On Sat, 27 Feb 2016, li....@siat.ac.cn wrote: > For the parallel implementation of adaptive refinement in Libmesh, > I am using "--enable-parmesh". > it seems that after one adaptive refinement step, the changed mesh is not > redistributed over all processors, > some processors may have many refined mesh nodes but others may have a few > coarsed nodes, > so it is not load balancing, am I right?
We should be doing load balancing even with ParallelMesh as of last August, in https://github.com/libMesh/libmesh/pull/625 We rely on ParMETIS for the repartitioning, though, and if the load imbalance isn't too bad I don't know under what conditions it will give us a redistribution vs a no-op. --- Roy ------------------------------------------------------------------------------ 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
