![]() |
LINE Solver (C++)
Templated C++ port of the LINE queueing solver
|
What the analyzer returns, the reference's metrics plus its actualmethod. More...
#include <line/solvers/mva/solver_mva_sjn.h>
Public Attributes | |
| MvaSolution< T > | sol |
| std::string | actualmethod |
| std::string | warning |
| Non-empty when the reference would have called line_warning AND RETURNED, carrying its text. | |
What the analyzer returns, the reference's metrics plus its actualmethod.
Definition at line 92 of file solver_mva_sjn.h.
| std::string line::mva::SjnAnalyzerResult< T >::actualmethod |
Definition at line 94 of file solver_mva_sjn.h.
Referenced by line::mva::mva_dispatch(), and line::mva::solver_mva_sjn_analyzer().
| MvaSolution<T> line::mva::SjnAnalyzerResult< T >::sol |
Definition at line 93 of file solver_mva_sjn.h.
Referenced by line::mva::mva_dispatch(), and line::mva::solver_mva_sjn_analyzer().
| std::string line::mva::SjnAnalyzerResult< T >::warning |
Non-empty when the reference would have called line_warning AND RETURNED, carrying its text.
Today: the utilization cap binding, and the fixed point running out of iterations.
This is the shape solver_nc_cdf.h:83 established for a reference warning that accompanies a usable answer, as opposed to solver_mam_retrial.h:437, which throws because its warning accompanies an answer the port declines to stand behind. The SJN cap is the first kind: the numbers are stable and the population law still holds exactly, but their ACCURACY is not warranted, and MATLAB says so out loud. Dropping it would leave the caller an authoritative-looking number with a silent accuracy claim the reference explicitly declines to make.
Definition at line 109 of file solver_mva_sjn.h.
Referenced by line::mva::mva_dispatch(), and line::mva::solver_mva_sjn_analyzer().