On Fri, Feb 19, 2016 at 5:39 PM, John Peterson <jwpeter...@gmail.com> wrote:
> > > On Fri, Feb 19, 2016 at 2:47 PM, David Knezevic < > david.kneze...@akselos.com> wrote: > >> If I do a solve with different numbers of processes, and write out Exodus >> files in each case, and then use ncdump to compare at the ASCII version of >> each file, I can see that the two files are the same (up to machine >> precision). >> >> I initially assumed that the exodus data would be partition dependent, but >> I guess that isn't the case? Do we reorder to node numbering to ensure >> that >> the output is partition-independent. I know we do that for xdr files >> (EquationSystems::write() takes a partition_agnostic argument), but I >> didn't realize we do it for other formats as well? If someone can shed >> light on this, that'd be appreciated. >> > > I believe the Exodus files may always be the same since, for parallel > meshes, they get passed through a MeshSerializer object before being > written... > I'm using a SerialMesh, though. Even in the SerialMesh case I would have thought the exodus file would depend on the numbering of nodes and elems, which is partition dependent, but apparently not? ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 _______________________________________________ Libmesh-users mailing list Libmesh-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-users
