On Mon, Mar 7, 2011 at 8:43 AM, Gaurish Telang <span dir="ltr"><<a href="mailto:gaurish108@gmail.com">gaurish108@gmail.com</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;">
Oh thank you, this was helpful. I am interested in iterative solvers, so what is the minimum matrix size you<br>think that strong scalability will show up for such methods? <br></blockquote><div><br></div><div>Such blanket predictions are not worth much for strong scaling since they depend on the</div>
<div>architecture, interconnect, etc. What is most important is to understand the timing output</div><div>in -log_summary and see what is not scaling correctly. Dave pointed out that linear iterations</div><div>must also scale correctly.</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;">
<br><div class="gmail_quote">On Mon, Mar 7, 2011 at 9:38 AM, Matthew Knepley <span dir="ltr"><<a href="mailto:knepley@gmail.com" target="_blank">knepley@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
<div>On Mon, Mar 7, 2011 at 8:20 AM, Gaurish Telang <span dir="ltr"><<a href="mailto:gaurish108@gmail.com" target="_blank">gaurish108@gmail.com</a>></span> wrote:<br></div><div class="gmail_quote"><div>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
Hi,<div><br></div><div>I have been testing PETSc's scalability on clusters for matrices of sizes 2000, 10,000, uptill 60,000.</div></blockquote><div><br></div></div><div>1) These matrices are incredibly small. We usually recommend 10,000 unknowns/process for weak scaling. You</div>
<div> might get some benefit from a shared memory implementation on a multicore.</div><div><div> </div><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
<div>All I did was try to solve Ax=b for these matrices. I found that the solution time dips if I use upto 16 or 32 processors. However for a larger number of processors however the solution time seems to go up rather than down. IS there anyway I can make my code strongly scalable ?</div>
</blockquote><div><br></div></div><div>2) These are small enough that direct factorization should be the fastest alternative. I would try UMFPack, SuperLU, and MUMPS.</div><div><br></div><div> Matt</div><div>
<div> </div><blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204, 204, 204);padding-left:1ex">
<div>I am measuring the total time (sec) and KSP_SOLVE time in the -log_summary output. Both times show the same behaviour described above. </div><div><br></div><font color="#888888"><div>Gaurish</div></font></blockquote>
</div></div><font color="#888888">-- <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<br>
</font></blockquote></div><br>
</blockquote></div><br><br clear="all"><br>-- <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<br>