Adaptive fault-tolerant routing in cube-based multicomputers using safety vectors

摘要

Reliable communication in cube-based multicomputers using the safety vector concept is studied in this paper. In our approach, each node in a cube-based multicomputer of dimension n is associated with a safety vector of n bits, which is an approximated measure of the number and distribution of faults in the neighborhood. The safety vector of each node can be easily calculated through n-1 rounds of information exchange among neighboring nodes. Optimal unicasting between two nodes is guaranteed if the kth bit of the safety vector of the source node is one, where k is the Hamming distance between the source and destination nodes. The concept of dynamic adaptivity is introduced, representing the ability of a routing algorithm to dynamically adjust its routing adaptivity based on fault distribution in the neighborhood. The feasibility of the proposed unicasting can be easily determined at the source node by comparing its safety vector with the Hamming distance between the source and destination nodes. The proposed unicasting can also be used in disconnected hypercubes, where nodes in a hypercube are disjointed (into two or more parts). We then extend the safety vector concept to general cube-based multicomputers.