Hyperswitch network for the hypercube computer

Abstract

The performance of a parallel algorithm depends in a large part on the interconnection topology of the multicomputer system. The method presented in this paper realizes a kind of interconnection network, called a hyperswitch network, that is achieved using a mixture of static and dynamic topologies. Here, available or fault free paths need not be specified by a source because the routing header can be modified in response to congestion or faults encountered as a path is established. This method can be accomplished in a static topology such as the hypercube network if the nodes have switching elements which are capable of performing the necessary routing header revisions dynamically. Detailed simulation results show that the hyperswitch network is consistently more efficient than fixed path routing for large message traffic conditions. The simulation results also show that the hyperswitch network has equivalent latency overhead for messages with localized and antilocal destinations (i.e., less then a 25% difference between diameter 1 and 5).