Tags: SCOREC/core
Tags
phasta_0.5 Ken ran an axisymmetric test with this version that was "better" than previous versions. renamed from phasta_5
phasta_0.4 Both the crossflow and the suchannel test cases ran successfully with this version. These phasta runs include also the verification of - a new UR implementation with BL snapping - a preload balancing strategy with Parma instead of global Zoltan before for instance tetrahedronization This version of Chef could generate a 3.1B tet mesh with the subchannel case on 1k Tukey cores after two levels of UR, and partition it to 8k parts on Mira (ALCF).A renamed from phasta_4
phasta_0.3 matching scheme for ilwork and iper figured out. several matching bugs fixed. also fixed phasta file subdirs and allow no outMeshFile. this version is important enough to tag since most future work will be in meshadapt and size fields, the phasta format side of things is stable. renamed from phasta_3
phasta_0.2 mostly stable implementation of "chef" The superior BC system seems to work, restart file parser is more robust. The main thing missing is parma, some BuildMapping details need fixing too. renamed from phasta_2
ph_test produces non-diverging phasta files renamed tag from phasta_1
PreviousNext