There are two possible ways to obtain the ELPA library:
a) tar-ball archive:
We are beginning to offer pre-build packages for different Linux distributions.
At the moment, packages for Debian, openSUSE 13.1, openSUSE Factory, openSUSE Tumbleweed, and Fedora are available, which you can obtain here. This packages are marked as "unstable", since they
are offered by us and not (yet) in the official distribution repositories.
If you want to contribute to the ELPA library (e.g. put your improvements or ports to other architectures (and so forth) in the official ELPA version) the access to the git server is the best way to go.
Note that, in the git repository you can find various development branches, recent changes, etc., which are still untested and thus not production ready! Feel free to use them already, however, a guarantee of the correctness of the code can
not be provided. For users, who prefer to simply use ELPA without the urge to participate in the development of the ELPA package, we thus recommend the tar-ball archive (see above) to obtain the latest releases, which have been tested
to a fair extend.
Note, that the previous git server of the ELPA git repository at elpa-lib.fhi-berlin.mpg.de is not used anymore!
To obtain ELPA sources from the new git repository, please go to https://gitlab.mpcdf.mpg.de/elpa/elpa and clone from there. if you have already a clone of the ELPA source from the previous server, please do
git remote set-url origin https://gitlab.mpcdf.mpg.de/elpa/elpa.git
in your local checkout to switch to the new server.
Licencing of ELPA:
The license for ELPA is the LGPL . That said, for the time being we kindly ask you to cite the original ELPA library in your work whenever it is useful, and that you actively contribute back any amendments / improvements that you may have.
To cite ELPA, please use the following references:
T. Auckenthaler, V. Blum, H.-J. Bungartz, T. Huckle, R. Johanni, L. Krämer, B. Lang, H. Lederer, and P. R. Willems: "Parallel solution of partial symmetric eigenvalue problems from electronic structure calculations" Parallel Computing 37, 783-794 (2011). A preprint (pdf) can be found here.
Andreas Marek, Volker Blum, Rainer Johanni, Ville Havu, Bruno Lang, Thomas Auckenthaler, Alexander Heinecke, Hans-Joachim Bungartz, and Hermann Lederer, The ELPA Library - Scalable Parallel Eigenvalue Solutions for Electronic Structure Theory and Computational Science The Journal of Physics: Condensed Matter 26, 213201 (2014). http://stacks.iop.org/0953-8984/26/213201
If you run into problems using the ELPA library, we recommend the following steps:
2.) In the ELPA library software, we provide an additional email address to contact the developers of the ELPA library.