Package jline.api.pfqn.ld

Class Pfqn_ljdfunKt

    • Nested Class Summary

      Nested Classes 
      Modifier and Type Class Description

    • Field Summary

      Fields 
      Modifier and Type Field Description

    • Constructor Summary

      Constructors 
      Constructor Description

    • Enum Constant Summary

      Enum Constants 
      Enum Constant Description

    • Constructor Detail

    • Method Detail

      • pfqn_ljdfun

         final static Matrix pfqn_ljdfun(Matrix nvec, List<Matrix> ljdscaling, List<Matrix> ljdcutoffs, Integer M, Matrix nservers)

        Evaluate limited joint-dependent (LJD) function

        Parameters:
        nvec - Per-class population vector n1, n2, ...
        ljdscaling - Linearized scaling tables indexed by station index, or null if none
        ljdcutoffs - Per-class cutoffs N1, N2, ...
        M - Number of stations
        nservers - Number of servers per station (optional)
        Returns:

        Scaling factor vector (M x 1)

      • ljd_linearize

         final static Integer ljd_linearize(Matrix nvec, Matrix cutoffs)

        Convert per-class population vector to linearized index (0-indexed)

        Parameters:
        nvec - Per-class population vector n1, n2, ...
        cutoffs - Per-class cutoffs N1, N2, ...
        Returns:

        0-indexed linearized index

      • ljd_delinearize

         final static Matrix ljd_delinearize(Integer idx, Matrix cutoffs)

        Convert linearized index back to per-class population vector

        Parameters:
        idx - 0-indexed linearized index
        cutoffs - Per-class cutoffs N1, N2, ...
        Returns:

        Per-class population vector n1, n2, ..., nK

      • ljcd_interpolate

         final static Double ljcd_interpolate(Matrix nvec, Matrix cutoffs, Matrix table, Integer K)

        Multi-linear interpolation for LJCD scaling tables.

        Performs multi-linear interpolation of a throughput value from an LJCD scaling table for non-integer population vectors. For K classes, interpolates between 2^K corner points of the hypercube containing the population vector.

        The interpolation uses floor/ceil corners in each dimension with fractional weights, producing smooth transitions between discrete table entries.

        Parameters:
        nvec - Continuous population vector n1, n2, ...
        cutoffs - Per-class cutoffs N1, N2, ...
        table - Linearized throughput table (1-D matrix indexed by ljd_linearize)
        K - Number of classes
        Returns:

        Interpolated throughput value