<div dir="ltr">We should generally not be using wall clock timing for performance measurement.</div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Nov 15, 2017 at 11:35 AM, Matt Brown <span dir="ltr"><<a href="mailto:matt.brown@kitware.com" target="_blank">matt.brown@kitware.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div>Do we have or can we decide on a standard practice for logging processing time debug. I have gotten into the habit of wrapping the _step in my processes with<br><br>kwiver::vital::wall_timer timer;<br>timer.start();<br>...<br>timer.stop();<br>double elapsed_time = timer.elapsed();<br>LOG_DEBUG( logger(), "Elapsed time detecting objects: " << elapsed_time );<br><br></div>Some questions:<br><ul><li>Should this be TRACE or DEBUG?</li><li>Should we should wall and/or CPU timing?</li><li>Will all of this be superseded by a more-unified framework that does this automatically?</li></ul><p>Best,<br>Matt<br></p></div>
<br>______________________________<wbr>_________________<br>
Kwiver-users mailing list<br>
<a href="mailto:Kwiver-users@public.kitware.com">Kwiver-users@public.kitware.<wbr>com</a><br>
<a href="http://public.kitware.com/mailman/listinfo/kwiver-users" rel="noreferrer" target="_blank">http://public.kitware.com/<wbr>mailman/listinfo/kwiver-users</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div>Paul Tunison<br>Senior R&D Engineer</div><div>Kitware, Inc.</div><div>28 Corporate Drive</div><div>Clifton Park, NY 12065 USA</div><div><br></div><div>Phone: (518) 371-3971 Ext.164</div></div></div></div>
</div>