<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Wed, May 6, 2015 at 8:29 AM, Jed Brown <span dir="ltr"><<a href="mailto:jed@jedbrown.org" target="_blank">jed@jedbrown.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">Matthew Knepley <<a href="mailto:knepley@gmail.com">knepley@gmail.com</a>> writes:<br>
> I think the idea is to be explicit. I would just use something like<br>
><br>
> # Vecs * Vec size * 8 bytes/double + <same for Mat><br>
><br>
> and forget about small stuff. This model is uncached,<br>
<br>
</span>This is for a perfect cache model -- each byte of the data structures<br>
needs to be fetched from DRAM only once.<br></blockquote><div><br></div><div>I meant uncached, in which you count # Vecs for any operation you are</div><div>doing. If you count # Vecs for the whole program, then you have perfect cache.</div><div><br></div><div> Matt</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Make sure that "# Vecs" actually means the number that are accessed.<br>
For example, GMRES accesses an increasing number of vectors on each<br>
iteration (until a restart).<br>
<div class="HOEnZb"><div class="h5"><br>
> and you just state that up front. Someone else can take the data and<br>
> produce a more nuanced model.<br>
><br>
> You can use the Flop counts (not Flop/s) from PETSc, and get a rough<br>
> estimate<br>
> of flop/byte ratio.<br>
<br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">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></div>