学术讲座【Conditional diagnosability of Cayley graphs generated by transpositions trees under the PMC model】

09.12.2014  18:50

时间:2014年12月10日(星期三)下午15:30

地点:仓山校区成功楼603报告厅

主讲: 台湾成功大学 张乃文博士 

主办:数学与计算机科学学院、福建省网络安全与密码技术重点实验室

专家简介:张乃文,2001年6月毕业于台湾大学数学系,2005年6月、2010年6月于台湾成功大学获计算机专业硕士和博士学位,2011年6月-2013年7月在台湾成功大学从事博士后研究工作,2014年至今任台湾成功大学资讯工程系助理研究员。主要从事大规模计算机系统的故障诊断与检测以及网络系统的容错性能分析等领域的研究工作,主要科研成果发表于《IEEE Transactions on Computers》、《IEEE Transactions on on Parallel and Distributed Systems》、《IEEE Transactions on Reliability》、《IEEE Transactions on Dependable and Secure Computing》、《ACM Transactions on Design Automation of Electronic Systems》。

报告摘要:Processor fault diagnosis has played an essential role in measuring the reliability of a multiprocessor system; the diagnosability of many well-known multiprocessor systems has been widely investigated. Conditional diagnosability is a novel measure of diagnosability by adding a further condition that any fault set cannot contain all the neighbors of every node in the system. Several known structural properties of Cayley graphs are exhibited. Based on these properties, we investigate the conditional diagnosability of Cayley graphs generated by transposition trees under the PMC model, and show that it is 4n-11 for n>= 4, except for the n-dimensional star graph, for which it has been shown to be 8n-21 for n>= 5 (refer to [Chang and Hsieh 2014]).