Fault Tolerant Gamma-Minus Network

Abstract

In large-scale supercomputers, thousands of processors are connected together to their respective memory modules which are controlled by several control units connected in parallel. Large data streams have to be communicated between these processors and memories through interconnection networks. A multistage interconnection network (MIN) is an efficient way to provide these communications at a very reasonable cost. For such large systems, MIN should be highly reliable and fast to meet the desired specifications of these high speed switch fabrics.Here, two new architectures of MIN are proposed, denoted as fault-tolerant Gamma-Minus (FTGM-1 & FTGM-2) networks. These proposed architectures are multistage multipath interconnection networks with totally disjoint paths and are highly reliable and fault-tolerant networks. The routing algorithm proposed for these structures is simple distance tag routing with non-backtracking which overcomes rerouting overheads and hence improve communication delays. The proposed MIN ensures multiple fault tolerance at each stage including input and output stage with minimal horizontal distance between source node to destination node. A comparison of different recently proposed MINs with these new designs are also given on reliability basis to prove the supremacy of these designs over the previously proposed networks.