The Domain Name System (DNS) has a direct impact on the performance and dependability of nearly all aspects of interactions on the Internet. DNS relies on a delegation-based architecture, where resolution of a name to its IP address requires resolving the names of the servers responsible for that name. The graphs of the inter-dependencies that exist between name servers associated with each zone are called Dependency Graphs. We constructed a DNS Dependency Model as a unified representation of these Dependency Graphs. We utilize a set of Structural Metrics defined over this model as indicators of external quality attributes of the domain name system. We explore the inter-metric and inter-quality relations further in order to quantify the indicative power of each metric. We apply some machine learning algorithms in order to construct Prediction Models of the perceived quality attributes of the operational system out of the structural metrics of the model. Assessing these quality attributes at an early stage of the design/deployment enables us to avoid the implications of defective and low-quality designs and deployment choices and identify configuration changes that might improve the availability, security, stability and resiliency postures of the DNS.

The Domain Name System (DNS) is one of the most important components of the Internet infrastructure. DNS relies on a delegation-based architecture, where resolution of names to their IP addresses requires resolving the names of the servers responsible for those names. The recursive structures of the inter-dependencies that exist between name servers associated with each zone are called dependency graphs. System administrators' operational decisions have far reaching effects on the DNSs qualities. They need to be soundly made to create a balance between the availability, security and resilience of the system. We utilize dependency graphs to identify, detect and catalogue operational bad smells. Our method deals with smells on a high-level of abstraction using a consistent taxonomy and reusable vocabulary, defined by a DNS Operational Model. The method will be used to build a diagnostic advisory tool that will detect configuration changes that might decrease the robustness or security posture of domain names before they become into production.Detecting and Refactoring Operational Smells within the Domain Name System secondary server hosting should take into consideration the impact of transitional trust and administrative complexity [28,15].While the original DNS design documents [18,22,23,5,29] call for diverse placement of authoritative name servers for a zone, bad configurations may lead to cyclic dependencies while bad deployment choices may lead to diminished and false server redundancy. It was also assumed that redundant DNS servers fail independently; previous measurements [28,11] showed that operational deployment choices made at individual zones can introduce excessive zone influence that severely affect the availability, security and resiliency of other zones. This research is motivated by the lack of formal analysis of the DNS interdependencies stemming from the delegation-based architecture as well as operational deployment choices made by system administrators. We approached the problem from a design point of view that takes into consideration the DNS zone configuration and server deployment choices rather than from the dynamic behavioural view [8] which includes statistical and post-deployment measurements. We propose a method to identify, specify and detect misconfigurations and bad deployment choices in the form of operational bad smells.The method utilizes a set of structural metrics defined over a DNS operational model to detect the smells in early stages of the DNS deployment. It also suggests graph-based refactoring rules as correction mechanisms for the bad smells. We apply and validate the method using several representative case studies. The method will be used to build a pre-emptive diagnostic advisory tool that will detect and flag configuration changes that might decrease the robustness or security posture of a domain name, before even the changes become into production. The contributions of this research are: