Package jline.api.nc

Class Me_oqn_blk

java.lang.Object
jline.api.nc.Me_oqn_blk

public final class Me_oqn_blk extends Object
  • Field Details

    • RULE_LOSS

      public static final int RULE_LOSS
      Drop rule marker: a blocked job is discarded.
      See Also:
    • RULE_BAS

      public static final int RULE_BAS
      Drop rule marker: a blocked job is held in the upstream server.
      See Also:
  • Method Details

    • me_oqn_blk

      public static MeOqnBlkResult me_oqn_blk(int M, Matrix lambda0, Matrix Ca0, Matrix mu, Matrix Cs, Matrix P, Matrix c, Matrix N, int[] blockrule)
    • me_oqn_blk

      public static MeOqnBlkResult me_oqn_blk(int M, Matrix lambda0, Matrix Ca0, Matrix mu, Matrix Cs, Matrix P, Matrix c, Matrix N, int[] blockrule, MeOqnOptions options, double damping)
      Maximum Entropy algorithm for a single-class open network with finite buffers.
      Parameters:
      M - number of stations
      lambda0 - external arrival rates [M x 1]
      Ca0 - external interarrival scvs [M x 1]
      mu - service rates [M x 1]
      Cs - service scvs [M x 1]
      P - routing probabilities [M x M]; a row sum below one sends the residual flow out of the network
      c - servers per station [M x 1], infinite marking an IS
      N - buffer capacity per station [M x 1] in jobs, in service included, infinite marking an unbounded buffer
      blockrule - policy at each finite buffer, RULE_LOSS or RULE_BAS
      options - tolerance, iteration cap and verbosity
      damping - relaxation weight applied to the blocking probabilities
      Returns:
      station metrics and the flow moments at the fixed point