Thanks for the clarification Bogdan. Regards, Agalya
From: users-boun...@lists.opensips.org [mailto:users-boun...@lists.opensips.org] On Behalf Of Bogdan-Andrei Iancu Sent: Tuesday, September 20, 2016 7:06 AM To: Ramachandran, Agalya (Contractor) <agalya_ramachand...@comcast.com>; Liviu Chircu <li...@opensips.org>; OpenSIPS users mailling list <users@lists.opensips.org> Subject: Re: [OpenSIPS-Users] Load testing in opensips with async in 2.1 version Hi Agalya, With rest_client in async mode, you should see ~ cps rate as without having it (considering that the HTTP queries are finishing) - of course, this is valid for longer test intervals (much longer than the maximum time to reply for the HTTP queries). In async, the time to query will translate in 0 extra load for OpenSIPS. Of course, the processing of a call will be affected by the HTTP time to query, but on longer periods of time, the CPS will not be affected (as opensips will be able to handle the load). Regards, Bogdan-Andrei Iancu OpenSIPS Founder and Developer http://www.opensips-solutions.com On 19.09.2016 18:10, Ramachandran, Agalya (Contractor) wrote: Hi team, Have you performed load test with OpenSIPS 2.1 version by placing HTTP request as async ? If so can you share the throughput you have achieved? Because that would be bench mark for me as well to achieve the results. 1) Without HTTP being used at all, I see the below throughput . 150 calls per second 5000+ calls in full dialog 2) With HTPP as synchronous mode, it is been blocked till the response is received. As a workaround, I tried to increase the no of children parameter to handle many number of calls. Is this OK to do this way? 3) I just want to know the throughput data you guys have achieved for a load test with HTTP being async mode. This will be very useful to scale up and compare with my data. Regards, Agalya
_______________________________________________ Users mailing list Users@lists.opensips.org http://lists.opensips.org/cgi-bin/mailman/listinfo/users
