Fast and Efficient Submesh Determination in Faulty Tori

In a faulty torus/mesh, finding the maximal fault-free submesh is the main problem of reconfiguration. Chen and Hu [1] proposed a distributed method to determine the maximal fault-free submesh in a faulty torus. In this paper, we show that it is sufficient to apply the distributed algorithm proposed by Chen and Hu [1] to only few nodes of a torus. The time for determination of the maximal fault free submesh/submeshes (MFSS) is considerably reduced, by reduction in the number of messages needed for determination of MFSS. In addition, it also reduces the congestion in the network. We present an algorithm to determine the smallest submesh containing all faulty nodes in a torus. The proposed algorithm has a time complexity of O(n(m + k)) for a k-ary n-cube with m faults. Intensive simulation study reveals that number of messages is significantly reduced compared to Chen and Hu’s [1] method.