SGpp / DisCoTec

MPI-based code for distributed HPC simulations with the sparse grid combination technique
https://sparsegrids.org/
GNU Lesser General Public License v3.0
8 stars 6 forks source link

[JOSS] State of the field #130

Open EmilyBourne opened 3 weeks ago

EmilyBourne commented 3 weeks ago

Regarding https://github.com/openjournals/joss-reviews/issues/7018 "State of the field: Do the authors describe how this software compares to other commonly-used packages?"

The paper does not seem to mention other packages. Are the authors aware of any other packages for the sparse grid combination technique? Or other packages which could solve similar higher-dimensional problems to which this package could be compared?

freifrauvonbleifrei commented 3 weeks ago

I can quote from my thesis:

The Sparse Grids Matlab Kit [126, 127] by Piazzola and Tamellini is targeted towards teaching and uncertainty quantification with the CT. It provides a user friendly MATLAB interface for the CT, and also dimensional adaptivity [53] for nested and non-nested sequences of component grid collocation points. The code’s parallelism is limited to shared memory only, as it does not target HPC applications. The same holds true for the sparseSpACE code [120], which provides an implementation for automatic spatial adaptivity in the CT for integration [121], interpolation, uncertainty quantification, machine learning methods, and PDE solvers.

I am not aware of something that targets distributed-memory combination technique.

@EmilyBourne @jakelangham what is your opinion, should we rather say "as far as we know there are no other codes for large-scale combination technique" or mention the above codes along with the limited comparability?

EmilyBourne commented 3 weeks ago

In my opinion it would be good to mention the codes that are mentioned in your thesis. You can also say that there are no distributed-memory codes. This should clearly illustrate the gap that DisCoTec aims to fill