基于信号解释Petri网的可重构调度模型分析研究  

Research on Reconfigurable Scheduling Model Based on Signal Interpretation Petri Net

作　　者：栾岚 胡丹 LUAN Lan;HU Dan(College of Computer and Information Engineering,Guizhou University of Commerce,Guiyang Guizhou 550061,China;College of Big Data and Information Engineering,Guizhou University,Guiyang Guizhou 550025,China)

机构地区：[1]贵州商学院计算机与信息工程学院,贵州贵阳550061 [2]贵州大学大数据与信息工程学院,贵州贵阳550025

出　　处：《计算机仿真》2021年第12期411-414,479,共5页Computer Simulation

基　　金：云计算背景下的可重构计算技术策略优化(黔教合KY字[2018]279)。

摘　　要：复杂系统的任务调度对于提升系统性能至关重要,很多应用场景引入可重构调度模型来处理复杂系统中的并发、互斥和非同步任务。由于系统中的任务时间具有模糊特性,导致现有研究逻辑控制的效果不佳,为此提出了基于信号解释Petri网的可重构调度模型。考虑到Petri网擅长描述并发、互斥,以及非同步任务,设计了一种扩展Petri网,将重构系统时的影响因素,定义为包含七个变量的集合,并对资源变迁的过程和规则做了相应分析。根据任务执行对时间的要求,将其分成固定时间调度、周期调度,以及触发调度三种模式,并依次进行模型的分析优化。为了部署逻辑芯片,资源即为库所,操作即为变迁规则,资源与操作的联系即为信号解释,最终基于FPGA实现了信号解释Petri网的可重构调度模型。通过仿真验证了信号解释Petri网的可重构调度模型能够对复杂的系统任务进行良好的静态与动态重构,重构准确率和任务执行效率均得到显著提升。Task scheduling of a complex system is very important to improve system performance. In many application scenarios, a reconfigurable scheduling model is introduced to deal with concurrent, mutually exclusive and asynchronous tasks in a complex system. Due to the fuzzy characteristics of task time in the system, the effect of existing research logic control is not good. Therefore, a reconfigurable scheduling model based on signal interpretation Petri net is proposed. Considering that Petri nets are good at describing concurrent, mutually exclusive and asynchronous tasks, an extended Petri net was designed. The influencing factors of system reconstruction were defined as a set of seven variables, and the process and rules of resource transformation were analyzed. According to the time requirements of task execution, it was divided into three modes: fixed time scheduling, periodic scheduling, and trigger scheduling, and the model was analyzed and optimized in turn. In order to deploy a logic chip, the resource was treated as the library, the operation was treated as the transition rule, and the connection between resource and operation was treated as signal interpretation. Finally, the reconfigurable scheduling model of signal interpretation Petri net was implemented based on FPGA. The simulation results show that the reconfigurable scheduling model of signal interpretation Petri net can reconstruct the complex system tasks both statically and dynamically, and the reconstruction accuracy and task execution efficiency are significantly improved.

关 键 词：信号解释 变迁规则 可重构调度模型 逻辑芯片 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]