I should note that the [trr-multi](https://mozilla-necko.github.io/tests/dns/trr_multi_domain.html) test was designed as a distilled example of a scenario in which DoH showed higher latency than the os resolver. (Making a large number of dns requests while at the same time processing numerous other http requests). We don't necessarily need parity with the os resolver in this test to give us comparable overall performance since there are other scenarios in which DoH can outperform. (e.g. with DoH we see fewer of the sporadic multi-second requests, the DNS cache not cleared on change of network, etc).
Bug 2009372 Comment 22 Edit History
Note: The actual edited comment in the bug view page will always show the original commenter’s name and original timestamp.
I should note that the [trr-multi](https://mozilla-necko.github.io/tests/dns/trr_multi_domain.html) test was designed as a distilled example of a scenario in which DoH showed higher latency than the os resolver. (Making a large number of dns requests while at the same time processing numerous other http requests). We don't necessarily need parity with the os resolver in this test to give us comparable overall performance since there are other scenarios in which DoH can outperform. (e.g. with DoH we see fewer of the sporadic multi-second requests, the DNS cache is not cleared on change of network, etc).