International Conference «Mathematical and Informational Technologies, MIT-2011»
(IX Conference «Computational and Informational Technologies for Science,
Engineering and Education»)

Vrnjacka Banja, Serbia, August, 27–31, 2011

Budva, Montenegro, August, 31 – September, 5, 2011

Amirshahi B.  

A new algorithm of clustering grid-resources for optimization of data exchanges at distributed computing networks

In this paper, I will discuss the problem of aggregating the results of a distributed computation, in a multi processor network. As we know, data aggregation provides a challenge, especially when we deal with large sets of data points, in a noisy background. As a solution, I will represent a new parallel clustering algorithm, which employs a minimum spanning tree (MST) of the graph and solves data aggregation problem for GRID-resources. The computational bottleneck of my algorithm is the construction of an MST of a graph, for which a parallel algorithm is employed.

Keywords: GRID-computing, Cluster, parallel clustering algorithm, hierarchical clustering, Minimum spanning tree.

Abstracts file: Abstract.doc
Full text file: Full text.doc

To reports list

© 1996-2019, Institute of computational technologies of SB RAS, Novosibirsk