A Comparative Study on Different DNS Transports for Different Resolvers

In this study, we compare two different transports that a DNS query may undergo when a message is passed from the user to the server. We compare two DNS transports, plain DNS over UDP and DNS over HTTPS (DoH), for three public resolvers (Google, Cloudflare and Quad9) from an Indian hostel network, using about 28,000 queries over roughly 30 hours. Over plain DNS, all three resolvers looked the same: median latency was about 6.8 ms, whereas over DoH, the resolvers separated: median latency was 14 ms for Cloudflare, 36 ms for Google and 65 ms for Quad9. We found that the plain DNS queries appear to be intercepted by the hostel network that we used and redirected to a single resolver, that is, the Google resolver, whereas DoH actually send the queries to the actual servers. Apart from this, we also compare the latency for three different resolvers(Google, Cloudflare, Quad9). Over plain DNS, all three resolvers returned almost identical latency, with medians of 6.83 ms (Cloudflare), 6.83 ms (Google) and 6.95 ms (Quad9), and no statistically significant difference between any pair (median differences of 0.03 to 0.15 ms). Even though there have been many papers about the performance metrics of different resolvers over different International websites, we also focus on some Indian sites and compare the latency of both these categories of websites. We found that Indian domains were slower than international ones over DoH for every resolver, by 2.4 ms for Cloudflare, 8.5 ms for Google and 36.4 ms for Quad9, and the ranking of the resolvers stayed the same

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24358432
Primary Topic
Network Traffic and Congestion Control
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

A Comparative Study on Different DNS Transports for Different Resolvers

Harshit Dugar
Zenodo (CERN European Organization for Nuclear Research)
Network Traffic and Congestion Control
preprint

A Comparative Study on Different DNS Transports for Different Resolvers

Harshit Dugar
preprint en

Abstract

In this study, we compare two different transports that a DNS query may undergo when a message is passed from the user to the server. We compare two DNS transports, plain DNS over UDP and DNS over HTTPS (DoH), for three public resolvers (Google, Cloudflare and Quad9) from an Indian hostel network, using about 28,000 queries over roughly 30 hours. Over plain DNS, all three resolvers looked the same: median latency was about 6.8 ms, whereas over DoH, the resolvers separated: median latency was 14 ms for Cloudflare, 36 ms for Google and 65 ms for Quad9. We found that the plain DNS queries appear to be intercepted by the hostel network that we used and redirected to a single resolver, that is, the Google resolver, whereas DoH actually send the queries to the actual servers. Apart from this, we also compare the latency for three different resolvers(Google, Cloudflare, Quad9). Over plain DNS, all three resolvers returned almost identical latency, with medians of 6.83 ms (Cloudflare), 6.83 ms (Google) and 6.95 ms (Quad9), and no statistically significant difference between any pair (median differences of 0.03 to 0.15 ms). Even though there have been many papers about the performance metrics of different resolvers over different International websites, we also focus on some Indian sites and compare the latency of both these categories of websites. We found that Indian domains were slower than international ones over DoH for every resolver, by 2.4 ms for Cloudflare, 8.5 ms for Google and 36.4 ms for Quad9, and the ranking of the resolvers stayed the same

Zenodo (CERN European Organization for Nuclear Research)
Vellore Institute of Technology University (IN)
Network Traffic and Congestion Control
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon