首页> 外文会议>International conference on computers and their applications >Model Checking Approach for Deadlock Detection in an Operating System Process-Resource Graph Using Dynamic Model Generating and Computation Tree Logic Specification
【24h】

Model Checking Approach for Deadlock Detection in an Operating System Process-Resource Graph Using Dynamic Model Generating and Computation Tree Logic Specification

机译:使用动态模型生成和计算树逻辑规范的操作系统过程资源图中死锁检测的模型检查方法

获取原文

摘要

Deadlock between processes and resources is a serious problem in development of operating system. Multiple methods were invented to deal with deadlock issue. Deadlock detection is one method that allows a deadlock to take place then detects thereafter which processes and resources have caused it. In traditional process-resource graph, we propose an approach to detect a deadlock by implementing model checking technique and Computation Tree Logic (CTL) specification. In this paper, we modified traditional process-resource graph such that the outcome graph satisfied valid model of Kripke structure, which overcame limitations of traditional representation of process-resource graph and still preserved every proposition, correctness, and property of the system. With the modified graph, we designed a CTL specification that verified whether or not there existed a deadlock caused by one or more pairs of process and resource. A Java application was developed to implement the proposed approach such that it was capable of dynamically generating a valid model for any process-resource graph input, dynamically generating CTL formula for specification, and verifying the model with corresponding CTL formula.
机译:进程和资源之间的死锁是操作系统开发中的一个严重问题。发明了多种方法来处理死锁问题。死锁检测是一种方法,它允许发生死锁,然后再检测导致死锁的进程和资源。在传统的过程资源图中,我们提出了一种通过实现模型检查技术和计算树逻辑(CTL)规范来检测死锁的方法。在本文中,我们对传统的过程资源图进行了修改,以使结果图满足Kripke结构的有效模型,从而克服了传统的过程资源图表示的局限性,并且仍然保留了系统的所有命题,正确性和属性。使用修改后的图,我们设计了一个CTL规范,该规范验证是否存在由一对或多对进程和资源引起的死锁。开发了Java应用程序以实现所提出的方法,从而使其能够为任何过程资源图输入动态生成有效模型,动态生成用于规格说明的CTL公式并使用相应的CTL公式验证模型。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号