Difference between revisions of "FFTW"
Line 10: | Line 10: | ||
The library contains parallel versions using either of [[pthreads]], [[OpenMP]] or [[MPI]]. Besides the current version 3.x, there is the older version 2.x, which have different application interfaces. It is not recommended to use version 2.x for new developments, however it is still widely used in legacy code. | The library contains parallel versions using either of [[pthreads]], [[OpenMP]] or [[MPI]]. Besides the current version 3.x, there is the older version 2.x, which have different application interfaces. It is not recommended to use version 2.x for new developments, however it is still widely used in legacy code. | ||
+ | |||
+ | == License == | ||
+ | {{show license}} | ||
== Availability == | == Availability == |
Revision as of 14:34, 14 September 2012
FFTW (Fastest Fourier Transform in the West) is a high performance library to calculate discrete Fourier transformations. The library is freely available and typically easy to install.
The software is self optimising, trying several algorithms at run time and choosing the best performing one. To achieve its high computational speed latest version supports the SSE and AVX instructions of the latest x86-processors.
The library contains parallel versions using either of pthreads, OpenMP or MPI. Besides the current version 3.x, there is the older version 2.x, which have different application interfaces. It is not recommended to use version 2.x for new developments, however it is still widely used in legacy code.
License
License: Free.
Availability
Resource | Centre | Description |
---|---|---|
Abisko | HPC2N | capability resource of 153 TFLOPS with full bisectional infiniband interconnect |
Alarik | LUNARC | throughput cluster resource of 40 TFLOPS |
Aurora | LUNARC | throughput/general purpose cluster resource |
Platon | LUNARC | throughput cluster resource of 26 TFLOPS |