[EMAIL PROTECTED] wrote on 10/10/2007 10:55:35 AM: > >With the simple UAC/UAS scenario I can do 10,000CPS (SIP only, no media) > >when using a two processor 2.6Gz dual-core processor AMD Opteron. > > > Is this 10000 CPS for INVITE scenario? Actually, this would be a MESSAGE scenario not an INVITE scenario, which makes a big difference (roughly 40% the messages).
> What is your proxy implemention and running on which OS? This is a back-to-back number, but you can get similar numbers with a proxy in the middle. > Is it based on C implementation or other language? C. > > > This is > >with two SIPp instances (one acting as a UAC and another acting as a UAS), > >which is required because of the single-threaded event driven > >architecture. When I need to generate more load, I simply run multiple > >machines with multiple instances in parallel. In general, this scenario > >is CPU limited, other scenarios that I've run end up being memory limited. > > Depending on the scenario, I've run up to 12 machines with 36 SIPp > >instances in parallel. > > > > > If one wants to run on multiple machines, how would we sync the runs and > consolidate the results? Is there anyway to do this in SIPP? You have to create tooling outside of SIPp to consolidate the results. Charles ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Sipp-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/sipp-users
