<div class="gmail_quote">On Fri, Oct 22, 2010 at 00:22, Xuan Yu <span dir="ltr"><<a href="mailto:xxy113@psu.edu">xxy113@psu.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
Thanks, at first we are using sundials. Now, we found Petsc is more popular, and powerful of parallel and precondition. So, do you think it is the best way to use sundials option under Petsc framework?</blockquote></div>
<br>Yes, if Sundials is working well for your problem (in terms of error control and algorithmics for the solves) then it is a good option to use it through PETSc. You can use all PETSc preconditioners this way, and if your problem/machine/parameters changes, or you just want to try a different method, you don't have to write any code to use other TS implementations. There isn't a performance penalty for calling Sundials through PETSc and you get a lot more flexibility.<br>
<br>Jed<br>