Chris@19: Chris@19: Chris@19: FFTW MPI Transposes - FFTW 3.3.4 Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19: Chris@19:
Chris@19: Chris@19:

Chris@19: Next: , Chris@19: Previous: Other Multi-dimensional Real-data MPI Transforms, Chris@19: Up: Distributed-memory FFTW with MPI Chris@19:


Chris@19:
Chris@19: Chris@19:

6.7 FFTW MPI Transposes

Chris@19: Chris@19:

Chris@19: The FFTW's MPI Fourier transforms rely on one or more global Chris@19: transposition step for their communications. For example, the Chris@19: multidimensional transforms work by transforming along some Chris@19: dimensions, then transposing to make the first dimension local and Chris@19: transforming that, then transposing back. Because global Chris@19: transposition of a block-distributed matrix has many other potential Chris@19: uses besides FFTs, FFTW's transpose routines can be called directly, Chris@19: as documented in this section. Chris@19: Chris@19:

Chris@19: Chris@19: Chris@19: