Fig. 1
Design of the AMUSE interface. This diagram represents the way in which a community code (“code”) is accessed from the AMUSE framework. The code has a thin layer of interface functions in its native language which communicate through an MPI message channel with the Python host process. On the Python side the user script (“AMUSE simulation script”) only accesses generic calls (“setup”, “evolve” etc.) to a high level interface. This high level interface calls the low level interface functions, hiding details about units and the code implementation. The communication through the MPI channel does not interfere with the code’s own parallelization because the latter has its own MPI_WORLD_COMM context.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.