Mapping functions and data redistribution for parallel files

  • Tagung:

    Zeitschriftenartikel 

  • Autoren:

    Florin Isaila
    Walter Tichy
     

  • Summary

    Data distribution in memory or on disks is an important factor influencingthe performance of parallel applications. On the other hand, programs or systems,like a parallel file system, frequently redistribute data between memory and disks.This paper presents a generalization of previous approaches of the redistributionproblem. We introduce algorithms for mapping between two arbitrary distributionsof a data set. The algorithms are optimized for multidimensional array partitions. Wemotivate our approach and present potential utilizations. The paper also presents acase study, the employment of mapping functions, and redistribution algorithms in aparallel file system. 

  • Jahr:

    2008 

  • Links:
Beteiligte Mitarbeiter (zufällige Reihenfolge)
Titel Vorname Nachname

Bibtex

@article{,
author={Florin Isaila, Walter Tichy},
title={Mapping functions and data redistribution for parallel files},
year=2008,
month=Dec,
publisher={Springer},
volume={46},
url={https://ps.ipd.kit.edu/downloads/za_2008_mapping_functions_data_redistribution.pdf},
abstract={Data distribution in memory or on disks is an important factor influencingthe performance of parallel applications. On the other hand, programs or systems,like a parallel file system, frequently redistribute data between memory and disks.This paper presents a generalization of previous approaches of the redistributionproblem. We introduce algorithms for mapping between two arbitrary distributionsof a data set. The algorithms are optimized for multidimensional array partitions. Wemotivate our approach and present potential utilizations. The paper also presents acase study, the employment of mapping functions, and redistribution algorithms in aparallel file system.},
number={3},
pages={213-236},
journal={Journal of Supercomputing},