Title :
Performance analysis of parallel database sort operations in a heterogenous Grid Environment
Author :
Mach, Werner ; Schikuta, Erich
Author_Institution :
Dept. of Knowledge & Bus. Eng., Univ. of Vienna, Vienna
Abstract :
In this paper an analytical comparison of the performance behavior of parallel flavors of the well-known Binary Merge Sort and Bitonic Sort algorithm in a grid environment is presented. To keep the analysis clear and focused we concentrate on a limited number of characteristic parameters and develop a concise but comprehensive analytical model both for a generalized multiprocessor framework and a simplified heterogeneous grid environment. We justify that a meaningful model can be built upon only three characteristic parameter sets, describing node processing performance, the I/O and the disk bandwidth, which are the parameters for the optimization the Grid workflow by a smart brokerage mechanism. Based on these results the paper proves that by a smart enhancement of the algorithms exploiting the specifics of the grid the well-known results of Bitton et al. for a homogenous multi-processor architecture are invalidated and reversed for a heterogenous Grid environment.
Keywords :
grid computing; merging; parallel databases; software performance evaluation; sorting; I/O performance; binary merge sort; bitonic sort algorithm; disk bandwidth; generalized multiprocessor framework; grid workflow optimization; heterogeneous grid environment; node processing performance; parallel database sort operation; performance analysis; smart brokerage mechanism; Algorithm design and analysis; Bandwidth; Computer architecture; Database machines; Database systems; Distributed databases; Parallel processing; Performance analysis; Sorting; Supercomputers;
Conference_Titel :
Cluster Computing, 2007 IEEE International Conference on
Conference_Location :
Austin, TX
Print_ISBN :
978-1-4244-1387-4
Electronic_ISBN :
1552-5244
DOI :
10.1109/CLUSTR.2007.4629279