Class MMAPPH1PRPRKt
-
- All Implemented Interfaces:
public final class MMAPPH1PRPRKt
-
-
Method Summary
Modifier and Type Method Description final static Map<String, Map<Integer, Matrix>>MMAPPH1PRPR(MatrixCell D, MatrixCell sigma, MatrixCell S, Integer numOfQLMoms, Integer numOfQLProbs, Integer numOfSTMoms, Matrix stCdfPoints, Double prec, Integer erlMaxOrder_, Matrix classes_)Analyzes multi-class MMAPK/PHK/1 queue with preemptive-resume (PRPR) priority. -
-
Method Detail
-
MMAPPH1PRPR
final static Map<String, Map<Integer, Matrix>> MMAPPH1PRPR(MatrixCell D, MatrixCell sigma, MatrixCell S, Integer numOfQLMoms, Integer numOfQLProbs, Integer numOfSTMoms, Matrix stCdfPoints, Double prec, Integer erlMaxOrder_, Matrix classes_)
Analyzes multi-class MMAPK/PHK/1 queue with preemptive-resume (PRPR) priority.
This function computes the moments of queue length and service time distributions for a multi-class Markovian Arrival Process (MMAP) with class-dependent Phase-Type (PH) service time distributions in a single-server queue with preemptive-resume priority scheduling.
Preemptive-resume means that when a higher-priority job arrives, it preempts the current job in service. The preempted job resumes its service later from where it was interrupted.
Note: The current implementation is based on extended algorithms for FCFS queues. For exact preemptive-resume results, refer to: G. Horvath, "Efficient analysis of the MMAPK/PHK/1 priority queue", European Journal of Operational Research, 246(1), 128-139, 2015.
- Parameters:
D- The MMAP arrival process matrices (D0...sigma- The initial phase probability vectors for service distributionsS- The transient generators of PH service distributionsnumOfQLMoms- Number of queue length moments to compute (default: null)numOfQLProbs- Number of queue length probability values (default: null)numOfSTMoms- Number of service time moments (default: null)stCdfPoints- Matrix of points for service time CDF evaluation (default: null)prec- Numerical precision for iterations (default: null)erlMaxOrder_- Maximum order for Erlang approximation (default: null)classes_- Indices of classes to analyze (default: null)- Returns:
Map of performance measures including "ncMoms", "ncDistr", "stMoms", "stDistr"
-
-
-
-