Sequential diagnosability of star graphs

System-level diagnosis for n — star multi-computer systems is considered. In this approach, processors test each other and the test results are collected and analyzed to determine faulty processors. Three basic diagnosis strategies based on this approach are considered for use in n — star: one — t, one — t t , and seq — t. The one — t and one — t t are one-step diagnosis strategies which involve only one testing phase and one repair phase. The goal here is to identify and replace all faulty processors through one mutual testing phase. The other method is sequential diagnosis strategy (seq — t) which involves multiple iterations of testing and repair phases. The cost and diagnostic power of these three basic strategies for n — star is considered. In this paper some new properties of the sequential diagnosis strategy are discussed. In particular, we give a lower bound for the degree of diagnosability of the seq — t strategy in n − star multi-computer system.