Information Technology Reference
In-Depth Information
Table 8. Allocation of modules for Cluster C 0
m 00
P 0
P 1
P 2
P 3
P 4
E ijkn
15
7.5
15
15
7.5
T prkn
10
12
10
13
12
∑ (B ihj * D kl )
0
0
0
0
0
NEC ikn
25
19.5
25
28
19.5 (BEC 04 )
m 10
P 0
P 1
P 2
P 3
P 4
E ijkn
20
10
20
20
10
T prkn
10
12
10
13
19.5
∑ (B ihj * D kl )
6
9
6
3
0
29.5
(BEC 14 )
NEC ikn
36
31
36
36
m 20
P 0
P 1
P 2
P 3
P 4
E ijkn
17.5
8.75
17.5
17.5
8.75
T prkn
10
12
10
13
29.5
∑ (B ihj * D kl )
6
9
6
3
0
29.75
(BEC 21 )
NEC ikn
33.5
33.5
33.5
38.25
m 30
P 0
P 1
P 2
P 3
P 4
E ijkn
10
5
10
10
5
T prkn
10
29.75
10
13
29.5
∑ (B ihj * D kl )
4
6
4
2
0
NEC ikn
24
40.75
24(BEC 32 )
25
34.5
m 40
P 0
P 1
P 2
P 3
P 4
E ijkn
20
10
20
20
10
T prkn
10
29.75
24
13
29.5
∑ (B ihj * D kl )
8
9
12
7
6
38
(BEC 40 )
NEC ikn
48.75
56
40
45.5
which is 0.005 in this case. This is shown in Table
10 along with the random replacements for the
modules on the sick nodes.
So the new allocation becomes as shown in
Table 11.
The node numbers shown in the brackets are
the duplicate copies of the modules lying on the
sick nodes, which becomes active as soon as the
corresponding node fails. Assuming all the sick
nodes fail, the new allocation becomes Table 12.
For the new allocation, now, the turnaround
time can be calculated in the same way as shown
in Table 8 as per equation (i). The new turnaround
time calculation is shown in Table 13 for thresh-
 
Search WWH ::




Custom Search