> This is a very interesting idea. Do you have plans to develop a set of > test and benchmark criteria to objectively assess the effectiveness of > different cost functions?
Yes! My cost function is purely theoretical, based on Mathis's formula. The purpose is to maximize TCP throughput. I fully agree that a set of test and benchmark is required. However I can only come up with a performance benchmark in a simulated packet-lossy environment, which was actually done by other people before. Do you have any other suggestions on the benchmark? > As Star said, he is using tunnels over home-broadband and datacenter > UDP. The high loss of packets could come from congestion or > ISP/state-level firewalls sabotaging connectivity. Yeah, thanks for your explanation. This is true, here in you know... China. _______________________________________________ Babel-users mailing list [email protected] https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/babel-users
