A Predictive Approach for Selecting Suitable Computing Nodes in Grid Environment by Using Data Mining Technique - Advances in Computational Science and Engineering - page 202

Table 3. The computed result in 7.00 to 9.00 o'clock

Node's Name A1 A2 A3 A4 A5 A6 A7 A8 A9

Node 1 0.89 0.9 0.9 0.9 1 0.55 0 0.54 0.02

Node 2 0.87 0.94 0.9 0.85 0.95 0.8 0.21 0.77 0.04

Node 3 0.9 0.92 0.9 0.85 1 0.67 0.39 0.97 0.05

Node 4 0.94 0.95 0.95 1 1 0.9 0.48 0.87 0.43

Node 5 0.95 0.96 1 0.9 1 0.85 0.48 0.98 0.39

Node 6 0.96 0.95 1 1 1 0.7 0.48 0.16 0.38

Node 7 0.92 0.93 0.9 0.95 1 0.85 0.48 0.8 0.28

Node 8 0.8 1 0.85 1 1 1 0.78 0.65 1

this purpose. Then job sent to this node for execution and after a little time, job fin-
ished successfully on Node4.
The priority list nodes for this job were as following (high reliability priority for
job): Node 4 > Node 5 > Node 8 > Node 7 > Node 6 > Node 3 > Node 2 > Node 1
If this job had a real-time priority, the below order was selected by broker provider
Application: Node 8 > Node 4 > Node 5 > Node 7 > Node 3 > Node 2 > Node 6 > Node 1
In following days, all measures were based-on broker provider application. After
doing 120 measures, we took a below results (Table 4) to execute a job with 10.24
MB and 22 minutes for approximate time. The following results sent by each partici-
pated nodes at time 12-15:

Table 4. The computed result in 12.00 to 15.00 in 21August (job size 10.24 MB at 12:45
o'clock)

Node's
Name

A1 A2 A3 A4 A5 A6 A7 A8 A9

Node 1 0.902 0.87 0.95 0.94 0.8 0.45 0 0.78 0.02

Node 3 0.908 0.93 1 0.86 0.8 0.53 0.39 0.92 0.05

Node 5 0.961 0.96 0.95 0.89 1 0.89 0.48 0.96 0.39

Node 6 0.964 0.93 0.95 0.94 0.9 0.66 0.48 0.80 0.38

Node 7 0.92 0.94 0.95 0.92 0.6 0.91 0.48 0.93 0.28

Node 8 0.81 1 0.9 0.89 0.8 1 0.78 0.75 1

As you see, there are only 6 nodes in available. This job is considered as a Real-
time job, thus Node 8 was selected as the best node by proposed application. The selec-
tion priority of nodes is as following:

Node 8 > Node 5 > Node 6 > Node 7 > Node 3 > Node 1
If the job had a very high reliability priority, then the selection priority will be as
following: Node 5 > Node 8 > Node 6 > Node 7 > Node 3 > Node 1