<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Fri, Jun 10, 2016 at 6:13 AM, Victor Stinner <span dir="ltr"><<a href="mailto:victor.stinner@gmail.com" target="_blank">victor.stinner@gmail.com</a>></span> wrote:</div><div class="gmail_quote"><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">The second result is a new perf module which includes all "tricks"<br>
discovered in my research: compute average and standard deviation,<br>
spawn multiple worker child processes, automatically calibrate the<br>
number of outter-loop iterations, automatically pin worker processes<br>
to isolated CPUs, and more.<br></blockquote><div><br></div><div>Apologies in advance if this is answered in one of the links you posted, but</div><div>out of curiosity was geometric mean considered?</div><div><br></div><div>In the compiler world this is a very common way of aggregating performance</div><div>results.</div><div><br></div><div>-- Meador</div></div>
</div></div>