ecTrans is a library for performing efficient and scalable spectral transformations. It is used for transforming fields from a grid point space on the sphere (e.g. latitude-longitude) to a spectral space based on spherical harmonics (for global transformations) or bifourier harmonics (for limited area transformations), which constitutes a direct transform. A corresponding inverse transform can also be performed. A transform consists of a Fourier transform in the longitudinal direction and either a Legendre transform (global) or another Fourier transform (limited area) in the latitudinal direction. ecTrans can also operate on fields which are distributed across separate MPI tasks and performs the necessary communication to ensure all data needed for a particular transform are resident on a local task.
After co-development as part of the Integrated Forecasting System (IFS) atmospheric model of the European Centre for Medium-Range Weather Forecasts for several decades, ecTrans became a standalone software package in 2022. It constitutes one of the most important and expensive parts of the IFS and neatly encapsulates both computational and communicational paradigms and bottlenecks exhibited by the IFS model as a whole.
ecTrans primarily targets conventional CPU platforms, requiring FFTW- and BLAS-implementing libraries. It can also operate efficiently on GPU accelerators making use of offloading directives (either OpenACC or OpenMP) and vendor library routines (cuBLAS/cuFFT or hipBLAS/hipFFT). ecTrans performs efficiently and stably on Nvidia platforms but is currently less mature on AMD platforms.
To learn more about ecTrans, please consult the documentation (which is under construction).
ecTrans is distributed under the Apache License Version 2.0.
See LICENSE file for details.
Please consult the documentation.
Please report bugs using a GitHub issue. Support is given on a best-effort basis by package developers.
Contributions to ecTrans are welcome. In order to do so, please open a GitHub issue where a feature request or bug can be discussed. Then create a pull request with your contribution. All contributors to the pull request need to sign the contributors license agreement (CLA).