Abstract:
Grid computing can provide abundant
resources to assist many studies and it can be
employed in more and more projects that have so
far yielded numerous findings. However, in grid
environment, many distributed resources are
integrated in different sites and resources are
often shared by multiple user communities and
can vary greatly in performance and load
characteristics. Therefore, a central problem in
grid environment is the selection of computation
nodes for execution. Automatic selection of node
is complex as the best choice depends on the user
request structure as well as the expected
availability of computation and communication
resources. This paper considers the ongoing
research on Rough Sets based selecting the best
resource node in computational grid. Our node
selection method is based on the CPU capacity,
communication bandwidth and memory and disk
availability on the resource nodes. The goal of
this paper is to enable the automatically select
the best available resource node on the network
for execution within a reasonable time. In this
paper, random data is used to test the validity of
proposed method. The result showed that this
method can provide the suitable probability of
the best resource selection.
