Performance Analysis for Large IaaS Clouds - Large Scale and Big Data: Processing and Management - page 566

Probability ( P h ) that a hot pool can accept a job for provisioning is given by

n h

P =1( ) B (18.27)

h h

Observe that P h is used as an input parameter in the RPDE submodel (Figure 18.2).

18.3.4 sharPe C oDe For h ot Pm s ubmoDel
We present the SHARPE code for hot PM submodel below.

format 8
bind
* Dummy value of P_block
P_block 0.01
P_not_blocked 1-P_block
lambda 1000
beta_h 12
m 4
Lh 2
mu 1
n_h 10
epsilon_zero 0.000001
end
* Function to compute effective arrival rate for each hot PM
func lambda_h(i)
if (i==0)

0

else

lambda*P_not_blocked/i

end
end
* Markov model to describe VM provisioning model for hot PM
markov hot(num_hot)
loop i,0,m-1

0_0_$(i) 0_1_$(i) lambda_h(num_hot)

end
loop j,0,m-1

loop i,0,Lh-1

$(i)_1_$(j) $(i+1)_1_$(j) lambda_h(num_hot)

end

end
loop i,0,Lh-1

$(i)_0_$(m) $(i+1)_0_$(m) lambda_h(num_hot)

end
loop i,0,m-1

0_1_$(i) 0_0_$(i+1) beta_h

end
loop j,0,m-2