<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Could you load freeswitch with a couple hundred calls then run the test again.. and do the same to asterisk and see how the numbers stack up then? I'm just curious to see what happens at that point.<div><br></div><div>-Ray</div><div><div apple-content-edited="true"> </div><br><div><div>On Aug 21, 2009, at 3:15 PM, Rogelio Perez wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div>Hi Everyone,</div><div><br></div><div>I'm working on a PBX project for the <a href="http://www.google.com/url?sa=t&source=web&ct=res&cd=1&url=http%253A%252F%252Fwww.marvell.com%252Fproducts%252Fembedded_processors%252Fdeveloper%252Fkirkwood%252Fsheevaplug.jsp&ei=EOOOSo6ILcyOtgeNruTOBA&usg=AFQjCNFREhfy_erj5irBWk8XFUjkQOP-aw&sig2=pAHYlI15IbZ5Kcw0n-nIvA">Sheevaplug</a> ARM based computer, with the following specs: CPU 1.2 GHz, 512MB DDR2, no FPU.</div><div>So far I've found a big difference between Freeswitch and Asterisk performance times.</div><div>This is a comparison of the time it takes them to perform different actions:</div><div><br></div><blockquote class="webkit-indent-blockquote" style="margin: 0 0 0 40px; border: none; padding: 0px;">startup Freeswitch:<span class="Apple-tab-span" style="white-space:pre"> </span>3 min.<br>startup Asterisk:<span class="Apple-tab-span" style="white-space:pre"> </span>2 sec.<br><br>call extension Freeswitch:<span class="Apple-tab-span" style="white-space:pre"> </span>6 sec.<br>call extension Asterisk:<span class="Apple-tab-span" style="white-space:pre"> </span>0 sec.<br><br></blockquote><blockquote class="webkit-indent-blockquote" style="margin: 0 0 0 40px; border: none; padding: 0px;">shutdown Freeswitch:<span class="Apple-tab-span" style="white-space:pre"> </span>6.5 sec</blockquote><blockquote class="webkit-indent-blockquote" style="margin: 0 0 0 40px; border: none; padding: 0px;">shutdown Asterisk:<span class="Apple-tab-span" style="white-space:pre"> </span>0 sec.</blockquote><blockquote class="webkit-indent-blockquote" style="margin: 0 0 0 40px; border: none; padding: 0px;"><span class="Apple-tab-span" style="white-space:pre"></span><br>reload config Freeswitch: <span class="Apple-tab-span" style="white-space:pre"> </span>1 sec.<br>reload config Asterisk:<span class="Apple-tab-span" style="white-space:pre"> </span>1 sec.</blockquote><div><br></div><div>Both were built from sources natively (no cross-compiling), and they use the default startup configurations.</div><div>I have managed to lower the Freeswitch times by disabling most of the modules and recompiling, but it is still far away from Asterisk (i.e. FS startup time 2.5 min).</div><div><br></div><div>1. Is there any way to further improve Freeswitch performance for the ARM architecture?</div><div>2. Can this be related to the lack of a FPU (the Sheevalug emulates the floating point operations).</div><div>3. On the startup I see this error repeated many times: [ERR] switch_core_sqldb.c:95 SQL ERR [database is locked]. Can this be related?</div><div><br></div><div>Thanks,</div><div>Rogelio Perez</div><div><br></div></div>_______________________________________________<br>FreeSWITCH-users mailing list<br><a href="mailto:FreeSWITCH-users@lists.freeswitch.org">FreeSWITCH-users@lists.freeswitch.org</a><br>http://lists.freeswitch.org/mailman/listinfo/freeswitch-users<br>UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users<br>http://www.freeswitch.org<br></blockquote></div><br></div></body></html>