<html><head><meta http-equiv="Content-Type" content="text/html charset=us-ascii"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;">Not a whole lot of cpu from any modern non embedded processor can probably handle this. Test to see exactly how it behaves with your exact requirements. Sound card would do nothing at all for transcoding, thats all cpu.<div><br></div><div>Mike</div><div><br><div><div>On Jun 4, 2014, at 6:20 PM, Oscar Franzen <<a href="mailto:oscar.franzen@advisa.se">oscar.franzen@advisa.se</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div dir="ltr">Hi!<div><br></div><div>I'm looking for some help with hardware for a production setup of freeswitch. I need at least 150 concurrent calls, but would prefer to buy a server which can handle more to facilitate growth.</div>
<div><br></div><div>I am looking at using the same codec (i think it is g711) for both legs of the call. I want to record all calls. Either to whatever codec and then transcode them separately or directly to mp3/ogg. I am going to use ESL to control FSW and also listen to most events over the socket.</div>
<div><br></div><div>What would you recommend for hardware? Is it recommended to buy a sound card for transcoding or just buy enough CPU? I will probably run from SSD, record to same SSD or other local disk and continuously move recordings to external (network) storage.</div>
<div><br></div><div>So:</div><div>150 concurrent calls</div><div>~10 initiations/second</div><div>mod_event_socket - for control and events</div><div>A limited amount of eavesdropping</div><div>recording (either to original codec or ogg/mp3)</div>
<div>Some "standard" modules.</div><div><div><br></div><div dir="ltr"><div><font face="arial, helvetica, sans-serif">Regards</font></div><div><font face="arial, helvetica, sans-serif">Oscar</font></div></div></div></div></blockquote></div></div></body></html>