Package jline.api.nc

Class Me_cqn

java.lang.Object
jline.api.nc.Me_cqn
public final class Me_cqn extends Object

  • Method Details

    • me_cqn

      public static MeCqnResult me_cqn(int M, int R, Matrix N, Matrix mu, Matrix Cs, Matrix[][] P, Matrix c, int[] refstat)

    • me_cqn

      public static MeCqnResult me_cqn(int M, int R, Matrix N, Matrix mu, Matrix Cs, Matrix[][] P, Matrix c, int[] refstat, MeOqnOptions options)
      Maximum Entropy algorithm for Closed Queueing Networks. Stage 1 solves a pseudo-open network (no external arrivals) subject to job flow conservation and the population constraints sum_i L(i,r)=N(r), using the GE-type fixed point of the open algorithm (Section 3.2) on class-composed flows. Stage 2 builds the ME product-form solution (3.8) from the Lagrangian coefficients of Stage 1, computes the normalising constant Z(N) by a multiclass convolution, and iterates the flow (work rate) equations until the class throughputs implied by the closed ME solution agree with those used to parametrise the building blocks.
      Parameters:
      M - number of queues (stations)
      R - number of job classes
      N - class populations [1 x R]
      mu - service rates [M x R]
      Cs - service scvs [M x R]
      P - routing probabilities, P[j][i].get(r,0) = p_ji,r
      c - servers per queue [M x 1]; Double.POSITIVE_INFINITY marks an IS queue; finite values must be 1
      refstat - reference station per class [R], 0-based; negative entries select the first station visited by the class
      options - algorithm options (tolerance, max iterations, verbosity)
      Returns:
      closed mean queue lengths, response times, pseudo-open flow scvs, per-station throughputs, closed utilizations, class throughputs at the reference stations and iteration count

    • me_cqn

      public static MeCqnResult me_cqn(int M, int R, Matrix N, Matrix mu, Matrix Cs, Matrix[][] P, Matrix c, int[] refstat, boolean[] insens, MeOqnOptions options)
      Maximum Entropy algorithm for Closed Queueing Networks with discipline-aware building blocks: insens[i] marks a station with an insensitive scheduling discipline (PS, LCFS-PR), solved with the product-form mean queue length instead of the FCFS GE formula.