Many theoretical models of parallel computation are based on overly simplistic assumptions on the performance of the interconnection network. For example they assume constant latency for any communication pattern or innnite bandwidth. This paper presents a case study based on the FFT transpose algorithm, which is mapped on two families of scalable interconnection networks, the k-ary n-trees and...

We consider perfect 1-error correcting codes over a finite field with q elements (briefly q-ary 1-perfect codes). In this paper, a generalized concatenation construction for q-ary 1-perfect codes is presented that allows us to construct q-ary 1-perfect codes of length (q − 1)nm + n +m from the given q-ary 1-perfect codes of length n = (q1 − 1)/(q − 1) and m = (q2 − 1)/(q − 1), where s1, s2 are ...

Low-dimensional k-ary n-cubes have been popular in recent multicomputers. These networks, however, suffer from high switching delays due to their high message distance. To overcome this problem, Dally has proposed a variation, called express k-ary n-cubes, with express channels that allow non-local messages to partially bypass clusters of nodes within a dimension. K-ary n-cubes are graph topolo...

In this paper, we present a strongly fault-tolerant design for the k-ary n-cube multiprocessor and examine its reconfigurability. Our design augments the k-ary n-cube with ( j ) n spare nodes; each set of j regular nodes is connected to a spare node and the spare nodes are interconnected as a ( j )-ary n-cube. Our approach utilizes the circuit-switched capabilities of the communication modules ...