From: Timo Heister Date: Tue, 2 Jun 2020 21:29:21 +0000 (-0400) Subject: fix large scale Stokes section, remove todo X-Git-Url: https://gitweb.dealii.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=85e1c769c318997d1f8f0bd43e939ec37ee74eba;p=release-papers.git fix large scale Stokes section, remove todo --- diff --git a/9.2/paper.tex b/9.2/paper.tex index 0fc1b2a..a12cad7 100644 --- a/9.2/paper.tex +++ b/9.2/paper.tex @@ -535,17 +535,12 @@ Stokes system and achieved weak and strong scaling up to 114K MPI ranks with up \dealii{} and required various optimizations and fixes on top of the ones mentioned above: (i) Bug fixes to concurrent point to point communications. -(ii) Fixes multigrid transfer with adaptive refinement and more than \num{4e9} unknowns. +(ii) Fixes to multigrid transfer with adaptive refinement and more than \num{4e9} unknowns. (iii) Fixes to index sets in block indices with more than \num{4e9} unknowns. (iv) Fixes to computations with more than \num{4e9} active cells. (v) Implementation of IDR(s) solvers to reduce memory overhead. For more details, see \cite{clevenger_stokes19}. - -\todo[inline]{Timo added the paragraph above. Thoughts on how much we should mention (none of this worked with 9.1)?} - -\todo[inline]{Martin: I think (ii) worked in 9.1/9.0/8.5, and (iv) definitely worked in the uniform case for 8.5, see Fig 1 in that paper, data line `8B cells'. We just happened to break things in between. But ``fixes'' is a good wording here because it doesn't exclude that some problems with those configurations have been solved before. So I like the text.} - %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% \subsection{Better support for parallel $hp$-adaptive algorithms} \label{subsec:hp}