Hi Bogdan, Our data for measuring Post Dial Delay (PDD) was limited to what we could collect from SIPp. To create a simple chart that would reflect the change in in PDD as a function of calls per second we focused on the number of calls for each test where the time between the second 100 messsage and the Ringing message was greater than 6 seconds. We then divided this number by the total number of calls for the test to get a percentage value that could be compared to other tests with a different number of calls per second.
For example, consider the first test for OpenSER V1.1 at 50 calls per second. There were 143 calls in the test with a measured time from the 2nd 100 to Ringing was greater than 6 second. We then divided 143 call by 30,000 calls (the total calls for this test) to calulate 0.48% of calls in this test had a PDD greater than 6 seconds. We recognize this analysis is not perfect, but we believe it provides some useful information. We appreciate the feedback and input. PDD analysis will be a topic for improvement in our next performance test effort. Thanks, Di-Shi Sun. _______________________________________________ Devel mailing list Devel@openser.org http://openser.org/cgi-bin/mailman/listinfo/devel
