Welcome to dtFFT Documentation

Usage Guide

dtFFT (DataTyped Fast Fourier Transform) is a high-performance library designed for parallel data transpositions and optional Fast Fourier Transforms (FFTs) in multidimensional computing environments.

Initially developed to perform zero-copy transpositions using custom MPI datatypes on CPU clusters, dtFFT leverages these efficient data structures to minimize memory overhead in distributed systems. However, as the demand for GPU-accelerated computing grew, it became clear that MPI datatypes were suboptimal for GPU workflows. To address this, a parallel approach was crafted for GPU execution: instead of relying on custom datatypes, dtFFT compiles CUDA kernels at runtime using nvrtc, tailoring them to the specific plan and data layout.

The library supports MPI for distributed systems and GPU acceleration via CUDA, integrating seamlessly with external FFT libraries such as FFTW3, MKL DFTI, cuFFT, and VkFFT, or operating in transpose-only mode.

Whether you’re working on CPU clusters or GPU-enabled nodes, dtFFT provides a flexible and efficient framework for scientific computing tasks requiring large-scale data transformations.

This documentation covers the essentials of using, building, and extending dtFFT. Explore the sections below to get started.

Getting Started

To begin using dtFFT:

1. Build the Library: Follow the instructions in Building the Library to compile dtFFT with your desired features (e.g., CUDA or FFTW3 support).

2. Configure Runtime: Set environment variables as needed (see Environment Variables) to tweak logging or datatype selection.

3. Use the Library: Refer to the Usage Guide for step-by-step examples of creating plans, allocating memory, and executing transformations.

Detailed API specifications are available in the Fortran, C, and C++ sections.

Contributing

Feedback, bug reports, and contributions are welcome! Please submit issues or pull requests via the project’s repository. For API-specific details, consult the respective language sections.

Table of Contents

• Building the Library
  • Prerequisites
  • Configuration Options
  • Building the Library
  • Integration with CMake Projects
• Usage Guide
  • Error Handling and Macros
  • Plan Creation
  • Plan Creation Using Pencil Decomposition
  • Memory Allocation
  • Plan properties
  • Plan Execution
  • Plan Finalization
  • Complete Example
• Fortran API Reference
  • Error Codes
  • Basic types
  • Version handling
  • Abstract plan
  • Real-to-Real plan
  • Complex-to-Complex plan
  • Real-to-Complex plan
• C API Reference
  • Predefined Macros
  • Enumerators
  • Types
  • Functions
  • Plan constructors
  • Plan destructor
  • Memory allocation
  • Plan execution
  • Plan information
• C++ API Reference
  • Predefined Macros
  • Enumerators
  • Functions
  • Structs
  • Classes
• Environment Variables
  • DTFFT_ENABLE_LOG
  • DTFFT_MEASURE_WARMUP_ITERS
  • DTFFT_MEASURE_ITERS
  • DTFFT_PLATFORM
  • DTFFT_BACKEND
  • DTFFT_NCCL_BUFFER_REGISTER
  • DTFFT_ENABLE_Z_SLAB
  • DTFFT_ENABLE_MPI
  • DTFFT_ENABLE_NCCL
  • DTFFT_ENABLE_NVSHMEM
  • DTFFT_ENABLE_PIPE
  • DTFFT_ENABLE_KERNEL_OPTIMIZATION
  • DTFFT_CONFIGS_TO_TEST
  • DTFFT_FORCE_KERNEL_OPTIMIZATION
  • MPI Datatype Selection Variables
• Benchmark Overview
  • Test Environment
  • Strong Scaling Analysis
  • Weak Scaling Analysis
  • Detailed Performance Data
  • Conclusions