Minutes 20 March2006
Meeting held 11am Monday 20 March 2006
Present: Shane, Carey, Travis, Poul
Apologies: Andre, Chris, Karl, Greg, Peter
An audio recording of the meeting can be "downloaded":ftp://ftp.bioeng.auckland.ac.nz/cmiss/opencmiss/meeting-20.3.2006.ogg
Parallelization investigation
- Discussion from Karl and Travis
Travis spent a day investigating packages, in particular looking at Metis, has a multilevel partitioning algorithm which partitions an unstructured connectivity matrix. Looks like it could be some use to us, but needs further investigation.
Travis also looked at Diablo95, determined that it wasn't really useful to us at this point. GPL'ed, public version stripped down from what he used at LasAlomos.
Global Arrays is quite a different paradigm, and had build difficulties, and so haven't pursued it any further yet.
Travis noted he doesn't really have the time needed to do this properly.
Peter got the SUR grant so a cluster is coming.
Peter and Shane looking again at employing some more resource.
Travis proposes we need to move on getting the distributed field information, but this does not need fieldml to be completed.
Data structures.
Poul and Shane have had further FielML discussions and have resolved to make serialised examples this week.
- Examples from Poul
Development framework.
- What testing framework are we going to use?
- How are we going to develop code?