Each executable resides on a set of SMP nodes.
Components within an executable may overlap on different nodes or processors.
Instances within an executable may not overlap on different nodes or processors.
MPH4 contains the following functionality:
The major addition over MPH3 is the multi-instance executable for ensemble simulation.
Send us email for questions and technical issues. If you find MPH useful, please also drop us an email so we could send you update and new release notice. A short paragraph of what's your application, which platform you are running and the processor allocation info is much appreciated.
Note: 02/10/05 new release: MPH V4.1
More explanation and source code as a tar file here (please see copyright info here) or access the individual files as follows:
For more information, please go to web page:
http://hpcrd.lbl.gov/SCG/acpi/MPH
Last modified Feb 10, 2005. Yun Helen He (yhe@lbl.gov), Chris Ding (chqding@lbl.gov)
Back to NERSC ACPI Homepage