### Table 19.1: Tetrahedral Mesh Smoothing Variables 3n

2004

Cited by 2

### Table 2: Statistics on coarsening of tetrahedral mesh in a prism

1999

Cited by 1

### Table 5.3: Properties of some tetrahedral meshes used in the study.

### Table 4: Partitioning of a tetrahedral mesh between concentric spheres.

### Table I. Relation between number of entities in a mesh. Tetrahedral mesh T Hexahedral mesh H

### Table 3.1: This table shows performance gures for our technique on an AMD X2 5000+. The columns from left to right contain: Model name, gure number in the paper, number of input mesh vertices, number of tet mesh vertices/tetrahedra; thereafter, timings are given for pre-factorization of the left-hand side matrix of Eq. (2.1), precalculation of pose transfer coe cients, deformation transfer to input mesh; the rightmost column gives frames per second during interactive editing of the tetrahedral mesh. All other timings are given in seconds.

2007

### Table 6.4. Fichera corner: results for tetrahedral mesh (columns 2 and 3) and hexahedral mesh (column 4)

### Table 1: Results for various graph representations with a FORGE3 mesh consisting of 231846 tetrahedral elements and 49666 nodes and divided in 4 subdomains. The initial mesh distribution is heavily unbalanced (about half of the mesh resides on one processor). ParMETIS has been called with 1%, 2%, and 5% imbalance, PJOSTLE with local and global options.

"... In PAGE 11: ...rocessors). The re-partitioning time is the time needed by the DRAMA library. The relative execution time of ParMETIS is for the nodal graph and the dual graph about 25% and 40%, respectively. The results in Table1 show that excellent re-partitioning is obtained for both the nodal and the dual graph, despite the fact that the node-based calculation costs and the communication latency are neglected in the dual graph representation. The nodal graph representation clearly leads to a good solution in much less time and with less memory usage.... In PAGE 12: ...DRAMA library costs for Forge3 meshes were given in Table1 , for partitioning with both nodal and element graphs. Since the natural option for the PAM-CRASH code is the use the combined graph, and because the PAM-CRASH meshes have a very different structure, especially since virtual elements are employed for the contact-impact cost capture (with non-local communications links), additional performance tests with the DRAMA library are reported here.... ..."

### Table 1: Results for various graph representations with a FORGE3 mesh consisting of 231846 tetrahedral elements and 49666 nodes and divided in 4 subdomains. The initial mesh distribution is heavily unbalanced (about half of the mesh resides on one processor). ParMETIS has been called with 1%, 2%, and 5% imbalance, PJOSTLE with local and global options.

"... In PAGE 11: ...rocessors). The re-partitioning time is the time needed by the DRAMA library. The relative execution time of ParMETIS is for the nodal graph and the dual graph about 25% and 40%, respectively. The results in Table1 show that excellent re-partitioning is obtained for both the nodal and the dual graph, despite the fact that the node-based calculation costs and the communication latency are neglected in the dual graph representation. The nodal graph representation clearly leads to a good solution in much less time and with less memory usage.... In PAGE 12: ...DRAMA library costs for Forge3 meshes were given in Table1 , for partitioning with both nodal and element graphs. Since the natural option for the PAM-CRASH code is the use the combined graph, and because the PAM-CRASH meshes have a very different structure, especially since virtual elements are employed for the contact-impact cost capture (with non-local communications links), additional performance tests with the DRAMA library are reported here.... ..."