Kenneth E. Jansen

kjansen@cc014: /projects/TRAssembly_2/kjansen/Models/BoeingBump/matchedReaderTest/1B $ mpirun -np 32 --hostfile $COBALT_NODEFILE /projects/TRAssembly_2/SCOREC-core/build-matchedReader/test/matchedNodeElmReader box_WMLES_1B.cnn box_WMLES_1B.crd NULL mod mesh/ numElms 1000000000 numVerts 1003003001 isMatched 0 seconds to create mesh 3589.092 mesh verified in 158.253273 seconds...

Note all of the above is with NULL passed for matching. Please let me know when you think the code is ready to test matching. Also, it looks like you...

I think we are facing a tradeoff between IO bottlenecks at large process/per node and the scalability of the mds element creation. To test the extreme of this I tried...

Since this approach appears promising, I am going to take some time here to plan the remaining steps to putting this into a production run. I will put tentative assignments...

This looks good. I am going to add to task 5 what you probably already know goes with it -- modifying your code to write the model classification and boundary...

Unless Cameron says otherwise, I suggest you follow the order of the connectivity array. Model vertices start with 0 at the "origin" which we can define as minx, miny, minz...

Cameron, if it makes your matching work easier, It would be easy for us to restrict ourselves to cases where there was only on-part matching. This is as easy as...

We should also plan for which stage we will switch the file formats (e.g., add the header with the array shape and ditch the ID column) .

Since, in the above, there is a general rule that classification is always set by the classification of the node on the closure with the highest dimension classification, I think...

What broke that version was that, when you were testing 2D you added the reading of three lines at the top of the .cnn file 1 1 nsd numel nvtxPerElement...