Parallel Three-Dimensional Fast Fourier Transforms

Parallel Three-Dimensional Fast Fourier Transforms, dubbed P3DFFT, is a library for computational computing in a wide range of sciences, such as physics, climatology, chemistry.

This project was developed at SDSC by Dmitry Pekurovsky as a product of a Strategic Applications Collaborations (SAC) project.

Features:

• Parallel implementation with 2D data decomposition, overcoming an important limitation to scalability of other 3D FFT libraries implementing 1D, or slab, decomposition.
• Optimized for parallel communication and single-CPU performance.
• Built on top of well-optimized and flexible 1D FFT libraries.

What is new in this release:

• This version adds assorted bugfixes and enhancements.

What is new in version 2.4:

• Several small bugs were fixed. When the STRIDE1 option is used, the Fourier-space complex array is now stored in the form (Z,Y,X) rather than (Z,X,Y).
• The C interface is now more robust, and C++ support is available (in beta phase).
• A new installation process based on the GNU configure tool is used to facilitate installation on a variety of platforms.
• The documentation was updated to clarify available options.
• New examples were added.
• The project was relocated to Google Project Hosting.

What is new in version 2.3.1:

• An option has been added to overwrite the (complex) input array when using the complex-to-real transform with FFTW.
• An option has been added to use data layout using unit-stride on output of the real-to-complex routine, i.e. (Z,X,Y) ordering.

What is new in version 2.3:

• A C interface has been added.
• More examples have been added: in-place transform and power spectrum calculation.
• Performance improvements have been made.
• Ghost cell support was added for nearest-neighbor communication.
• Several bugs have been fixed.