Domain name system without root servers

0Citations
Citations of this article
4Readers
Mendeley users who have this article in their library.
Get full text

Abstract

We present a variation to the infrastructure of the Domain Name System (DNS) that works without DNS root servers. This allows to switch from a centralized trust model (root) to a decentralized trust model (top-level domains). By dropping DNS root in our approach, users have one entity less that they must trust. Besides trust issues, not relying on DNS root means that DNS root servers are no longer a central point of failure. Our approach is minimally invasive, builds on established DNS architecture and protocols and supports the DNS Security Extensions (DNSSEC). Furthermore, we designed our approach as an opt-in technology. Thus, each top-level domain operator can decide whether to support rootless DNS or not. The challenge of a rootless DNS is to keep track of changing IP addresses of top-level domain servers and to handle key rollovers, which are part of normal DNSSEC operation. Top-level domains opting in to rootless DNS must follow constraints regarding the frequency of changes of IP addresses and DNSSEC keys. We conducted a four-year measurement to show that 82% respectively 72% of top-level domains fulfill these constraints already.

Cite

CITATION STYLE

APA

Wander, M., Boelmann, C., & Weis, T. (2018). Domain name system without root servers. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10694 LNCS, pp. 203–216). Springer Verlag. https://doi.org/10.1007/978-3-319-76687-4_14

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free