Thursday, 27 November 2008

MPI enabled

Enabled MPI in Manchester using YAIM and the recipe from Stephen Childs I found in the links below:


1) The documentation will move to probably the YAIM official pages

2) The location of the gip files is now under /opt/glite not /opt/lcg

3) Scripts DO interfere with the current setup on the WNs if run on their own so you need to reconfigure the whole node (I made the mistake of running only MPI_WN). On the CE instead it's enough to run MPI_CE.

4) MPI_SUBMIT_FILTER variable in site-info.def is not documented (yet). It enables the section of the scripts that rewrites torque submit_filter script that allocates the correct number of CPUs

5) Yaim doesn't publish MPICH (yet?) so I had to add the following lines

GlueHostApplicationSoftwareRunTimeEnvironment: MPICH
GlueHostApplicationSoftwareRunTimeEnvironment: MPICH-1.2.7

to /opt/glite/etc/gip/ldif/static-file-Cluster.ldif manually.


Stephen Childs said...

The MPI_CE bit should add tags - I think you need to run at least the config_gip bit of YAIM on the CE as well. The problem is that I ended up having to develop the YAIM functions despite not using YAIM ...

Hopefully a new maintainer fluent in YAIM can be found.

Stephen Childs said...

Yes it adds entries to the variable CE_RUNTIMEENV so you would need to run config_gip after CE_MPI.

Alessandra said...

Ok, thanks.