Revisiting Comparative Performance of DNS Resolvers in the IPv6 and ECS Era

by   Rami Al-Dalky, et al.

This paper revisits the issue of the performance of DNS resolution services available to Internet users. While several prior studies addressed this important issue, significant developments, namely, the IPv6 finally getting traction and the adoption of the ECS extension to DNS by major DNS resolution services, warrant a reassessment under these new realities. We find that DNS resolution services differ drastically – by an order of magnitude in some locations – in their query response time. We also find established resolvers (Google DNS and OpenDNS) to lag far behind relative newcomers (Cloudflair and Quad9) in terms of DNS latency, and trace the cause to drastically lower cache hit rates, which we further trace to less cache sharing within the resolver platform. In addition, we find that public resolvers have largely closed the gap with ISP resolvers in the quality of CDNs' client-to-edge-server mappings as measured by latency. Finally, in most locations, we observe IPv6 penalty in the latency of client-to-CDN-edge-server mappings produced by the resolvers. Moreover, this penalty, while often significant, still does not rise above typical thresholds employed by the Happy Eyeballs algorithm for preferring IPv4 communication. resolvers. Thus, dual-stacked clients in these locations may experience suboptimal performance.


page 1

page 2

page 3

page 4


Accelerating QUIC's Connection Establishment on High-Latency Access Networks

A significant amount of connection establishments on the web require a p...

DarkHorse: A UDP-based Framework to Improve the Latency of Tor Onion Services

Tor is the most popular anonymous communication overlay network which hi...

A Theory of RPC Calculi for Client-Server Model

With multi-tier programming languages, programmers can specify the locat...

Gemini: Reducing DRAM Cache Hit Latency by Hybrid Mappings

Die-stacked DRAM caches are increasingly advocated to bridge the perform...

Client Selection Approach in Support of Clustered Federated Learning over Wireless Edge Networks

Clustered Federated Multitask Learning (CFL) was introduced as an effici...

Predictive Simulation: Using Regression and Artificial Neural Networks to Negate Latency in Networked Interactive Virtual Reality

Current virtual reality systems are typically limited by performance/cos...

Please sign up or login with your details

Forgot password? Click here to reset