<div dir="ltr">We don't need to recall the Schrödinger cat for this kind of experiments !! Lincoln do you see that startup time is really higher when you start YourKit ??<br><div><br></div><div style>Which version are you using ?? In my experience the last one has a really good agent that impact not so much on performance in tracing or instrumentation.</div>
<div style><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">2013/1/30 Thomas Frühbeck <span dir="ltr"><<a href="mailto:fruehbeck@aon.at" target="_blank">fruehbeck@aon.at</a>></span><br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
IMHO both tools have their up/downsides.<br>
- instrumentation gives you a quite precise _relative_ performance<br>
indicator<br>
- sampling gives a _relatively precise_ performance indicator<br>
both have impact on the runtime, so you always have to account for the<br>
"uncertainty principle" :-)<br>
<br>
Am 29.01.2013 23:33, schrieb Stuart Douglas:<br>
<div class="im HOEnZb">> Tracing is pretty useless at identifying slow parts, as the overhead it<br>
> adds distorts the profile too much. The only thing it is really good for<br>
> is identifying methods that are called too often.<br>
><br>
> Sampling is generally much better.<br>
><br>
</div><div class="HOEnZb"><div class="h5">_______________________________________________<br>
forge-dev mailing list<br>
<a href="mailto:forge-dev@lists.jboss.org">forge-dev@lists.jboss.org</a><br>
<a href="https://lists.jboss.org/mailman/listinfo/forge-dev" target="_blank">https://lists.jboss.org/mailman/listinfo/forge-dev</a><br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br>****************************************<br><a href="http://www.lucamasini.net">http://www.lucamasini.net</a><br><a href="http://twitter.com/lmasini">http://twitter.com/lmasini</a><br>
<a href="http://www.linkedin.com/pub/luca-masini/7/10/2b9">http://www.linkedin.com/pub/luca-masini/7/10/2b9</a><br>****************************************
</div>