<div dir="ltr"><div class="gmail_extra">On Wed, May 8, 2013 at 7:30 AM, Jed Brown <span dir="ltr"><<a href="mailto:jedbrown@mcs.anl.gov" target="_blank">jedbrown@mcs.anl.gov</a>></span> wrote:<br><div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">Matthew Knepley <<a href="mailto:knepley@gmail.com">knepley@gmail.com</a>> writes:<br>
<br>
> On Tue, May 7, 2013 at 10:53 PM, Jed Brown <<a href="mailto:jedbrown@mcs.anl.gov">jedbrown@mcs.anl.gov</a>> wrote:<br>
><br>
>> This thread is complaining about DMDA costing multiple vectors worth<br>
>> of memory.<br>
>><br>
>> <a href="https://groups.google.com/d/msg/claw-dev/JwIjL5e48No/NOizc6i88gkJ" target="_blank">https://groups.google.com/d/msg/claw-dev/JwIjL5e48No/NOizc6i88gkJ</a><br>
>><br>
>> We currently set a lot of stuff up eagerly so that it can be accessed<br>
>> using non-collective accessors. When a Krylov method is used or a<br>
>> matrix is assembled, that stuff is in the noise, but for explicit<br>
>> methods, it can be the limiting factor for problem size. Should we<br>
>> do something about this?<br>
>><br>
><br>
> I would at least like to know what it is, and how the interface would<br>
> have to change. This can't be all scatter memory.<br>
<br>
</div>At the very least, it's global-to-local scatter, local-to-local scatter,<br>
and local-to-global mapping (scalar and block).<br>
</blockquote></div><br>Well, the scatter operations are collective, so making creation lazy here seems fine.</div><div class="gmail_extra">I have no problem forcing a setup of local-to-global since I do not think many people</div>
<div class="gmail_extra">use it directly.</div><div class="gmail_extra"><br></div><div class="gmail_extra"> Matt<br clear="all"><div><br></div>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener
</div></div>