Simulations at extreme scale generate large amounts of data. As sufficiently large resources to handle the large data sets are generally available only in a distributed form, the complete tool chain deployed to obtain simulation results needs to be usable on distributed systems. In this work we present the APES framework which provides such a tool chain for mesh based simulations. We show the scaling behavior of two solvers from that framework, both of which are capable to deal with complex geometries but target different application domains. The scalability is investigated on the SuperMUC system installed at the Leibniz-Rechenzentrum in Munich including an analysis of single node behavior and the sustained performance.
IOS Press, Inc.
6751 Tepper Drive
Clifton, VA 20124
Tel.: +1 703 830 6300
Fax: +1 703 830 2300 firstname.lastname@example.org
(Corporate matters and books only) IOS Press c/o Accucoms US, Inc.
For North America Sales and Customer Service
West Point Commons
Lansdale PA 19446
Tel.: +1 866 855 8967
Fax: +1 215 660 5042 email@example.com