Global handshaking among different component models. Each component will run on a set of nodes (processors). Processors in different sets are determined in "processors_map.in" file. They could overlap. Different components could be sequentially called. Most flexible.
You could get the source code as a tar file here or access the individual files as follows:
To compile and run on NERSC IBM SP:
===============================
1) Type "make", 1 executable will be created: master.
2) to run interactive job:
a) unset environment variable "MP_TASKS_PER_NODE"
b) type in ONE LINE:
poe master -nodes 5 -procs
9
-stdoutmode ordered -infolevel
2 > & output &
3) to run batch job:
type "llsubmit script".
To compile and run on other machines:
==============================
Please refer to the corresponding sections of README for MPH1.
Last modified on June 27, 2002.
Back to NERSC ACPI Homepage