摘要
超立方体因其特殊的结构和良好的性质成为多处理机系统最常用的互联网络之一.在实际的超立方体网络中两个处理器间的双向连接常常是通过两个方向相反的单向信道物理实现的.诊断度是度量系统识别故障能力的指标,而PMC模型是一种基于测试的系统级故障诊断模型.研究了当丢失一些单向信道时双向超立方体网络在PMC模型下的诊断度并确定了具有特定诊断度的双向超立方体所需的最小测试数.
The hypercube is one of the most popular intereonnection networks for multiprocessor systems because of its special structure and good properties. In most hypereube multiprocessor architectures, each link is physically implemented by two opposite unidirectional channels. The diagnosability is an indicator for measuring the system' s ability to identify faults, while PMC model is a system level fault diagnosis model based on testing. The diagnosability of bidirectional hypereube networks with missing links under the PMC model is studied and the minimum number of tests in a bidirectional hypercube with a specific diagnosability is determined.
作者
张雯丽
林上为
景小飞
ZHANG Wenli;LIN Shangwei;JING Xiaofei(School of Mathematical Sciences,Shanxi University,Taiyuan 03006,China)
出处
《河南科学》
2018年第11期1684-1688,共5页
Henan Science
基金
国家自然科学基金(61202017)
关键词
网络
有向图
超立方体
诊断度
容错性
network
digraph
hypercube
diagnosability
fault tolerance