Database Reference
In-Depth Information
18.3.4.2 Warm PM Submodel Outputs
Steady-state probability (
B
w
) that a warm PM cannot accept a job for VM provision-
ing is given by
−
∑
m
1
B
w
=
φ
()
w
+
φ
()
w
+
φ
()
w
+
φ
()
w
+
φ
()
w
(18.29)
(
L
, *, )
10
(
L
, ,)
10
(
L
, **,)
10
(
LLi
,,)
1
(
Lm
, ,)
0
w
w
w
w
w
i
=
1
Assuming
n
w
as independent warm PM submodels for the whole pool, the prob-
ability (
P
w
) that warm PM can accept a job for provisioning is computed as
n
w
P
=1( )
B
(18.30)
w
w
18.3.4.3 Cold PM CTMC
Cold PM CTMC submodel is shown in Figure 18.6. The overall cold pool submodel
is the set of
n
c
independent cold PM submodels. Key differences between a warm and
a cold PM submodel are (i) effective arrival rates (λ
w
vs. λ
c
), (ii) startup rates (γ
w
vs. γ
c
),
λ
c
λ
c
λ
c
0,0,0
0,1*,0
L
c
,1*,0
…
γ
c
γ
c
λ
c
λ
c
µ
0,1,0
L
c
,1,0
λ
c
…
λ
c
0,1**,0
L
c
,1**,0
β
c
µ
β
h
β
c
µ
µ
β
h
0,1,1
…
(
L
c
−1),1,1
0,0,1
L
c
,1,1
λ
c
λ
c
λ
c
β
h
β
h
2µ
2µ
2µ
(
m
- 1)µ
β
h
β
h
β
h
(
m
- 1)µ
(
m
- 1)µ
(
m
- 1)µ
…
λ
c
0,0,(
m
−
1)
0,1,(
m
−
1)
(
L
c
−
1),1,(
m
−
1)
L
c
,1,(
m
−
1)
λ
c
λ
c
λ
c
β
h
m
µ
β
h
β
h
β
h
m
µ
m
µ
…
0,0,
m
1,0,
m
L
c
,0,
m
λ
c
λ
c
λ
c
FIGURE 18.6
VM provisioning submodel for each cold PM.
Search WWH ::
Custom Search