling algorithm on multi-core wireless array,and to optimize the performance of system.Task scheduling algorithms aims to assign tasks to appropriate processor resources and determine the order of tasks execution in each processor resource.Among them,it is necessary to consider the dependencies between tasks,because it has the impact on information transferring between the tasks,but also need to consider the entire task scheduling delays,power consumption and other performance indicators.For a general task graph,its optimal task scheduling problem is an np-complete problem.Therefore,it is necessary to study the efficient scheduling algorithm for the parallel computation of multi-core array platforms.In this paper,we first introduce the related technology of wireless multi-core array platform,including interconnection model and network routing,and give the basic background of task scheduling.After that,we give the mathematical definition and modeling in the process of task scheduling on wireless multi-core array platform,which is a formal description of the problem.Next,we take two simple instance tasks as examples,using three kinds of typical random search algorithms for performance simulation.After a certain understanding of the performance of typical algorithms,we introduce the focus of this paper: a task map formed by the PDSCH signal processing flow in LTE.Through the performance index analysis,we find that the typical random search algorithm cannot meet the performance requirements,and therefore we introduces a new algorithm: Chunking splicing algorithm.After comparison,the new algorithm has certain advantages over the typical random algorithm in terms of optimization time and performance.At the same time,in order to verify the performance of the new algorithm,we introduce the random task graph,and use 4 kinds of typical table scheduling algorithms whose simulation times are very close to the new algorithm to carry on the performance comparison.