
The above Markov chain is countinous time and the
input of the C++ version of the tool should consist of matrixes that define its infinitesimal
generator. The order of the input matrices is showed in the following. Note
that the matrices in this example are simple scalars (the explanations on
the right should not be included in the input file).
1 
Size of
boundary portion 
1 
Size of
repetitive portion 
1 
Burst
size of the process 
0.000000000000001 
Numerical accuracy 
2.0 
Local transitions in boundary
portion 
3.0 
Backward transitions to boundary
portion 
2.0 
Forward transitions from boundary
portion 
3.0 
Backward transitions in the
repetitive portion 
5.0 
Local transitions in the repetitive
portion 
2.0 
Forward transitions in the repetitive
portion 
> load QBDtestcase1
> G = QBD_G_ETAQA(A);
> pe = QBD_pi_ETAQA(B,A,B0,G);
> meanqlen = QBD_qlen_ETAQA(B,A,B0,pe,1);

The above Markov chain is countinous time and the
input of the C++ version of the tool should consist of matrixes that define its infinitesimal
generator. The order of the input matrices is showed in the following. Note
that the matrices in this example are simple scalars (the explanations on
the right should not be included in the input file).
1 
Size
of boundary portion 
2 
Size
of repetitive portion 
1 
Burst
size of the process 
0.000000000000001 
Numerical accuracy 
2.0 
Local transitions in boundary
portion 
6.0 3.0 
Backward transitions to boundary
portion 
2.0 0.0 
Forward transitions from boundary
portion 
6.0 0.0 3.0 0.0 
Backward transitions in the
repetitive portion 
12.0 4.0 0.0 5.0 
Local transitions in the repetitive
portion 
2.0 0.0 0.0 2.0 
Forward transitions in the
repetitive portion 
> load QBDtestcase2
> G = QBD_G_ETAQA(A);
> pe = QBD_pi_ETAQA(B,A,B0,G);
> meanqlen = QBD_qlen_ETAQA(B,A,B0,pe,1);

The above Markov chain is countinous time and the
input of the tool should consist of matrixes that define its infinitesimal
generator. The burst size in this example is 3. The order of the input matrices
is showed in the following (the explanations on the right should not be included
in the input file):
1 
Size
of boundary portion 
1 
Size
of repetitive portion 
3 
Burst
size of the process 
0.000000000000001 
Numerical accuracy 
2.0 
Local transitions in boundary
portion 
2.0  Forward transitions from boundary
portion 
6.0 
Backward transitions to boundary
portion 
3.0 

1.0 

8.0 
Backward transitions to the
first set of the repetitive portion 
3.0 

2.0 

1.0 

2.0 
Forward transitions in the
repetitive portion 
8.0 
Local transitions in the repetitive
portion 
3.0 
Backward transitions tin the
repetitive portion 
2.0 

1.0 
The input to the Matlab version of the tool should be transition matrices as defined above. The input format can be found in a Matlab mat file: GIM1testcase. To use the mat file and to use the functions of MAMSolver, see the following example.
> load GIM1testcase
> R = GIM1_R_ETAQA(A);
> pe = GIM1_pi_ETAQA(B,A,R,'Boundary',B0);
> meanqlen = GIM1_qlen_ETAQA(B,A,R,pe,1,'Boundary',B0);

The above Markov chain is countinous time and the
input of the tool should consist of matrixes that define its infinitesimal
generator. The burst size in this example is 3. The order of the input matrices
is showed in the following (the explanations on the right should not be included
in the input file):
1 
Size
of boundary portion 
2 
Size
of repetitive portion 
3 
Burst
size of the process 
0.000000000000001 
Numerical accuracy 
3.5 
Local transitions in boundary
portion 
10.0 12.0 
Backward transitions to boundary
portion 
2.0 0.0 
Forward transitions from boundary
portion 
1.0 0.0 

0.5 0.0 

10.0 0.0 12.0 0.0 
Backward transitions in the
repetitive portion 
17.5 4.0 0.0 15.5 
Local transitions in the repetitive
portion 
2.0 0.0 0.0 2.0 
Forward transitions in the
repetitive portion 
1.0 0.0 0.0 1.0 

0.5 0.0 0.0 0.5 
The input to the Matlab version of the tool should be transition matrices as defined above. The input format can be found in a Matlab mat file: MG1testcase1. To use the mat file and to use the functions of MAMSolver, see the following example.
> load MG1testcase1;
> G = MG1_G_ETAQA(A);
> pe = MG1_pi_ETAQA(B,A,G,'Boundary',C0);
> meanqlen = MG1_qlen_ETAQA(B,A,pe,1,'Boundary',C0);