Component Fault Diagnosis and Fault Tolerance of Alternating Group Graphs

Abstract

Abstract Reliability of a multiprocessor system becomes an important issue for parallel computing. Component diagnosability and component connectivity of a graph play crucial roles in assessing the vulnerability of an interconnection network, which are two significant indicators for the reliability and fault tolerance of a multiprocessor system. Until now, only a little knowledge of results have been known on $r$-component diagnosability and $r$-component connectivity. In this paper, we first propose the $r$-component diagnosability of $n$-dimensional alternating group graph $AG_{n}$ under PMC model. And then we promote our research on $AG_{n}$ by a fairly good construction for general $r$-component connectivity of $AG_{n}$, where $6\leq r\leq n-1$. The theoretical analysis and simulation show that the general $r$-component connectivity of $AG_{n}$ is larger than those of $Q_{n}$, $D_n$ and $FQ_{n}$.