Class ColumnSixStepFNTStrategy

  • All Implemented Interfaces:
    Parallelizable, NTTStrategy
    Direct Known Subclasses:
    IntAparapiSixStepFNTStrategy, LongAparapiSixStepFNTStrategy

    public class ColumnSixStepFNTStrategy
    extends SixStepFNTStrategy
    Six-step NTT implementation that processes the data in the columns of the matrix.

    This transform only works together with an NTTStepStrategy implementation that processes the data in columns instead of rows and a MatrixStrategy implementation that can transpose the data.

    The data size should be sufficiently large to meet the parallelization needs of the GPU. The GPU global size i.e. the number of columns in the data matrix should be at least 1024.

    Since:
    1.8.3
    Version:
    1.8.3
    Author:
    Mikko Tommila
    • Constructor Detail

      • ColumnSixStepFNTStrategy

        public ColumnSixStepFNTStrategy​(NTTStepStrategy stepStrategy,
                                        MatrixStrategy matrixStrategy)
        Basic constructor.
        Parameters:
        stepStrategy - A step strategy that can process data in columns.
        matrixStrategy - A matrix strategy that can process the data.
    • Method Detail

      • transposeInitial

        protected void transposeInitial​(ArrayAccess arrayAccess,
                                        int n1,
                                        int n2,
                                        boolean isInverse)
        Description copied from class: SixStepFNTStrategy
        The initial transpose of the forward transform, or the final transpose of the inverse transform, to transpose the columns of the matrix to be rows. This step is needed in the six-step algorithm but is omitted in the four-step algorithm.
        Overrides:
        transposeInitial in class SixStepFNTStrategy
        Parameters:
        arrayAccess - Accessor to the matrix data. This data will be transposed.
        n1 - Number of rows in the matrix.
        n2 - Number of columns in the matrix.
        isInverse - true if an inverse transform is performed, false if a forward transform is performed.
      • transposeMiddle

        protected void transposeMiddle​(ArrayAccess arrayAccess,
                                       int n1,
                                       int n2,
                                       boolean isInverse)
        Description copied from class: SixStepFNTStrategy
        The second transpose of either the forward or inverse transform. Normally this step is always required as the four-step algorithm only transforms columns of the matrix and the six-step algorithm transforms only rows.
        Overrides:
        transposeMiddle in class SixStepFNTStrategy
        Parameters:
        arrayAccess - Accessor to the matrix data. This data will be transposed.
        n1 - Number of rows in the matrix.
        n2 - Number of columns in the matrix.
        isInverse - true if an inverse transform is performed, false if a forward transform is performed.
      • multiplyElements

        protected void multiplyElements​(ArrayAccess arrayAccess,
                                        int rows,
                                        int columns,
                                        long length,
                                        long totalTransformLength,
                                        boolean isInverse,
                                        int modulus)
        Description copied from class: SixStepFNTStrategy
        Multiply each matrix element by a power of the n:th root of unity.
        Overrides:
        multiplyElements in class SixStepFNTStrategy
        Parameters:
        arrayAccess - The memory array to multiply.
        rows - The number of rows in the arrayAccess to multiply.
        columns - The number of columns in the matrix (= n2).
        length - The length of data in the matrix being transformed.
        totalTransformLength - The total transform length, for the scaling factor. Used only for the inverse case.
        isInverse - If the multiplication is done for the inverse transform or not.
        modulus - Index of the modulus.