Physics and mathematics are at the heart of most simulation codes. There are normally parts of the code where external physics and maths libraries can be used and these parts of the code can quickly become a bottleneck if the wrong library or method is used.

Effective parallel linear algebra solvers with good parallel scaling properties are complex to code but frequently required in CCP codes. We will accumulate information and guidance on these solvers to meet the needs of the CCPs/HECs on our linear algebra page. By improving the underlying methods, this can dramatically improve the execution times and energy consumption values for the computational simulation.

Fast Fourier Transforms are another common kernel in CCP and HEC codes. The 2012 PRACE survey of FFT codes focussed on MPI codes. We will update the guidance from this survey and also consider the requirements of codes that use other parallelism methods or have specific license needs. Further details, including a list of FFT libraries and their attributes, are available via our FFT page.