Hi Rakesh, Authors et al, I much appreciate it if you can help me understand what is expected to be standardized in Section 4.2.3 <https://datatracker.ietf.org/doc/html/draft-gandhi-spring-stamp-srpm#section-4.2.3> of the draft. As I understand it, the transmitting test probe node constructs a packet that it receives back without any modification. It appears that there's no interaction with any other node in the network is defined. It is not clear how that is related to STAMP. As I can see, a test probe in the loopback mode can be anything since that is completely local to the sending system decision. As a result, I cannot find what needs standardization in that mechanism. I further have a question about the quality of the delay measurement that the loopback method can provide. It appears that you are concerned with how local processing at the STAMP Session-Reflector affects the accuracy of the measurement. But the STAMP enables a real one-way delay measurement that provides accurate statistical information about the condition of the particular path. The loopback method described in Section 4.2.3 measures the round-trip delay and, as a result, the accuracy of the downstream one-way delay is impacted by the choice of the upstream part of the engineered route that loops the test probe back to the sender. As a result, using the loopback mechanism may produce measurements that don't represent the network conditions in the downstream direction, on the path that is of interest and the real importance to the operator. The loopback mode may be producing results that are very far from the real conditions, lowering or inflating the measured delay.
I am looking forward to your feedback. Regards, Greg
_______________________________________________ spring mailing list spring@ietf.org https://www.ietf.org/mailman/listinfo/spring
