On 7/22/19, 5:42 PM, "DNSOP on behalf of Jason Livingood": > On 7/21/19, 4:27 PM, "DNSOP on behalf of Puneet Sood": >> Running the experiment from locations which are further away from cloud >> providers would provide another interesting set of data.
> [JL] +1. I understand the US FCC is considering running DoH tests from their > Measuring Broadband America end points around the US. Those devices are > managed by SamKnows, which just announced they added a DoH test to their > suite of measurements. >> Conclusion on benefit (or lack) of ECS. Did the page load measurements >> include content that would benefit from proximity to the end user, e.g. >> streaming videos or large downloads? This kind of content benefits from ECS >> when the resolver is further away from the client. > [JL] I wonder if it might be useful for CDNs to publish some measurement test > URLs that are highly localized (e.g. distributed widely & as close as > possible to users). That might enable better testing of performance for > accessing content that benefits from proximity to users. [AC] This would be helpful given it appears some (all?) DoH resolvers have indicated that they will not pass sufficient information to (rival) CDN vendors to allow geographic routing. Clearly if this is commonplace then performance gains from DoH, if any, could be lost from an end-user perspective due to inefficient routing of content. Andrew _______________________________________________ DNSOP mailing list DNSOP@ietf.org https://www.ietf.org/mailman/listinfo/dnsop
