Multidestination Message Passing in Wormhole k-ary n-cube Networks with Base Routing Conformed Paths
Issue No.01 - January (1999 vol.10)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/71.744844
<p><b>Abstract</b>—This paper proposes <it>multidestination message passing</it> on wormhole <it>k</it>-ary <it>n</it>-cube networks using a new <it>base-routing-conformed-path</it> (BRCP) model. This model allows both unicast (single-destination) and multidestination messages to co-exist in a given network without leading to deadlock. The model is illustrated with several common routing schemes (deterministic, as well as adaptive), and the associated deadlock-freedom properties are analyzed. Using this model, a set of new algorithms for popular collective communication operations, <it>broadcast</it> and <it>multicast</it>, are proposed and evaluated. It is shown that the proposed algorithms can considerably reduce the latency of these operations compared to the Umesh (unicast-based multicast) [<ref rid="bibl00761" type="bib">1</ref>] and the Hamiltonian path-based [<ref rid="bibl00762" type="bib">2</ref>] schemes. A very interesting result that is presented shows that a multicast can be implemented with <it>reduced</it> or <it>near-constant</it> latency as the number of processors participating in the multicast <it>increases</it> beyond a certain number. It is also shown that the BRCP model can take advantage of adaptivity in routing schemes to further reduce the latency of these operations. The multidestination mechanism and the BRCP model establish a new foundation to provide <it>fast</it> and <it>scalable collective communication</it> support on wormhole-routed systems.</p>
Wormhole routing, collective communication, broadcast, multicast, k-ary n-cubes, meshes, interconnection networks, deadlock-freedom, and interprocessor communication.
Dhabaleswar K. Panda, Sanjay Singal, Ram Kesavan, "Multidestination Message Passing in Wormhole k-ary n-cube Networks with Base Routing Conformed Paths", IEEE Transactions on Parallel & Distributed Systems, vol.10, no. 1, pp. 76-96, January 1999, doi:10.1109/71.744844