Thanks Bob. <br>Now it's clear and I didn't know how the comm overhead could be.<br><br><div class="gmail_quote">On Tue, Jul 14, 2009 at 4:49 PM, bob ilgner <span dir="ltr"><<a href="mailto:bobilgner@gmail.com">bobilgner@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">hi, late at night here and forgot to include my timings also done on a 2 core PC:<br><br>1 Node(-n 1 i.e no comms)<br>
Master wall clock time 0.074823<br><br>2 Node(-n 2) <br>Master wall clock time 0.041086<br><br>4 Node(-n 4)<br>
Master Wall clock time 0.047954<br><br>8 Node(-n 8)<br>Master Wall clock time 0.062776<br><br>Which is what I would expect from a 2 core machine. I imagine that there is an increase for n>2 as we have additional comms overhead.<br>
<br><br><br>regards, bob<br><div><div></div><div class="h5"><br><div class="gmail_quote">On Tue, Jul 14, 2009 at 5:25 PM, bob ilgner <span dir="ltr"><<a href="mailto:bobilgner@gmail.com" target="_blank">bobilgner@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi WL,<br><br>Had a quick look at the code and note that you can NOT run this as a serial process, i.e. using only 1 node. You will need to change code for that. -n 1 does not run anything.<br><br>I've added a few printf comments and wall clock timings to show when the code is running, and not running, i.e. with -n 1 it does not run. Try it out with these markers and see how it works out.<br>
<br>Regards, bob<div><div></div><div><br><br><br><br><div class="gmail_quote">On Tue, Jul 14, 2009 at 4:00 PM, samantha lin <span dir="ltr"><<a href="mailto:wl8150@googlemail.com" target="_blank">wl8150@googlemail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi Bob,<br><br>Yes, that was the command I executed. Apart from '2' processes, I also tried a variety of numbers<br>and found '-n 1' is the fastest. I also made another similar C program without using the MPI libraries.<br>
The result was faster than '-n 1'. I didn't put a timing marks in the program; instead, I use 'time' command<br>prefixing the executing command, e.g. "time mpiexec -n 2 myprog". (should be okay?)<br>
I added my program to this email. Hope that helps.<br><br>Regards,<br>WL<div><div></div><div><br><div class="gmail_quote">On Tue, Jul 14, 2009 at 2:25 PM, bob ilgner <span dir="ltr"><<a href="mailto:bobilgner@gmail.com" target="_blank">bobilgner@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Hi WL,<br><br>What mpiexec command are you using to run your program ? i.e. "mpiexec -n 2 xprogy"<br>
where xprogy is the name of your program ?<br><br>What is your timing for the serial case and the timing for the multiple core case ? Is it a short program that you can list here so we can have a look at it ? Have you tried to place timing marks in the program to analyse what is happening ?<br>
<br>Regards, bob<div><div></div><div><br><br><div class="gmail_quote">On Tue, Jul 14, 2009 at 1:42 PM, samantha lin <span dir="ltr"><<a href="mailto:wl8150@googlemail.com" target="_blank">wl8150@googlemail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi bob,<br><br>It's running on a macbook pro(intel cpu 2 cores); just one laptop and filesystem <br>is local.<br><br>Regards,<br>WL<br><div class="gmail_quote">On Tue, Jul 14, 2009 at 5:37 AM, bob ilgner <span dir="ltr"><<a href="mailto:bobilgner@gmail.com" target="_blank">bobilgner@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div>Hi WL,</div>
<div> </div>
<div>What hardware are you running the mpiexec on and what sort of process is this ? A little description please.</div>
<div> </div>
<div>Regards, bob<br><br></div><div><div></div><div>
<div class="gmail_quote">On Tue, Jul 14, 2009 at 12:49 AM, samantha lin <span dir="ltr"><<a href="mailto:wl8150@googlemail.com" target="_blank">wl8150@googlemail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0px 0px 0px 0.8ex; padding-left: 1ex;">Hi there,
<div><br></div>
<div>I am new to MPICH. I just finished a test program which does</div>
<div>matrix multiplication. When I used 'mpiexec' to execute the </div>
<div>program, I found that the more processes, the longer it takes.</div>
<div>Not sure if the result is correct or not? Or I should do some </div>
<div>things more to improve it?</div>
<div><br></div>
<div>WL</div></blockquote></div><br>
</div></div></blockquote></div><br>
</blockquote></div><br>
</div></div></blockquote></div><br>
</div></div></blockquote></div><br>
</div></div></blockquote></div><br>
</div></div></blockquote></div><br>