The Study on Adaptive Routing Algorithm of 2-D Torus Network with Fault Tolerance

Abstract

A 2-D torus network is one of the most popular networks for parallel processing. Many algorithms have been proposed based on the turn model, but most of them cannot be applied to a torus network without modification. In this paper, we mention the North-South First (NSF) routing that is applicable to a 2-D torus and combines the North-First method (NF) and the South-First method (SF). Our methods focused on the proposal of routing algorithms aimed at avoiding congestion of the coupled network, so we have not evaluated the fault tolerance. In fact, since the proposed method was a routing algorithm that guarantees the shortest path between source and destination, it was superior in congestion tolerance, but it was not known whether it had fault tolerance. In this paper, we evaluate the congestion tolerance of NSF by software simulation and evaluate the fault tolerance by simulation with faulty PE. Moreover, we propose an improved North-South First method (Improved NSF, NSF-IP) which is a new routing algorithm with improved fault tolerance by correcting the conventional NSF algorithm. For the proposed method, we evaluated both of congestion resistance and fault tolerance by dynamic communication performance evaluation by simulation. The software simulation showed that improved NSF method has higher performance.