Fault-tolerant table generating algorithm: further work on MNLIRS routing scheme

In networks of clusters, a collection of computing nodes works in concert to solve large application problems. The nodes exchange messages and coordinate their efforts by sending and receiving messages through a routing network. Consequently the achieved performance of such machines depends critically on that of their routing networks. Commonly there is a routing table at each node of the network. When a message is to be sent, the routing table is consulted for the direction. Multi-node label interval routing scheme (MNLIRS) is a routing framework which provides a compressed routing table for efficient routing with fault-tolerance. In the scheme, fault-tolerant tables can be introduced to improve the capability of end-to-end fault-tolerant delivery. In this paper, the algorithms to generate the fault-tolerant table are presented and evaluated.