I am looking for a realistic phonon calculation, to be used as a benchmark to identify bottlenecks and improve performances on parallel machines. The ideal benchmark should have about 100 atoms (let's say in the 50- to 200-atom range) and run on up to a few hundreds processors. If you have something like this and you are willing to provide it for testing, please contact me
Paolo --- Paolo Giannozzi, IOM-Democritos and University of Udine, Italy
