• DocumentCode
    3375726
  • Title

    FLUX Networks: Interconnects on Demand

  • Author

    Vassiliadis, Stamatis ; Sourdis, Ioannis

  • Author_Institution
    TU Delft
  • fYear
    2006
  • fDate
    38899
  • Firstpage
    160
  • Lastpage
    167
  • Abstract
    In this paper, we introduce the FLUX interconnection networks, a scheme where the interconnections of a parallel system are established on demand before or during program execution. We present a programming paradigm which can be utilized to make the proposed solution feasible. We perform several experiments to show the viability of our approach. We experiment on three case studies, evaluate different algorithms, developed for meshes or binary trees, and map them on "grid"-like physical interconnection networks. Our results clearly show that, based on the underlying network, different mappings are suitable for different algorithms. Even for a single algorithm different mappings are more appropriate, when the processing data size or the number of utilized nodes changes. The implication of the above is that changing interconnection topologies/mappings (dynamically) on demand depending on the program needs can be beneficial
  • Keywords
    multiprocessor interconnection networks; parallel algorithms; parallel programming; FLUX interconnection network; binary tree algorithm; grid-like physical interconnection network; interconnection topology; mesh algorithm; parallel system interconnection; program execution; programming paradigm; Computer networks; Concurrent computing; Microarchitecture; Multiprocessor interconnection networks; Network topology; Programming; Proposals; Scalability; Software performance; Very large scale integration;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Embedded Computer Systems: Architectures, Modeling and Simulation, 2006. IC-SAMOS 2006. International Conference on
  • Conference_Location
    Samos
  • Print_ISBN
    1-4244-0155-0
  • Type

    conf

  • DOI
    10.1109/ICSAMOS.2006.300823
  • Filename
    4084764